WO2025255384A1

WO2025255384A1 - Dynamically managing treatment of parkinson's disease

Info

Publication number: WO2025255384A1
Application number: PCT/US2025/032513
Authority: WO
Inventors: Joshua Windmiller; Hifza NAJIB; Gordon Fletcher; Rush BARTLETT; Abdullah Abbasi; Paul Knutson
Original assignee: Dexcom Inc
Current assignee: Dexcom Inc
Priority date: 2024-06-06
Filing date: 2025-06-05
Publication date: 2025-12-11
Anticipated expiration: 2026-12-06

Abstract

In some embodiments, a system includes a processor in data communication with a memory having executable instructions. The processor is configured to execute the executable instructions to identify a current levodopa (L-DOPA) level in a patient's system and to determine an impact of the current L-DOPA level on the patient. The processor is also configured to execute the executable instructions to adjust a dosage of L-DOPA administered to the patient based on the current L-DOPA level and the impact.

Description

DYNAMICALLY MANAGING TREATMENT OF PARKINSON’S DISEASE

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to and benefit of U.S. Provisional Patent Application No. 63/656,893 filed June 6, 2024. This application also claims priority to and benefit of U.S. Provisional Patent Application No. 63/696,228 filed September 18, 2024. The aforementioned applications are hereby expressly incorporated by reference herein in their entirety as if fully set forth below and for all applicable purposes.

INTRODUCTION

[0002] Parkinson's disease (PD) is a chronic degenerative disorder of the central nervous system that mainly affects the motor system. The motor symptoms of PD result from the death of nerve cells in the substantia nigra, a region of the midbrain that supplies dopamine to the basal ganglia. Non-motor systems can include, for example, anxiety, cognitive problems, and orthostatic hypotension. It is estimated that, in the United States alone, approximately one million people may have PD. Since no cure for PD is currently known, treatment of PD generally focuses on reducing the effects of the symptoms.

[0003] Although PD can result in many debilitating conditions, few durable therapies exist. Treatment of PD typically involves administering levodopa (L-DOPA). L-DOPA is the precursor to the neurotransmitters dopamine, norepinephrine (noradrenaline), and epinephrine (adrenaline), collectively known as catecholamines. L-DOPA crosses the protective blood-brain barrier, whereas dopamine itself cannot. Thus, L-DOPA is used to increase dopamine concentrations in the treatment of PD.

SUMMARY

[0004] In some embodiments, one general aspect includes a method. The method includes identifying a current levodopa (L-DOPA) level in a patient’s system and determining an impact of the current L-DOPA level on the patient. The method also includes adjusting a dosage of L-DOPA administered to the patient based on the current L-DOPA level and the impact. [0005] In some embodiments, another general aspect includes a system. The system includes a memory having executable instructions. The system also includes a processor in data communication with the memory. The processor is configured to execute the executable instructions to identify a current levodopa (L-DOPA) level in a patient’s system and to determine an impact of the current L-DOPA level on the patient. The processor is also configured to execute the executable instructions to adjust a dosage of L-DOPA administered to the patient based on the current L-DOPA level and the impact.

[0006] In some embodiments, another general aspect includes a computer-program product. The computer-program product includes a non-transitory computer-usable medium having computer-readable program code embodied therein. The computer-readable program code is adapted to be executed to implement a method. The method includes identifying a current levodopa (L-DOPA) level in a patient’s system and determining an impact of the current L-DOPA level on the patient. The method also includes adjusting a dosage of L-DOPA administered to the patient based on the current L-DOPA level and the impact.

[0007] In some embodiments, another general aspect includes a monitoring system. The monitoring system includes a continuous levodopa (L-DOPA) sensor configured to generate L- DOPA measurements associated with L-DOPA levels of a patient, a symptom feedback sensor configured to generate symptom information measuring a physical response of the patient to the L-DOPA levels, a memory having executable instructions, and a processor in data communication with the memory, the continuous L-DOPA sensor, and the symptom feedback sensor. The processor is configured to execute the executable instructions to receive, from the continuous L- DOPA sensor, the L-DOPA measurements, and to receive, from the symptom feedback sensor, the symptom information. The processor is configured to execute the executable instructions to characterize the L-DOPA measurements in achieving symptom control for the patient based on the symptom information, and to define a personalized therapeutic window for the patient based on the characterization of the L-DOPA measurements. The personalized therapeutic window includes a target range of L-DOPA levels to minimize symptoms.

[0008] In some embodiments, another general aspect includes a method. The method includes identifying a current levodopa (L-DOPA) level in a patient’s system and receiving an input from a surface electromyography (sEMG) sensor positioned in relation to a muscle of the patient. The method also includes determining an impact of the current L-DOPA level on the patient based on the input from the sEMG sensor. The method also includes adjusting a dosage of L-DOPA administered to the patient based on the current L-DOPA level and the impact.

[0009] In some embodiments, another general aspect includes a system. The system includes a memory having executable instructions and a processor in data communication with the memory. The processor is configured to execute the executable instructions to identify a current levodopa (L-DOPA) level in a patient’s system and receive an input from a surface electromyography (sEMG) sensor positioned in relation to a muscle of the patient. The processor is also configured to execute the executable instructions to determine an impact of the current L-DOPA level on the patient based on the input from the sEMG sensor. The processor is also configured to execute the executable instructions to adjust a dosage of L-DOPA administered to the patient based on the current L-DOPA level and the impact.

[0010] In some embodiments, another general aspect includes a computer-program product. The computer-program product includes a non-transitory computer-usable medium having computer-readable program code embodied therein. The computer-readable program code is adapted to be executed to implement a method. The method includes identifying a current levodopa (L-DOPA) level in a patient’s system and receiving an input from a surface electromyography (sEMG) sensor positioned in relation to a muscle of the patient. The method also includes determining an impact of the current L-DOPA level on the patient based on the input from the sEMG sensor. The method also includes adjusting a dosage of L-DOPA administered to the patient based on the current L-DOPA level and the impact.

[0011] In some embodiments, another general aspect includes a monitoring system. The monitoring system includes a continuous levodopa (L-DOPA) sensor configured to generate L-DOPA measurements associated with L-DOPA levels of a patient, a surface electromyography (sEMG) sensor positioned in relation to a muscle of the patient and configured to generate symptom information measuring activity of the muscle, a memory having executable instructions, and a processor in data communication with the memory, the continuous L-DOPA sensor, and the sEMG sensor. The processor is configured to execute the executable instructions to receive, from the continuous L-DOPA sensor, the L-DOPA measurements, and to receive, from the sEMG sensor, the symptom information measuring the activity of the muscle. The processor is also configured to execute the executable instructions to characterize the L-DOPA measurements in achieving symptom control for the patient based on the symptom information, and to define a personalized therapeutic window for the patient based on the characterization of the L-DOPA measurements. The personalized therapeutic window includes a target range of L-DOPA levels to minimize symptoms.

[0012] In some embodiments, another general aspect includes a method. The method includes identifying a future time interval for Parkinson’s disease (PD) symptom control for a patient. The method also includes automatically determining levodopa (L-DOPA) dosing information that prioritizes PD symptom control during the future time interval over PD symptom control in at least one other future time interval. The method also includes facilitating treatment of the patient based on the automatically determined L-DOPA dosing information.

[0013] In some embodiments, another general aspect includes a method. The method includes receiving a levodopa (L-DOPA) dosage for a patient. The method also includes automatically determining recommended timing for the L-DOPA dosage based on a therapeutic window for the patient. The method also includes facilitating treatment of the patient based on the recommended timing for the L-DOPA dosage.

[0014] In some embodiments, another general aspect includes a method. The method includes identifying a current levodopa (L-DOPA) level in a patient’s system. The method also includes receiving a real-time recording of the patient’s voice. The method also includes determining an impact of the current L-DOPA level on the patient based on an analysis of the patient’s voice in the recording. The method also includes adjusting a dosage of L-DOPA administered to the patient based on the current L-DOPA level and the impact.

[0015] In some embodiments, another general aspect includes a method. The method includes identifying a current levodopa (L-DOPA) level in a patient’s system. The method also includes identifying one or more current physical characteristics of the patient based on information received from an inertial measurement unit associated with the patient. The method also includes determining an impact of the current L-DOPA level on the patient based on the one or more current physical characteristics. The method also includes adjusting a dosage of L-DOPA administered to the patient based on the current L-DOPA level and the impact. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] So that the manner in which the above-recited features of the present disclosure can be understood in detail, a more particular description, briefly summarized above, may be had by reference to aspects, some of which are illustrated in the drawings. It is to be noted, however, that the appended drawings illustrate only certain typical aspects of this disclosure and are therefore not to be considered limiting of its scope, for the description may admit to other equally effective aspects.

[0017] FIG. 1 illustrates an example therapy management system, in accordance with certain aspects of the present disclosure.

[0018] FIG. 2 illustrates an example of a continuous L-DOPA monitor (CLM) system, in accordance with certain aspects of the present disclosure.

[0019] FIG. 3 illustrates example inputs and example metrics that are calculated based on the inputs for use by the therapy management system of FIG. 1, in accordance with certain aspects of the present disclosure.

[0020] FIG. 4 illustrates example operation of a continuous L-DOPA sensor, in accordance with certain aspects of the present disclosure.

[0021] FIG. 5 illustrates examples of data sources that can provide, for example, inputs as described relative to FIG. 3, in accordance with certain aspects of the present disclosure.

[0022] FIG. 6A illustrates an example of a sensor configuration relative to a patient, in accordance with certain aspects of the present disclosure.

[0023] FIG. 6B illustrates an example architecture for a surface electromyography (sEMG) sensor having inertial measurement unit (IMU) features, in accordance with certain aspects of the present disclosure.

[0024] FIG. 6C illustrates an example communications framework based on the sensor configuration of FIG. 6A, in accordance with certain aspects of the present disclosure.

[0025] FIG. 7 illustrates an example of a process for establishing a sensor configuration, in accordance with certain aspects of the present disclosure. [0026] FIG. 8 illustrates an example of a process for monitoring and tracking Parkinson’s disease symptoms, in accordance with certain aspects of the present disclosure

[0027] FIG. 9 illustrates an example of L-DOPA measurement information that can be generated, for example, by the CLM sensor system of FIG. 2, in accordance with certain aspects of the present disclosure.

[0028] FIG. 10 illustrates an example of user feedback that can be provided via a user interface, in accordance with certain aspects of the present disclosure.

[0029] FIG. 11 illustrates examples of accelerometer feedback, in accordance with certain aspects of the present disclosure.

[0030] FIG. 12 illustrates an example of motor function feedback that can be provided by a surface electromyography (sEMG) sensor, in accordance with certain aspects of the present disclosure.

[0031] FIG. 13 A illustrates an example of time-based accelerometer magnitude data indicating a sleeping patient, in accordance with certain aspects of the present disclosure.

[0032] FIG. 13B illustrates a time-based spectrogram showing which frequencies have energy for the sleeping patient of FIG. 13A, in accordance with certain aspects of the present disclosure.

[0033] FIG. 14A illustrates an example of time-based accelerometer magnitude data indicating various activities of a patient, in accordance with certain aspects of the present disclosure.

[0034] FIG. 14B illustrates a time-based spectrogram showing which frequencies have energy for the patient of FIG. 14A, in accordance with certain aspects of the present disclosure.

[0035] FIG. 15 A illustrates an example of motor function feedback, in accordance with certain aspects of the present disclosure.

[0036] FIG. 15B another example of motor function feedback, in accordance with certain aspects of the present disclosure.

[0037] FIG. 16 illustrates an example of correlating user feedback, accelerometer feedback, and L-DOPA sensor feedback, in accordance with certain aspects of the present disclosure. [0038] FIG. 17 illustrates an example of correlating motor function feedback from an sEMG sensor and/or an accelerometer with L-DOPA sensor feedback, in accordance with certain aspects of the present disclosure.

[0039] FIG. 18 illustrates examples of tracking an effectiveness of L-DOPA treatment, in accordance with certain aspects of the present disclosure.

[0040] FIG. 19 illustrates examples of tracking L-DOPA treatment in relation to tremor values, in accordance with certain aspects of the present disclosure.

[0041] FIG. 20 illustrates examples of tracking an effectiveness of L-DOPA treatment in relation to meals, meal timing, and activities, in accordance with certain aspects of the present disclosure.

[0042] FIG. 21 illustrates an example process for determining a personalized L-DOPA range and administering an L-DOPA dose, in accordance with certain aspects of the present disclosure.

[0043] FIG. 22 illustrates an example of a scenario 2200 for a patient experiencing a fall, in accordance with certain aspects of the present disclosure.

[0044] FIG. 23 illustrates an example process for continuous L-DOPA monitoring and therapy management, in accordance with certain aspects of the present disclosure.

[0045] FIG. 24A is a graph illustrating an example time-based correlation of L-DOPA measurement information, symptom information, and L-DOPA dosing information, in accordance with certain aspects of the present disclosure.

[0046] FIG. 24B is a graph plotting symptoms versus L-DOPA concentrations over time, in accordance with certain aspects of the present disclosure.

[0047] FIG. 25A illustrates an example user interface output indicating a dangerous zone for L-DOPA levels, in accordance with certain aspects of the present disclosure.

[0048] FIG. 25B illustrates another example user interface output indicating a dangerous zone for L-DOPA levels, in accordance with certain aspects of the present disclosure.

[0049] FIG. 26 illustrates an example of a process for automatic treatment determination, in accordance with certain aspects of the present disclosure. [0050] FIG. 27 is a block diagram depicting a computing device configured for continuous L-DOPA monitoring and therapy management, in accordance with certain aspects of the present disclosure.

[0051] To facilitate understanding, identical reference numerals have been used, where possible, to designate identical elements that are common to the figures. It is contemplated that elements disclosed in one aspect may be beneficially utilized on other aspects without specific recitation.

DETAILED DESCRIPTION

[0052] Management of PD can present complex challenges for patients, clinicians, and caregivers, particularly regarding treatment using L-DOPA. A confluence of numerous factors, such as dosage, time of administration and patient-specific factors (e.g., age, diet and activity level), may impact the effectiveness of L-DOPA. L-DOPA may also become less effective over time. Furthermore, prolonged use and/or high doses of L-DOPA can cause negative side effects, such as involuntary muscle movements and orthostatic hypotension. Further, although lower doses of L-DOPA may minimize side effects, such lower doses may also be ineffective in mitigating symptoms of PD. These challenges are exacerbated by the fact that L-DOPA generally has a narrow therapeutic profile and is characterized by relatively fast clearance from the body with around a 90-minute half-life.

[0053] Due to the aforementioned challenges, L-DOPA dosing is typically a trial-and-error process. There is currently no effective way for patients, providers, and caregivers to understand the bioavailability and efficacy of L-DOPA in tandem and in real time for patients initiating or modifying L-DOPA treatment. Likewise, there is currently no effective way for patients, providers, and caregivers to understand the impact of disease progression or lifestyle changes on PD symptoms. Thus, when it comes to treating PD with L-DOPA, it can be difficult to determine the timing and duration of symptom improvement, if any. As a general matter, it is desirable to maximize the benefits (and minimize the detriments) of the administration of L-DOPA, and it is also desirable to be able to predict a user’s physical response to an L-DOPA dosage at various times. However, current L-DOPA dosage management consists of manual symptom entry and manual trial-and-error dosage adjustments. This may negatively affect a PD patient who is taking L-DOPA. For example, a patient receiving a dose of L-DOPA may experience PD symptoms that are avoidable by administering a more accurate L-DOPA dose. In another example the patient may experience negative effects of an unnecessarily excessive administered dose of L-DOPA that could also be avoided by administering a more accurate L-DOPA dose.

[0054] Accordingly, certain aspects described herein provide a continuous L-DOPA monitoring (CLM) sensor system that can generate L-DOPA measurement information (e.g., a time series of L-DOPA levels) on a continuous basis (e.g., every 10 seconds, every 30 seconds, every 5 minutes, every 10 minutes, etc.). In various embodiments, the L-DOPA measurement information can be provided to a system, such as a user display device (e.g., a smartphone or smartwatch), on-demand, as the information is generated, according to a predetermined schedule (e.g., every 5 minutes, every 10 minutes, etc.), in response to a periodic request, for example, from the user display device that is sent according to such a predetermined schedule, and/or in other suitable ways. In some embodiments, the predetermined schedule can be driven by events, such that if certain types of events are detected (e.g., a fall or a certain level of muscle tremors), the L- DOPA measurement information is provided more frequently on at least a temporary basis (e.g., providing the L-DOPA measurement information every 5 minutes, instead of every 10 minutes, for the next hour). The CLM can include, for example, a wearable L-DOPA sensor that performs continuous monitoring of L-DOPA levels in interstitial fluid of a patient. In this way, the CLM sensor system can provide useful information as to the bioavailability of L-DOPA in the patient’s body. The CLM sensor system, however, introduces further technical challenges.

[0055] For example, it is technically difficult to continuously manually monitor L-DOPA measurement information in a manner that improves PD management. Manual monitoring would be impractical, for example, due to the high sampling rate (e.g., every minute, every 5 minutes, every 30 minutes, etc.), the resulting large quantities of data, and the complexity of identifying an appropriate response or change in treatment (e.g., a dosage change). Automatic monitoring, for example, of L-DOPA levels, would be of limited usefulness due to technical limitations in the field of PD management. For example, L-DOPA levels, by themselves, do not reliably indicate effectiveness in mitigating PD symptoms, the existence or degree of negative side effects, disease progression, or lifestyle factors.

[0056] The present disclosure describes examples of a therapy management system configured to provide therapy management as well as enable controlled administration of L-DOPA for treatment, for example, of PD. In various aspects, the therapy management system is configured to receive and analyze a set of patient-related information to infer the patient’s degree of symptom control (e.g., a degree of measurable physical symptoms such as muscle tremors or muscle rigidity), identify the timing and duration of symptom improvement in the patient, monitor the effects of PD on the patient, track the patient’s compliance with medication such as L-DOPA, recommend and/or command L-DOPA dosing (e.g., amounts and times of administration), and/or recommend other changes relative to the patient (e.g., a diet change). The set of patient-related information can include L-DOPA dosing information (e.g., dose and time of administration) from a continuous a L-DOPA delivery device (e.g., an L-DOPA pump) or user device (e.g., a smartphone), L-DOPA measurement information (e.g., a time series of L-DOPA levels) from a CLM sensor system, food consumption information from a user and/or another source, and symptom information from various feedback sources (e g., inertial measurement units such as accelerometers, vocal biomarker sensors such as microphones, surface electromyography (sEMG) sensors, galvanic skin response, user feedback, etc.).

[0057] In various aspects, the symptom information can be indicative of symptoms of PD and/or symptoms associated with side effects of medication such as L-DOPA. For example, the symptom information can include measurements, provided by one or more hardware or software sensors, of the patient’s physical response to current L-DOPA levels, the physical impact of L- DOPA and/or PD on the patient. In some aspects, these measurements can be used to infer a patient’s degree of motor symptom control. The sensors that supply measurements can include, for example, a vocal biomarker sensor, a mobile device having an inertial measurement unit (e.g., an accelerometer, gyroscope, magnetometer, or inclinometer), an sEMG sensor, or another suitable device or component. In other examples, the symptom information can include behavioral information (e.g., an amount of device usage), indicators of cognitive function (e.g., results of puzzles or tests), and/or outputs of sensors for one or more of electrocardiogram (ECG), electroencephalography (EEG), heart rate variability (HRV), heart rate, blood pressure, body impedance, skin conductance, sleep monitoring, body sounds (e.g., gastrointestinal acoustic monitoring), voice recognition, electrogastrography (e.g., for gastroparesis), and/or the like. In some aspects, the sensors that supply measurements can include, for example, L-DOPA sensors as discussed in U.S. Provisional Application No. 63/605,096 filed December 1, 2023, U.S. Patent Application No. 18/952,481 filed November 19, 2024, and/or International Application No.

PCT/US2024/056525 filed November 19, 2024, all of which are hereby incorporated by reference.

[0058] By way of more particular example, the therapy management system can monitor a patient’s muscle activity based on input from one or more sEMG sensors. sEMG sensors can provide data useful for understanding the intricacies of the movement disorder at a neuromuscular level including, for example, which muscle groups are most impacted at during different activities (e.g., rest, walking, running, etc.). In various embodiments, such data further aids in determining a solution plan to preserve muscle activity, pinpointing high risk activities, and identifying recommended mitigations.

[0059] Continuing the foregoing example, the sEMG sensors can be worn in relation to one or more selected muscles of the patient (e g., on or over the patient’s biceps, triceps, and/or other externally facing skeletal muscles). In various aspects, the EMG sensor(s) measure electrical activity produced by the selected muscle(s). The therapy management system can analyze the muscle activity in conjunction with other patient-related information. The other patient-related information can include, for example, L-DOPA dosing information (e.g., dose and time of administration) from a continuous L-DOPA pump or user device (e.g., a smartphone), L-DOPA measurement information (e.g., a time series of L-DOPA levels) from a continuous L-DOPA monitor (CLM) sensor system, food consumption information from a user and/or another source, and symptom information from other feedback sources such as inertial measurement units (IMUs) (e.g., accelerometers, magnetometers, and/or gyroscopes), vocal biomarker sensors such as microphones, galvanic skin response sensors, and/or user feedback (e.g., voice, textual, indications via a user interface, etc.).

[0060] In various aspects, the therapy management system can identify the timing and duration of symptom improvement relative to L-DOPA administration, for example, by correlating the L- DOPA measurement information, the symptom information, and/or the L-DOPA dosing information based on time. For example, following the time-based correlation, the therapy management system can characterize the L-DOPA measurement information for a given period by a corresponding degree of symptom control evidenced by the symptom information for the same period. In various aspects, the therapy management system can define a personalized therapeutic window for the patient (e.g., a target range of L-DOPA levels deemed effective for the patient) and determine, based on the time-based correlation, personalized (e.g., patient-specific) L-DOPA dosing (e.g., amounts and/or times) that produces the L-DOPA levels within the therapeutic window. Such determinations can include, for example, characterizing delay between dose administration and symptom relief as well as other factors that may affect symptom relief. In various aspects, the therapy management system can continuously adapt L-DOPA dosing, or recommend such adaptations, based on a continuous observation of the patient’s symptom control (e.g., motor symptom control) as evidenced by the symptom information (e.g., motor function information such as presence of muscle tremors or muscle rigidity).

[0061] In various aspects, the CLM sensor system and/or the therapy management system can facilitate personalized L-DOPA dosing. Typically, each patient has an individualized reaction to L-DOPA based on the progression of their PD and many other factors including but not limited to physiological characteristics like height, weight, genes, and lifestyle characteristics like diet, sleep patterns, exercise, gastric motility or gastrointestinal disease, etc. Likewise, the time of day a patient takes L-DOPA and the route of administration, including oral, injected, continuous pump delivery, etc. can greatly affect the L-DOPA levels and the body’s physiologic response to those levels. In various aspects, the CLM sensor system and/or the therapy management system can identify and track individualized reactions to L-DOPA based on factors and variables such as the foregoing.

[0062] In various aspects, the therapy management system can monitor the effects of PD on a patient, for example, to determine the progression of disease and how well controlled a patient’s symptoms are with a current dosage regime. The therapy management system can identify specific ways a patient’s symptom control is affected by activities or daily acts such as time of L-DOPA administration and/or lifestyle or other treatments the patient may be receiving for PD and/or a comorbidity. In addition, or alternatively, the therapy management system can analyze factors like sleep patterns, genetics, and aging among many other circumstances and situations for measurable impact on PD symptoms and/or treatment.

[0063] In various aspects, the therapy management system can measure effects of disease and/or treatment on a patient’s cognitive capacity for answering questions and/or puzzle challenges, muscle and motor control, speech control, bowel and bladder control, and/or cardiac activity. Many other neuro and muscular effects can be measured by various feedback sources (e.g., analyte and non-analyte sensors) and other measurement techniques to determine the level of symptoms and degree of PD severity and control under the patient’s current treatment regimen and situation. A measurement of L-DOPA concentrations overtime in combination with symptom information from the feedback sources, with or without other external data, can help healthcare providers, caregivers, and/or the patient understand if their dosage and/or frequency of L-DOPA should be changed.

[0064] In various aspects, for patients on an oral or periodic injection or other periodic route of administration of L-DOPA, the therapy management system can track compliance with L- DOPA medication. Compliance tracking can be helpful for patients, healthcare providers, caregivers, and/or insurance companies. For patients and caregivers, an increase in L-DOPA levels corresponding to an expected consumption of L-DOPA medication could indicate that the L- DOPA medication was taken. A lower or higher than expected rise could correspond to a patient or provider administering an incorrect L-DOPA dosage. If no L-DOPA change is detected when expected, such non-detection of change could indicate a missed dose. Additionally, if L-DOPA levels do not rise within an expected window, the therapy management system can cause the patient or a caregiver, for example, to be prompted with one or more reminders that the patient needs to take their L-DOPA medication.

[0065] In various cases, if a higher than expected level of L-DOPA is repeatedly detected in conjunction with a lower than expected level of motor symptom relief, such repeated detection may indicate that the patient has progressed in their disease and/or that some other factor has resulted in a more or less significant than expected bioavailability of L-DOPA (e.g., a diet or lifestyle change or a change in medication or dietary supplement used by the patient to treat another condition which may interfere with L-DOPA or another disease or condition impacting the therapy such as but not limited to diabetes, kidney disease, liver disease, or gastric dysfunction). If the patient has made a positive lifestyle change, they and/or their healthcare provider team could be given affirmational support and/or recommendations to reduce L-DOPA dosage. Similarly, if no such positive effect has been noted and the patient’s symptoms have worsened or remained the same, the therapy management system can recommend that the care team do additional tests and/or increase the dosage (e.g., via prompts from the therapy management system). [0066] In various aspects, in the case of a closed loop system including an L-DOPA pump and a CLM sensor system, the patient can be tracked via the CLM sensor system. Information from the CLM sensor system can be provided to an algorithm in the pump, CLM and/or therapy management system that would command the L-DOPA pump to modify its delivery. In various aspects, because proteins and other ingested materials can affect the L-DOPA levels within the body, other feedback (e.g., user feedback identifying meals and/or meal timing) can be incorporated into the algorithm to help determine the optimal modifications for L-DOPA delivery throughout the day. In some aspects, an artificial intelligence module can receive images of meals from users (e.g., images captured, provided, or indicated by the users) and automatically estimate nutritional content such as protein. Examples of the artificial intelligence module are described in U.S. Provisional Application No. 63/683,656 filed August 15, 2024, which application is hereby incorporated by reference. In these aspects, the automatically estimated nutritional content can be provided into the algorithm to help determine the optimal modifications for L-DOPA delivery, as discussed above.

[0067] In various aspects, in the case of L-DOPA therapy used to treat symptoms of PD or other motor neuron disease such as monoamine oxidase-B (MAO-B) inhibitors (e.g., Rasagiline (Azilect) and selegiline (Eldepryl) and/or dopamine agonists (e.g., Ropinirole (Requip), pramipexole (Mirapex), rotigotine (Neupro), and apomorphine (Apokyn)), tracking L-DOPA level interstitially and/or the analyte concentration of the additional therapy can be used to understand compliance and effectiveness of therapy. Monitoring one or more of these analytes in combination with symptom tracking (e.g., with patient self-reported outcomes and/or other sensors such as blood pressure, ECG, EEG, accelerometer, and/or other data) can help determine progression and severity of PD relative to the dosage and/or concentrations of these medications available to the patient’s body.

[0068] For example, for a patient on L-DOPA therapy, the symptoms of PD may be most attenuated when the concentration of L-DOPA is within a therapeutic window (e.g., a target range of L-DOPA levels deemed effective for the patient). If the concentration of L-DOPA is too high, that may lead to dyskinesia and/or muscle rigidity, and if the L-DOPA concentration is too low, that may lead to akinesias and parkinsonian symptoms. The therapeutic window may narrow with the progression of the disease. In certain aspects, the therapy management system can stage the disease by determining when the symptoms appear and what types of symptoms they are relative to the concentration of L-DOPA interstitially measured, for example, by the CLM sensor system.

[0069] For example, the symptoms may be measured by one or more symptom feedback sensors, such as but not limited to inertial measurement units (e.g., accelerometers, gyroscopes, or inclinometers), sensors that provide electrophysiological readings (e.g., EEG, ECG, EMG), sensors that provide skin measurements (e.g., galvanic skin response, bioelectrical impedance), voice monitors, etc. In addition, the therapy management system can recommend a disease staging measurement protocol on a temporary basis, one or more times throughout a period of time, for example once or twice per year, to gain a quantitative assessment of the stage and progression of PD. The patient may begin this disease staging measurement protocol at a well-controlled symptom level relative to a specific L-DOPA concentration that is within a personalized therapeutic window (e.g., a target range of L-DOPA levels deemed effective for that patient), as measured by sensors or tools.

[0070] In certain aspects, the therapy management system can thereafter recommend a gradual lessening of the therapeutic concentration until significant or noted symptoms appear (e.g., akinesias), establishing the lower concentration threshold of effect. Thereafter, the therapy management system may recommend a gradual increasing of therapeutic concentration until significant or noted symptoms appear (for example dyskinesia), establishing the higher concentration threshold effect. These two thresholds may then be compared to the last protocol for testing the individual patient’s therapeutic range and or against a standard population level therapeutic range vs disease progression metric. This comparison overtime will provide the patient and their healthcare providers with more information about how the therapeutic range of effectiveness of L-DOPA is changing over time to give an indication of disease progression, stabilization, or even improvement. Additionally, or alternatively, averages or absolute trough and peak measurement of concentrations at the time when symptoms begin, as measured throughout the course of multiple successive days and continuous sensor wear, could also be used to provide a guided real time measurement of progression over normal sensor wear and daily living instead or, in addition to, a specifically provided excursion trial.

[0071] Akinesias and other parkinsonian symptoms may become more pronounced when the concentration is cleared, such that the concentration of L-DOPA starts to drop to the lower end of the therapeutic window prior to the next dose of therapy. Dyskinesia symptoms may become more pronounced when the concentration is too high or rapidly changing in some patients. Measuring symptoms relative to the concentration of symptom attenuating therapies can help determine the progression or regression of disease for patients on symptom-mitigating therapy such as L-DOPA and a combination therapy that is meant to slow or reverse the disease. In this manner, disease progression can be monitored by the therapy management system even when disease modifying therapy is being used, and the concentration of the disease-modifying therapy can become an additional variable to inform symptom control.

[0072] In various aspects, the aforementioned features can result in a useful correlation of patient symptoms and/or other physical characteristics with L-DOPA levels of the patient. In certain aspects, this correlation, due to its effectiveness in treatment applications, can minimize future user adjustments of an L-DOPA implementation system, and associated network, storage, and compute requirements for such a system. In view of a large corpus of patients, the minimized adjustments can significantly reduce back-end network and computation requirements, thereby improving performance of such systems. In addition, or alternatively, the correlation can be leveraged to adjust one or more settings or functions of monitoring or dosing hardware or software. In certain cases, such adjustments may disable unnecessary features for certain conditions, which can improve performance and battery life of associated hardware.

[0073] In this way, by monitoring a current amount of L-DOPA in a patient's system in realtime, and also monitoring a current physical effect of this measured L-DOPA amount on the patient, this information may be analyzed to improve an accuracy of future L-DOPA dosages to be administered to the patient. This improved accuracy may in turn improve medicament dosing instructions (e.g., dosing instructions sent to a hardware L-DOPA pump), which may minimize the negative effects of L-DOPA and maximize the positive effects of L-DOPA on the patient, thereby improving a physical condition of the patient. This improved accuracy may also improve recommendations sent to the patient by the system. These improved recommendations (such as diet, exercise, and medication recommendations) may be followed by the patient, resulting in a favorable improvement of the patient’s PD symptoms. Improved medicament dosing and patient recommendations may also improve a patient’s motor (and potentially non-motor) functions. [0074] Also, as increased amounts of L-DOPA dosing data, in-patient L-DOPA levels, and associated L-DOPA physical effect data are received from a patient over time (as a result of the above improved procedures), a system may identify the results of earlier dosages/recommendations and may continually refine future dosages/recommendations based at least in part on this monitored data. This may in turn improve the accuracy of an L-DOPA implementation system.

[0075] In various aspects, the therapy management system can track historical user characteristics, symptoms, dosages, etc. This data can be used to improve treatment for other users who share one or more of the characteristics.

[0076] Although certain examples are described herein in relation to a patient with PD that takes L-DOPA, the aspects herein are likewise applicable and useful in connection with any disease or condition involving other treatments. For example, other catecholamine neurotransmitters (e.g., tyrosine, dopamine, epinephrine, and norepinephrine) may be measured and used in a similar way to the methods described here for other neurology applications. In particular, as shown by way of example relative to Table 1 below, sensors described herein may measure other elements in the same chemical pathway (e.g., other elements of L-DOPA ingestion).

Table 1

[0077] FIG. 1 illustrates an example of a therapy management system 100 for providing treatment recommendations, in relation to users 102 (individually referred to herein as a user and collectively referred to herein as users), using the CLM sensor system 104, including one or more L-DOPA sensors. A user 102, in certain aspects, may be the patient or, in some cases, the patient’s caregiver. In certain aspects, therapy management system 100 includes a CLM sensor system 104, a display device 107 that executes application 106, a therapy management engine 114, a patient database 110, a historical records database 112, a training server system 140, and a therapy management engine 114, each of which is described in more detail below.

[0078] In certain aspects, the CLM sensor system 104 is configured to continuously measure L-DOPA and transmit the L-DOPA measurements to display device 107 for use by application

106. In some aspects, the CLM sensor system 104 may primarily function as a monitoring device by pairing with the display device 107 and transmitting L-DOPA measurements to the display device 107 in a continuous or semi-continuous manner. In other aspects, the CLM sensor system 104 may primarily function as a diagnostic device that is configured to store and log the L-DOPA measurements. In such aspects, the data log stored by the CLM sensor system 104 may be transmitted to a remote service (e.g., a cloud server) without the involvement of the display device

107. In such aspects, the CLM sensor system 104 may be equipped with a mobile internet of things (loT) interface (e.g., LTE, Cat-Mi, NB-IoT, etc.), a cellular radio (e.g., 3G, 4G, LTE, 5G, 6G, etc.), or other means to directly communicate the L-DOPA measurements in the data log to the remote server.

[0079] In some aspects, CLM sensor system 104 transmits the L-DOPA measurements to display device 107 through a wireless connection (e.g., Bluetooth connection). In certain aspects, display device 107 is a smartphone. However, in certain other aspects, display device 107 may instead be any other type of computing device, such as a laptop computer, a smartwatch, a tablet, or any other computing device capable of executing application 106. In some aspects, CLM sensor system 104 and/or L-DOPA sensor application 106 transmit the L-DOPA measurements to one or more other individuals having an interest in the health of the patient (e.g., a family member or physician for real-time treatment and care of the patient). CLM sensor system 104 may be described in more detail with respect to FIG. 2.

[0080] Application 106 is a mobile health application that is configured to receive and analyze L-DOPA measurements from CLM sensor system 104. In particular, application 106 stores information about a patient, including the patient’s L-DOPA measurements, in a patient profile 118 associated with the patient for processing and analysis, as well as for use by therapy management engine 114 to provide therapy recommendations or guidance to the patient or other user. [0081] In certain aspects, the CLM sensor system 104 is configured to continuously measure L-DOPA and transmit the L-DOPA measurements to an electric medical records (EMR) system (not shown in FIG. 1). An EMR system is a software platform which allows for the electronic entry, storage, and maintenance of digital medical data. An EMR system is generally used throughout hospitals and/or other caregiver facilities to document clinical information on patients over long periods. EMR systems organize and present data in ways that assist clinicians with, for example, interpreting health conditions and providing ongoing care, scheduling, billing, and follow up. Data contained in an EMR system may also be used to create reports for clinical care and/or disease management for a patient. In certain aspects, the EMR may be in communication with therapy management engine 114 (e.g., via a network) for performing the techniques described herein. In other aspects, an EMR may be mined for population-level health statistics, health economics, and the generation of clinical evidence or assessment of healthcare outcomes. In particular, as described herein, therapy management engine 114 may obtain data associated with a user, use the obtained data as input into one or more trained model(s), and output a prediction. In some cases, the EMR may provide the data to therapy management engine 114 to be used as input into one or more models, e.g., machine learning (ML) models. Further, in some cases, therapy management engine 114, after making a prediction, may provide the output prediction to the EMR.

[0082] Therapy management engine 114 refers to a set of software instructions with one or more software modules, including data analysis module (DAM) 116. In certain aspects, therapy management engine 114 executes entirely on one or more computing devices in a private or a public cloud. In such aspects, application 106 communicates with therapy management engine 114 over a network (e.g., Internet). In some other aspects, therapy management engine 114 executes partially on one or more local devices, such as display device 107 and/or CLM sensor system 104, and partially on one or more computing devices in a private or a public cloud. In some other aspects, therapy management engine 114 executes entirely on one or more local devices, such as display device 107 and/or CLM sensor system 104. As discussed in more detail herein, therapy management engine 114 may provide therapy recommendations to the patient or other user via application 106. Therapy management engine 114 provides therapy recommendations based on information included in patient profile 118. [0083] Patient profile 118 may include information collected about the patient from application 106. For example, application 106 provides a set of inputs 128, including the L-DOPA measurements received from CLM sensor system 104, that are stored in patient profile 118. In certain aspects, inputs 128 provided by application 106 include other data in addition to L-DOPA measurements received from CLM sensor system 104. For example, application 106 may obtain additional inputs 128 through manual user input, a medical device such as an L-DOPA pump, one or more symptom feedback sensors, other applications executing on display device 107, etc. The symptom feedback sensors can provide, for example, symptom information that measures the patient’s physical response to current L-DOPA levels. The symptom information can indicate, for example, a physical impact of L-DOPA and/or PD using one or more hardware or software sensors. In some aspects, the measurements can be used to infer a patient’s degree of motor symptom control (e.g., via Fourier or wavelet analysis ). Examples of the symptom feedback sensors will be discussed in greater detail relative to FIG. 2. Inputs 128 of patient profile 118 provided by application 106 are described in further detail below with respect to FIG. 3.

[0084] DAM 116 of therapy management engine 114 is configured to process the set of inputs 128 to determine one or more metrics 130. Metrics 130, discussed in more detail below with respect to FIG. 3, may, at least in some cases, be generally indicative of the health or state of a patient, such as one or more of the patient’s physiological state, trends associated with the health or state of a patient, etc. In certain aspects, metrics 130 may then be used by therapy management engine 114 as input for providing guidance to the patient or other user. As shown, metrics 130 are also stored in patient profile 118.

[0085] Patient profile 118 also includes demographic information 120, disease info 122, and/or medication information 124 (e.g., type of medication, brand of medication, dosage, frequency of administration). In certain aspects, such information may be provided through user input or obtained from certain data stores (e.g., electronic medical records (EMRs), etc.). In certain aspects, demographic information 120 may include one or more of the patient’s age, body mass index (BMI), ethnicity, gender, etc. In certain aspects, disease info 122 may include information about a condition of a patient, such as a stage (if known) according to a disease staging measurement protocol, co-morbidities, etc. In certain aspects, information about a patient’s condition may also include the length of time since PD diagnosis, the level of control, level of compliance with condition management therapy, other types of diagnosis (e.g., heart disease, obesity) or measures of health (e.g., heart rate, exercise, stress, sleep, etc.), and/or the like.

[0086] In certain aspects, medication information 124 may include information about the amount, frequency, and type of a medication taken by a patient. In certain aspects, the amount, frequency, and type of a medication taken by a patient is time-stamped and correlated with the patient’s L-DOPA levels, thereby indicating the impact that the amount, frequency, and type of the medication had on the patient’s L-DOPA levels. In certain aspects, medication information 124 may include, for example, information about the prescribed dosage/frequency of L-DOPA and the consumption of one or more MAO-B inhibitors (e.g., Rasagiline (Azilect) and selegiline (Eldepryl) and/or dopamine agonists (e.g., Ropinirole (Requip), pramipexole (Mirapex), rotigotine (Neupro), and apomorphine (Apokyn)). Further, medication information 124 may include inhibitor action curves, and/or pharmacokinetic and/or pharmacodynamics properties to determine medication effectiveness, etc. MAO-B inhibitors may be prescribed to a patient for the purpose of managing PD symptoms.

[0087] As described in more detail below, therapy management system 100 may be configured to use medication information 124 to determine medication effectiveness and/or an optimal medication dosage and frequency for different patients. In particular, therapy management system 100 may be configured to identify one or more optimal prescriptions based on the health of the patient when one or more medications are prescribed, as well as the condition(s) of the patient to be treated. In certain aspects, the medication information 124 may include information about other medicaments or treatments. For example, for patients that have received a deep brain stimulator (DBS), the medication information 124 can include information related to a DBS waveform including, for example, frequency, repetition rate, magnitude of stimulus, and/or the like. By way of further example, for patients that have received a body-adorned bioelectronics device configured for transcutaneous afferent patterned stimulation (TAPS), the medication information 124 can include information related to configurations for tremor treatment via individualized stimulation of nerves. The medication information 124 may include information manually provided by the patient or other user and/or information provided by the CLM sensor system 104.

[0088] In certain aspects, patient profile 118 is dynamic because at least part of the information that is stored in patient profile 118 may be revised over time and/or new information may be added to patient profile 118 by therapy management engine 114 and/or application 106. Accordingly, information in patient profile 118 stored in patient database 110 provides an up-to-date repository of information related to a patient.

[0089] Patient database 110, in some aspects, refers to a storage server that operates in a public or private cloud. Patient database 110 may be implemented as any type of data store, such as relational databases, non-relational databases, key-value data stores, file systems including hierarchical file systems, and the like. In some exemplary implementations, patient database 110 is distributed. For example, patient database 110 may comprise a plurality of persistent storage devices, which are distributed. Furthermore, patient database 110 may be replicated so that the storage devices are geographically dispersed.

[0090] The patient database 110 may include patient profiles 118 associated with a plurality of patients who similarly interact with application 106 executing on the display devices 107 of the other patients. Patient profiles stored in patient database 110 may be accessible to not only application 10 but therapy management engine 114 as well. Patient profiles in the patient database 110 may be accessible to the application 106 and the therapy management engine 114 over one or more networks (not shown). As described above, the therapy management engine 114, and more specifically the DAM 116 of the therapy management engine 114, can fetch inputs 128 from the patient database 110 and compute a plurality of metrics 130 which can then be stored as application data 126 in the patient profile 118, and/or used in population data and statistics as may be required or desired to be used by the application 106. In some aspects, the patient database 110 may be used to train an inference engine to provide patients with better management of their fine motor symptoms or otherwise help predict medication wearing off, dyskinesia, etc.

[0091] In certain aspects, the patient profiles 118 stored in patient database 110 may also be stored in a historical records database 112. The patient profiles 118 stored in the historical records database 112 may provide a repository of up-to-date information and historical information for each patient or other user of the application 106. Thus, the historical records database 112 essentially provides all data related to each patient or other user of the application 106, where data is stored according to an associated timestamp. The timestamp associated with information stored in the historical records database 112 may identify, for example, when information related to a patient has been obtained and/or updated. [0092] Further, the historical records database 112 may maintain time series data collected for patients over a period of time, including for patients who use the CLM sensor system 104 and the application 106. For example, L-DOPA data for a patient who has used the CLM sensor system 104 and the application 106 for a period of five years may have time series L-DOPA data, associated with the patient, maintained over the five-year period. Further, in certain aspects, the historical records database 112 may also include data for one or more patients who are not users of the CLM sensor system 104 and/or the application 106. Data stored in the historical records database 112 may be referred to herein as population data.

[0093] Data related to each patient stored in the historical records database 112 may provide time series data collected over the disease lifetime of the patient. For example, the data may include information about the patient prior to being diagnosed and information associated with the patient during the lifetime of the treatment, including information related to level of treatment required, information related to other diseases or conditions or other relevant co-morbidities, demographic information, etc. Such information may indicate symptoms of the patient, physiological states of the patient, states/conditions of one or more organs of the patient, habits of the patient (e.g., activity levels, food consumption, etc.), medication prescribed, etc., throughout the lifetime of the treatment.

[0094] Although depicted as separate databases for conceptual clarity, in some aspects, the patient database 110 and the historical records database 112 may operate as a single database. In other words, the historical and current data related to users of the CLM sensor system 104 and the application 106, as well as historical data related to patients that were not previously users of the CLM sensor system 104 and the application 106, may be stored in a single database. The single database may be a storage server that operates in a public or private cloud or in another arrangement.

[0095] As mentioned previously, the therapy management system 100 is configured to provide a treatment recommendation for a patient using the CLM sensor system 104 including one or more L-DOPA sensors. In certain aspects, the therapy management engine 114 is configured to provide real-time and/or non-real-time therapy management based on L-DOPA levels to the patient and/or other users, including but not limited to, healthcare providers, family members of the patient, caregivers of the patient, researchers, artificial intelligence (Al) engines, and/or other individuals, systems, and/or groups supporting care or learning from the data. In particular, the therapy management engine 114 may be used to collect information associated with a patient in the patient profde 118, to perform analytics thereon for recommending treatments (e.g., recommending an optimal dosage of a MAO-B inhibitor or a dopamine agonist). The therapy management engine 114 may also be used to collect information for pharmaceutical research to develop new therapies or more efficacious therapies. The patient profile 118 may be accessible to the therapy management engine 114 over one or more networks (not shown) for performing such analytics.

[0096] In certain aspects, the therapy management system 100 is designed to predict the risk or likelihood of, or the presence and/or severity of, PD symptoms in real-time (including near realtime) or within a specified period of time for a patient. In certain aspects, to enable such prediction, the therapy management engine 114 is configured to collect information associated with a patient in the patient profile 118 stored in the patient database 110, to perform analytics thereon for: (1) automatically detecting and classifying L-DOPA levels; (2) predicting risk, likelihood, and/or severity of PD symptoms; and/or (3) assessing the effectiveness of the current treatment and other potential treatment dosages and frequencies.

[0097] In certain aspects, the therapy management engine 114 may utilize one or more trained machine learning models capable of determining the probability of the presence and/or occurrence of PD symptoms and/or treatment recommendation for a patient based on information provided by patient profile 118. In the illustrated aspect of FIG. 1, the therapy management engine 114 may utilize trained machine learning model(s) provided by a training server system 140. Although depicted as a separate server for conceptual clarity, in some aspects, the training server system 140 and the therapy management engine 114 may operate as a single server. That is, the model may be trained and used by a single server (e.g., a local device, a microprocessor, etc.) or may be trained by one or more servers and deployed for use on one or more other servers. In certain aspects, the model may be trained on one or many virtual machines (VMs) running, at least partially, on one or many physical servers in relational and or non-relational database formats.

[0098] The training server system 140 is configured to train the machine learning model(s) using training data, which may include data (e.g., from patient profiles) associated with one or more patients (e.g., users or non-users of CLM sensor system 104 and/or application 106) previously treated for PD, as well as patients not treated for PD (e.g., healthy patients). The training data may be stored in the historical records database 112 and may be accessible to the training server system 140 over one or more networks (not shown) for training the machine learning model(s). The training data may also, in some cases, include patient-specific data for a patient over time.

[0099] In some aspects, the training data refers to a dataset that has been featurized and labeled. For example, the dataset may include a plurality of data records, each including information corresponding to a different patient profile stored in the patient database 110, where each data record is featurized and labeled. In machine learning and pattern recognition, a feature is an individual measurable property or characteristic. Generally, featurizing is performed by selecting one or more features of the dataset that best characterize patterns in the data. These features may be used to create one or more predictive machine learning models. Data labeling is the process of adding one or more meaningful and informative labels to provide context to the data for learning by the machine learning model.

[0100] As an example, each relevant characteristic of a patient, which is reflected in a corresponding data record, may be a feature used in training the machine learning model. Such features may include age, gender, weight, height, body mass index, any therapies currently taken, when a therapy was last applied (e.g., L-DOPA), how much of a therapy was applied (e.g., units of L-DOPA), change (e.g., delta) in L-DOPA levels from a first timestamp to a second timestamp, change (e.g., delta) in L-DOPA thresholds of a patient under treatment for PD from a first timestamp to a subsequent timestamp, the derivative of the measured linear system of L-DOPA measurement at a point at a specific timestamp, rates of change in the slope of increase or decrease in L-DOPA values, etc. In addition, the data record may be labeled with an indication as to a PD diagnosis, an assigned severity, prescription information (e.g., dosage and frequency of consumption) for one or more other medications, and/or the like.

[0101] The model(s) are then trained by the training server system 140 using the featurized and labeled training data. In particular, the features of each data record may be used as input into the machine learning model(s), and the generated output may be compared to label(s) associated with the corresponding data record. The model(s) may compute a loss based on the difference between the generated output and the provided label(s). This loss is then used to modify the internal parameters or weights of the model. By iteratively processing each data record corresponding to each historical patient, in certain aspects, the model(s) may be iteratively refined, and the loss minimized, to generate, within a prescribed level of confidence, treatment recommendations (e.g., an optimal dosage/frequency of taking inhibitors) and predictions associated with PD symptom risk, presence, progression, improvement (e.g., regression), and/or severity in a patient. Further, in certain other aspects, by iteratively processing each data record corresponding to each historical patient, in certain aspects, the model(s) may be iteratively refined to generate accurate treatment recommendations and predictions of the risk and/or presence of PD symptoms.

[0102] As illustrated in FIG. 1, the training server system 140 deploys these trained model(s) to the therapy management engine 114 for use during runtime. For example, the therapy management engine 114 may obtain the patient profile 118 associated with a patient, use information in the patient profile 118 as input into the trained model(s), and output a treatment recommendation and/or PD symptom prediction. The treatment recommendation may be indicative of an efficacy of a current treatment based on the medication information 124, the L- DOPA data, and the like. In some aspects, the treatment recommendation includes a modification to an existing treatment or a recommendation for an alternative treatment based on the efficacy of a current treatment. For example, the treatment recommendation may include a change to a current dosage and/or frequency, a change in medication type, or a notice to consult a healthcare provider, etc.

[0103] The therapy management engine 114 may provide a prediction which may be indicative of the presence and/or severity of PD symptoms for the patient in real-time or within a certain time (e.g., shown as the output 144 in FIG. 1). For the example, the prediction can be, or can include, a prediction of the patient’s physical response to an L-DOPA dosage. The output 144 generated by the therapy management engine 114 may also provide one or more recommendations for treatment based on the predictions. The output 144 may be provided to the patient (e.g., through application 106), to a patient’s caretaker (e.g., a parent, a relative, a guardian, a teacher, a nurse, etc.), to a patient’s physician, or any other individual that has an interest in the wellbeing of the patient for purposes of improving the patient’s health, such as, in some cases by effectuating the recommended treatment.

[0104] In certain aspects, the patient’s own data is used to personalize the one or more models that are initially trained based on population data. For example, a model (e.g., trained using population data) may be deployed for use by therapy management engine 114 to provide a treatment recommendation and/or predict the presence and/or severity of PD symptoms for a specific patient. In some aspects, sometime after making a prediction using the model, the therapy management engine 114 may be configured to ask the patient, or a caretaker, physician, etc., whether the medication information 124 should be updated based on the recommended change in treatment. In other aspects, the therapy management engine 114 may provide a query as to whether the predicted presence and/or severity of PD symptoms was confirmed by, e.g., other methods (e.g., symptom information from symptom feedback sensors), and/or therapy management engine 114 may use one or more queries to the patient or other user. In some cases, the patient’s answer may indicate or deny the presence and/or severity of PD symptoms. Accordingly, the model may continue to be retrained and/or personalized using updated medication information 124, the patient’s answer, and/or symptom information from symptom feedback sensors. While specific examples are given, other data may also be used as input into the model to personalize the model for the patient.

[0105] In certain aspects, the output 144 generated by the therapy management engine 114 may be stored in the patient profile 118. In certain aspects, the output 144 may be patient-specific treatment recommendations, treatment efficacy, identification of one or more indicators of the presence and/or severity of PD symptoms, and the like. For example, in certain aspects, the output 144 may be a treatment recommendation for an update in medication, medication dosage, medication frequency of use, prediction as to the presence and/or severity of PD symptoms in a patient, and the like. In certain aspects, the output 144 may be a prediction as to the risk of the onset of PD symptoms, or of PD symptoms of a given type or severity. In certain aspects, the output 144 may be patient-specific treatment decisions or recommendations for management of PD symptoms for the patient. In specific aspects, the output 144 may be a recommendation relating to the use of L-DOPA, MAO-B inhibitors, dopamine agonist, DBS, TAPS, etc.

[0106] In some aspects, the output 144 stored in the patient profile 118 may be continuously updated by the therapy management engine 114. Accordingly, previous diagnoses and/or physiological parameters of the patient associated with PD management, originally stored as the outputs 144 in the patient profile 118 in the patient database 110 and then passed to the historical records database 1 12, may provide an indication of the effectiveness of the current treatment or may provide a likelihood of PD symptoms in a patient, or a likelihood of PD symptoms of a given severity, in a given time period. Additionally, previous diagnoses and/or physiological parameters of the patient associated with how well medication was tolerated and/or how efficacious a certain type/dose/frequency of administration of the medication was, originally stored as the outputs 144 in the patient profile 118 in the patient database 110 and then passed to the historical records database 112, may provide an indication of the effectiveness of the current treatment or may provide a likelihood of such PD symptoms in a patient in a given time period.

[0107] In certain aspects, a patient’s own historical data may be used to provide therapy management and insight around the patient’s L-DOPA levels and/or control of PD symptoms. For example, a patient’s historical data may be used by an algorithm as a baseline to indicate improvements or deterioration in the patient’s condition. As an illustrative example, a patient’s data from two weeks prior may be used as a baseline that can be compared with the patient’s current data to identify an improvement or deterioration in control of L-DOPA levels of the patient (e.g., relative to a personalized therapeutic window) and/or in the prevalence and/or severity of PD symptoms of the patient and, thereby, whether the risk associated with future PD symptoms (e.g., of a given type or severity) has increased or decreased. In certain aspects, the patient’s own historical data may be used by the training server system 140 to train a personalized model that may further be able to predict the presence and/or severity of PD symptoms, optimal treatments for reducing the predicted presence and/or severity of PD symptoms, and/or improvement/deteri oration in control of the patient’s L-DOPA levels (e.g., relative to a personalized therapeutic window) and/or the severity of PD symptoms based on the patient’s recent pattern of data (e.g., exercise data, food consumption data, etc.).

[0108] In certain aspects, the model may be trained to provide lifestyle recommendations, exercise recommendations, food intake recommendations, recommendations for adjusting a DBS waveform (e.g., frequency, repetition rate, magnitude of stimulus, etc.), and other types of therapy recommendations to help the patient improve treatment or prevent onset and/or progression of PD symptoms based on the patient’s historical data, including how different types of medication, food, and treatment (e.g., medication type, dosage, frequency) have impacted PD symptoms in the past. In certain aspects, the model may be trained to predict the underlying cause of certain improvements or deteriorations in PD symptom severity. For example, the application 106 may display a user interface with a graph that shows the patient’s symptoms or a measure thereof with trend lines and indicate, e.g., retrospectively, what caused the change in symptoms at certain points in time (e.g., administration of L-DOPA, administration of other medication such as a MAO-B inhibitor or a dopamine agonist, etc ).

[0109] FIG. 2 is a diagram 200 conceptually illustrating an example CLM sensor system 104 including an example continuous L-DOPA sensor(s) with sensor electronics, in accordance with certain aspects of the present disclosure. For example, the CLM sensor system 104 may be configured to continuously monitor L-DOPA levels of a patient, in accordance with certain aspects of the present disclosure.

[0110] Generally, real-time or continuous measurements of L-DOPA levels, rates of change, trends, clearance rates, and/or other L-DOPA data, as measured in interstitial fluid or blood by a CLM sensor system, can be used to provide treatment recommendations to the patient or other user. Such data can indicate a change in L-DOPA levels indicative of a treatment that is less than ideal. Therefore, continuous L-DOPA monitoring provides earlier, and/or improved treatment recommendations, such as improving the titration of pharmacologic agents with narrow therapeutic windows or evolving pharmacokinetic profiles. Some aspects may provide screening, diagnosis, prognosis, and/or staging of PD.

[0U1] In certain aspects, clinical indicators may be used to determine whether a CLM sensor system, e.g., the CLM sensor system 104, may be needed to assess an efficacy of a treatment or to assess a risk, likelihood, presence, and/or severity of PD symptoms in a patient. In one example, such clinical indicators include glucose measurements, ketone measurements, lactate measurements, cortisol measurements, dissolved oxygen measurements, ion measurements, blood pressure measurements, renal metrics, hydration measurements, physical activity metrics, sleep metrics, heart rate, respiration rate, core temperature, nutrition information, etc. L-DOPA levels and other information may generally indicate a needed optimization of a medication dosage and/or frequency or may indicate the risk, likelihood, presence, and/or severity of PD symptoms.

[0112] In yet another example, clinical indicators may include an assessment of patient adherence to treatment. A comparison of a prescribed treatment to an actual treatment to quantify how well a patient is complying with the prescribed treatment may allow for a more accurate assessment of the efficacy of the prescribed treatment. The comparison may be useful to healthcare providers to further adjust treatment or provide additional treatment instruction/education to the patient. The comparison may also be of interest to healthcare insurers and/or healthcare payers with respect to reimbursement or other considerations (e.g., rewards, discounts, etc.).

[0113] In yet another example, clinical indicators may include comorbidities often associated with, and/or increasing the risk, likelihood, presence, and/or severity of, PD symptoms. Such comorbidities may include, for example, cardiovascular disease, hypertension, diabetes, etc. In various aspects, analyte and/or non-analyte sensors can be incorporated, for example, into the therapy management system 100, to measure and provide guidance related to the comorbidities. For example, analyte data can be provided by a glucose sensor (e.g., for diabetes), a lactate sensor (e.g., for liver disease), a heart rate sensor, an ECG sensor, a blood pressure sensor (e g., for cardiovascular disease), combinations of the foregoing and/or the like. In various embodiments, such analyte and/or non-analyte sensors can provide helpful input, for example, to the therapy management system 100, to supplement a diagnosis of a disease listed. In various embodiments, data from such analyte and/or non-analyte sensors can provide near-term diagnostic, segmentation and/or severity information about the comorbidities.

[0114] As shown in FIG. 2, the CLM sensor system 104 in the illustrated aspect includes a sensor electronics module 204 and one or more continuous L-DOPA sensor(s) 202 (individually referred to herein as the continuous L-DOPA sensor 202 and collectively referred to herein as the continuous L-DOPA sensors 202) associated with a sensor electronics module 204. The sensor electronics module 204 may be in wireless communication (e.g., directly or indirectly) with one or more display devices 210, 220, 230, and 240. In certain aspects, the sensor electronics module 204 may also be in wireless communication (e.g., directly or indirectly) with an L-DOPA delivery device 208 (e.g., an L-DOPA pump), one or more symptom feedback sensors 206, and one or more other sensors 209. In other aspects, including, but not limited to, diagnostic implementations, the sensor electronics module 204 may be operated independently (e.g., unpaired with a display device) and queried at the end of a wear session to wirelessly transfer data logged during a session to a local device or cloud database for future review, retrieval, or execution of further analytics.

[0115] In certain aspects, a continuous L-DOPA sensor 202 may comprise a sensor for detecting and/or measuring L-DOPA. The continuous L-DOPA sensor 202 may be configured to continuously measure L-DOPA as a non-invasive device, a subcutaneous device, a transcutaneous device, a transdermal device, a dermal device, an intradermal device, a subdermal device, implanted device, and/or an intravascular device. In certain aspects, the continuous L-DOPA sensor 202 may be configured to continuously measure L-DOPA levels of a patient using one or more measurement techniques, such as enzymatic, immunometric, aptameric, amperometric, voltametric, potentiometric, impedimetric, conductimetric, conductometric, capacitive, chemical, physical, electrochemical, spectrophotometric, polarimetric, calorimetric, iontophoretic, radiometric, immunochemical, optical, ion-selective, photoplethysmological, and the like. In certain aspects, the continuous L-DOPA sensor 202 provides a data stream indicative of the concentration of L-DOPA in the patient. The data stream may include raw data signals which may be converted into a calibrated and/or filtered data stream used to provide estimated L-DOPA value(s) to the patient or other user.

[0116] In certain aspects, the sensor electronics module 204 includes electronic circuitry associated with measuring and processing the continuous L-DOPA sensor data, including prospective algorithms associated with processing and calibration of the sensor data. The sensor electronics module 204 can be physically connected to the continuous L-DOPA sensor(s) 202 and can be integral with (non-releasably attached to) or releasably attachable to the continuous L- DOPA sensor(s) 202. The sensor electronics module 204 may include hardware, firmware, and/or software that enables measurement of levels of L-DOPA via a continuous L-DOPA sensor(s) 202. For example, the sensor electronics module 204 can include an electrochemical analog front end (e.g., potentiostat, galvanostat, impedance measurement device), a power source for providing power to the sensor, a microprocessor for executing an embedded data processing or algorithmic routine, other components useful for signal processing and data storage, and a telemetry module for transmitting data from the sensor electronics module to one or more display devices or a centralized data repository. Electronics can be affixed to a printed circuit board (PCB), flexible PCB (flexPCB), or the like, and can take a variety of forms. For example, the electronics can take the form of an integrated circuit (IC), such as an Application-Specific Integrated Circuit (ASIC), a field-programmable gate array (FPGA), a system-on-a-chip (SoC), a microcontroller, and/or a processor.

[0117] In some aspects, the display devices 210, 220, 230, and/or 240 are configured for displaying displayable sensor data, including L-DOPA data, which may be transmitted by the sensor electronics module 204. Each of the display devices 210, 220, 230, or 240 can include a display such as a touchscreen display 212, 222, 232, or 242 for displaying sensor data to the patient or other user and/or receiving inputs from the patient or other user. For example, a graphical user interface (GUI) may be presented to the patient or other user for such purposes. In some aspects, the display devices 210, 220, 230, and 240 may include other types of user interfaces such as a voice user interface instead of, or in addition to, a touchscreen display for communicating sensor data to the user of the display device and/or receiving user inputs. The display devices 210, 220, 230, and 240 may be examples of the display device 107 illustrated in FIG. 1 used to display sensor data to a user of FIG. 1 and/or receive input from the user.

[0118] In some aspects, one, some, or all of the display devices are configured to display or otherwise communicate the sensor data as it is communicated from the sensor electronics module (e.g., in a data package that is transmitted to respective display devices), without any additional prospective processing required for calibration and real-time display of the sensor data.

[0119] The plurality of display devices may include a custom display device specially designed for displaying certain types of displayable sensor data associated with L-DOPA data received from sensor electronics module. In certain aspects, the plurality of display devices may be configured for providing alerts/alarms/notifications based on the displayable sensor data. The display device 210 is an example of such a custom device. In some aspects, one of the plurality of display devices is a smartphone, such as the display device 220 which represents a mobile phone, using a commercially available operating system (OS), and configured to display a graphical representation of the continuous sensor data (e.g., including current and historic data). Other display devices can include other hand-held devices, such as the display device 230 which represents a tablet, the display device 240 which represents a smartwatch, the L-DOPA delivery device 208 (e.g., an L-DOPA pump), and/or a desktop or laptop computer (not shown).

[0120] Because different display devices provide different user interfaces, the content of the data packages (e.g., amount, format, and/or type of data to be displayed, alarms, and the like) can be customized (e.g., programmed differently by the manufacture and/or by an end user) for each particular display device. Accordingly, in certain aspects, a plurality of different display devices can be in direct wireless communication with a sensor electronics module (e.g., such as an on-skin sensor electronics module 204 that is physically connected to continuous L-DOPA sensor(s) 202) during a sensor session to enable a plurality of different types and/or levels of display and/or functionality associated with the displayable sensor data. In certain aspects, the type of alarms customized for each particular display device, the number of alarms customized for each particular display device, the timing of alarms customized for each particular display device, and/or the threshold levels configured for each of the alarms (e.g., for triggering) are based on the output 144 (e.g., as mentioned, the output 144 may be indicative of the current health of a patient, the state of a patient’s L-DOPA levels and/or PD symptoms, current treatment recommended to a patient, and/or physiological parameters of a patient) stored in the patient profile 118 for each patient.

[0121] As mentioned, the sensor electronics module 204 may be in communication with the other sensor(s) 209 and the L-DOPA delivery device 208. The other sensor(s) 209 may include, for example, sensors of glucose, lactate, ketone, cortisol, ions, glycerol, amino acid, free fatty acid, and/or other analytes. The L-DOPA delivery device 208 may be, for example, a continuous L- DOPA pump for administering L-DOPA to a patient. In certain aspects, the L-DOPA delivery device 208 provides small doses of L-DOPA according to a configurable basal rate. In certain aspects, the continuous L-DOPA pump is operable to provide additional doses of L-DOPA as a bolus. In certain aspects, the therapy management engine 114, for example, is operable to adjust the basal rate, recommend an adjustment to the basal rate, recommend a bolus, initiate a bolus, and/or the like, in response to symptom information.

[0122] Further, as mentioned, the sensor electronics module 204 may also be in communication with the symptom feedback sensor(s) 206. The symptom feedback sensor(s) 206 may include one or more hardware or software sensors that are operable to provide symptom information such as, for example, information indicative of PD symptoms and/or symptoms associated with side effects of medication such as L-DOPA. In some examples, the symptom information can measure the patient’s physical response to current L-DOPA levels. The symptom information can indicate, for example, a physical impact of L-DOPA and/or PD using one or more hardware or software sensors. In some aspects, the measurements can be used to infer a patient’s degree of motor symptom control. The symptom feedback sensor(s) 206 can include, for example, a vocal biomarker sensor, a mobile device with an inertial measurement unit (e.g., an accelerometer, gyroscope, or inclinometer), an sEMG sensor, or another suitable device or component. In other examples, the symptom feedback sensor(s) 206 can include sensors for one or more of EEG, ECG, HRV, heart rate, respiration rate, blood pressure, body impedance, skin conductance, sleep monitoring, body sounds (e.g., for gastrointestinal acoustic monitoring), voice recognition, electrogastrography (e.g., for gastroparesis), and/or the like. In still other examples, the symptom feedback sensor(s) 206 can include software that provides, for example, behavioral information (e.g., an amount of device usage) or indicators of cognitive function (e.g., results of puzzles or tests). One or more of these symptom feedback sensor(s) 206 may provide data to the therapy management engine 114 described further below. In some aspects, a patient or other user may manually provide some of the data for processing by the training server system 140 and/or the therapy management engine 114 of FIG. 1.

[0123] In an example, the symptom feedback sensor(s) 206 can include a vocal biomarker sensor. The vocal biomarker sensor can be any patient device having, for example, a microphone and access to a software service for analyzing speech relative to one or more predetermined speech function signatures. The vocal biomarker sensor can be embodied, for example, on any of the display devices 107, 210, 220, 230, and/or 240 discussed previously, the CLM sensor system 104, another system in network communication with any of the foregoing devices, and/or any combination of the foregoing devices or systems.

[0124] In general, the vocal biomarker sensor may provide time-indexed information related to a patient’s speech function, such as time-indexed values indicative of a degree to which a patient’s speech function at a given time, or over a given period of time, matches one or more speech function signatures. The vocal biomarker sensor can record the patient’ s voice, in real time, to assess vocal biomarkers. The vocal biomarker sensor can thereby analyze how the patient speaks (e.g., prosody, intonation, pitch, etc.) for the one or more speech function signatures. The speech function signatures can each be indicative, for example, of muscle tremors or other physiological symptoms manifested in the patient’s speech. In some aspects, the analysis performed by the vocal biomarker sensor can include, for example, a Fourier-based decomposition or analysis of speech frequency content.

[0125] In some aspects, the vocal biomarker sensor can operate automatically. For example, in some aspects, the microphone of the vocal biomarker sensor can actively listen to a surrounding environment for the patient’s voice. A natural language processing (NLP) or pattern recognition algorithm can be used to extricate the patient’ s voice from noise and/or other voices in the patient's vicinity. In some aspects, operation of the vocal biomarker sensor can be triggered by the patient. For example, a keyword (e.g., 'Dexcom' or 'What's my L-DOPA level') can be used to query the vocal biomarker function. Either the keyword will be analyzed via the embedded algorithm or the ensuing speech will be analyzed.

[0126] In some aspects, once operation of the vocal biomarker sensor is triggered, a rules- based algorithm (e g., if-then, do-while, catch statements) can identify elements of speech that are indicative of deviations from a baseline (e.g. physiologic) state. The elements can include, for example, increased slurring, jitters, shimmers, pauses and/or the like relative to the baseline.

[0127] In some aspects, rules of the rules-based algorithm can be configured based on a patient-specific baseline. For example, the patient can be asked to dictate a generic text comprising several sentences. Optionally, the patient can be queried how they are feeling emotionally (e.g., stressed), physically (e.g., ill with fever), and/or with regard to symptoms (e.g., “high” or “low” motor function symptoms). In some aspects, a training regimen with text dictation can also be accompanied by an assessment by a physician and/or the provision of diagnostic data (e.g., laboratory results).

[0128] In some aspects, a model-based algorithm (e.g., machine learning) is applied to identify elements of speech that are indicative of deviations from a baseline (e.g., physiologic) state. The elements can include, for example, increased slurring, stuttering, jitters, shimmers, pauses and/or the like relative to the baseline. In some aspects, the model-based algorithm can be trained using either supervised or unsupervised learning methods. For example, the patient can be asked to dictate a generic text comprising several sentences. Optionally, the patient can be queried about how they are feeling emotionally (i.e., stressed) and/or physically (i.e., ill with fever). In some aspects, a training regimen with text dictation can also be accompanied by an assessment by a physician and/or the provision of diagnostic data (e.g., laboratory results).

[0129] In another example, the symptom feedback sensor(s) 206 can include a device having an inertial measurement unit (e.g., an accelerometer, gyroscope, and/or inclinometer). In general, the device may provide time-indexed information based on an output of the inertial measurement unit and an analysis of the same using software on the same or different device. The time-indexed information can include, for example, time-indexed tremor values, information related to a number of falls, and/or similar physiological information. [0130] In another example, the symptom feedback sensor(s) 206 and/or the display devices 107, 210, 220, 230, and/or 240 can include device-based behavior monitoring features such as but not limited to challenge tests on the phone or duration or frequency of use monitoring. For example, if a patient’s mobile phone utilization drops below the expected level that could indicate they are unable to sufficiently operate the phone with their current symptomatic state and they require a need for intervention. On a periodic basis, for example once per day, the patient can be presented with a challenge test such as tracing an Archimedes spiral around the phone following a dot. This could give an indication as to the patient’s symptom severity. This test could also be presented automatically when other measures are met to determine that it is likely the patient has had a decline in function but confirmation from an accepted clinical measure would be beneficial.

[0131] In another example, the symptom feedback sensor(s) 206 can include sensors for monitoring for patient activity such as movement, stride length, speech patterns and frequency, ability to hear sounds in a crowd and reason back a response, and patient reported outcomes from survey or question data can also be used to determine patient’s symptom frequency and severity.

[0132] In another example, the symptom feedback sensor(s) 206 can include a sensor for monitoring blood pressure (i.e., a blood pressure monitor). PD can sometimes cause cardiovascular changes that become evident in blood pressure. In addition, oral and infused L-DOPA commonly induce a reduction of blood pressure in humans. According to clinical research, L-DOPA intake significantly reduces systolic and diastolic blood pressure, heart rate and plasma noradrenaline and adrenaline in both the supine and upright positions. A significant reduction in stroke volume and cardiac output has also been seen with L-DOPA. While L-DOPA can have beneficial effects on blood pressure, L-DOPA can also cause or exaggerate hypotension for susceptible patients.

[0133] In certain aspects, blood pressure changes are most noticeable while the patient is sleeping. Therefore, in various aspects, the blood pressure monitor can be used to monitor average nocturnal blood pressure measured over long periods of time (e.g., over a period of six months) to help estimate disease progression and symptom control. In various aspects, monitoring blood pressure as a surrogate analyte for biological effectiveness of L-DOPA can clarify the dosage needed throughout the day and at night. In addition, or alternatively, blood pressure can be used to flag a potential falling or fainting event (e.g., due to hypotension induced or aggravated by L- DOPA). [0134] In another example, the symptom feedback sensor(s) 206 can include an sEMG sensor. In various aspects, the sEMG sensor can monitor electrical activity produced by muscles controlled by the somatic nervous system, often referred to as skeletal muscles. In some aspects, the sEMG sensor can be used in combination with an inertial measurement unit of the type discussed above to detect and track, for example, to detect and track muscle rigidity and freeze of gait. For example, outputs of the sEMG sensor and/or the inertial measurement unit can be cross-referenced with signatures corresponding to physical events (e.g., muscle tremors).

[0135] Table 2 below provides examples of various sensors and corresponding utilization in the therapy management system 100.

Table 2

[0136] In certain aspects, a wireless access point (WAP) may be used to couple one or more of the CLM sensor system 104, the plurality of display devices, the L-DOPA delivery device 208, and/or the symptom feedback sensor(s) 206 to one another. For example, the WAP may provide Wi-Fi, cellular, and/or loT (e.g., NB-IoT, LTE Cat-Mi) connectivity among these devices. Near Field Communication (NFC) and/or Bluetooth may also be used among devices depicted in the diagram 200 of FIG. 2.

[0137] FIG. 3 illustrates a diagram 300 of example inputs and example metrics that are calculated based on the inputs for use by the therapy management system 100 of FIG. 1, in accordance with certain aspects of the present disclosure. In particular, FIG. 3 provides a more detailed illustration of example inputs and example metrics introduced in FIG. 1.

[0138] FIG. 3 shows example inputs 128 on the left, the application 106 and the therapy management engine 114 including the DAM 116 in the middle, and metrics 130 on the right. In certain aspects, each one of the metrics 130 may correspond to one or more values, e.g., discrete numerical values, ranges, or qualitative values (high/medium/low, stable/unstable, rate of change, points of inflection, etc.). The application 106 obtains the inputs 128 through one or more channels (e.g., manual user input, sensors/monitors, other applications executing on the display device 107, EMRs, etc.). As mentioned previously, in certain aspects, the inputs 128 may be processed by the DAM 116 and/or the therapy management engine 114 to output the metrics 130. The inputs and metrics 130 may be used by the therapy management engine 114 to provide therapy management to the patient. For example, the inputs 128 and the metrics 130 may be used by the training server system 140 to train and deploy one or more machine learning models for use by the therapy management engine 114 for providing therapy management around treatment of the patient.

[0139] In certain aspects, starting with the inputs 128, food consumption information may include information about one or more of meals, snacks, and/or beverages, such as one or more of the size, content (milligrams (mg) of sodium, potassium, carbohydrate, fat, protein, etc.), sequence of consumption, and time of consumption. In certain aspects, food consumption may be provided by a user through manual entry, by providing a photograph through an application that is configured to recognize food types and quantities, by scanning a bar code or menu, and/or interrogating an NFC / RFID tag integrated into the packaging of the food item. In various examples, meal size may be manually entered as one or more of calories, quantity (e.g., “three cookies”), menu items (e.g., “Royale with Cheese”), and/or food exchanges (e.g., 1 fruit, 1 dairy). In some examples, meal information may be received via a convenient user interface provided by the application 106. In some examples, meal information may be provided via one or more other applications synchronized with the application 106, such as one or more other mobile health applications executed by the display device 107. In such examples, the synchronized applications may include, e.g., an electronic food diary application or photograph application.

[0140] In certain aspects, food consumption information entered by the patient or other user may relate to nutrients consumed by the patient. Consumption may include any natural or designed food or beverage. Food consumption information entered by a user may be related to L-DOPA levels. In some cases, food consumption information can be retrieved, in part, via an interface with a diet database and/or a dedicated food tracking application.

[0141] In some aspects, the food consumption information may include information about an impact certain foods have, for example, on an effectiveness of L-DOPA medication. For example, Table 3 below shows example interactive effects of example food types. In some aspects, such interactions can be identified and taken into account, for example, when recommending L-DOPA dosages as discussed herein. In addition, or alternatively, diet recommendations may also be generated and presented, for example, to influence or maximize L-DOPA absorption.

Table 3

[0142] In certain aspects, exercise information is also provided as an input. Exercise information may be any information surrounding activities, such as activities requiring physical exertion by the patient. For example, exercise information may range from information related to low intensity (e.g., walking a few steps) and high intensity (e.g., five mile run) physical exertion or it could take the form of a wattage (e.g., stationary cycle), speed (e.g., GPS-enabled smartwatch), and/or resistance (e.g., elliptical machine) over a specified time interval. In certain aspects, exercise information may be provided, for example, by an accelerometer sensor or a heart rate monitor on a wearable device such as a watch, fitness tracker, and/or patch. In certain aspects, exercise information may also be provided through manual user input, through workout machinery, and/or through a surrogate sensor and prediction algorithm measuring changes to heart rate (or other cardiac metrics). When predicting that a patient is exercising based on his/her sensor data, the patient may be asked to confirm if exercise is occurring, what type of exercise, and or the level of strenuous exertion being used during the exercise over a specific period. This data may be used to train the system to learn about the patient’s exercise patterns to reduce the need for confirmation questions as time progresses. Other analytes and sensor data may also be included in this training set, including analytes and other measured elements described herein including temporal elements, such as time and day. [0143] In certain aspects, patient statistics, such as one or more of age, height, weight, BMI, body composition (e.g., % body fat), stature, build, or other information may also be provided as an input. In certain aspects, patient statistics are provided through a user interface, by interfacing with an electronic source such as an electronic medical record, and/or from measurement devices. In certain aspects, the measurement devices include one or more of a wireless, e.g., Bluetooth- enabled, weight scale and/or camera, which may, for example, communicate with the display device 107 to provide patient data.

[0144] In certain aspects, treatment information is also provided as an input. Treatment information may include information about the type, dosage, and/or timing of when one or more medications (e.g., L-DOPA, MAO-B inhibitors, dopamine agonists etc.) are to be taken by the patient. The treatment information may include information regarding different lifestyle habits by the patient’s physician. For example, the patient’s physician may recommend that the patient increase their intake of water and fiber-rich foods, exercise for a minimum of thirty minutes a day, or increase an L-DOPA dosage or other medication to maintain, improve, and/or reduce PD symptoms. As another example, a healthcare professional may recommend that the patient engage in at-home treatment and/or treatment at a clinic. The treatment information may also indicate a patient’s compliance with the prescribed type, dosage, and/or timing of medications. For example, the treatment/medication information may indicate whether and when exactly and with what dosage/type the medication was taken.

[0145] In some aspects, the treatment information may include information about interactions between medications, such as an impact a given medication may have on an effectiveness of L-DOPA medication. For example, Table 4 below shows example interactive effects of example medications. In some aspects, such interactions can be identified and taken into account, for example, when recommending L-DOPA dosages as discussed herein.

Table 4

[0146] In certain aspects, continuous L-DOPA sensor data may also be provided as input, for example, through the CLM sensor system 104. In certain aspects, input may also be received from the symptom feedback sensor(s) 206 described with respect to FIG. 2. Input from the symptom feedback sensor(s) 206 may include any of the symptom information discussed previously.

[0147] In some aspects, the symptom feedback sensor(s) 206 may include an embedded scanner/reader to detect medication related information (e.g., type, brand, dosage, frequency). Examples of a scanner may include a reader configured to detect near-field communication (NFC) and/or radio frequency identification (RFID) information provided by a corresponding active or passive tag provided within the medication packaging or otherwise accompanying the medication. Another example of a scanner may be a barcode, QR, or other optical scanner capable of accessing information associated with a visual pattern provided on the packaging or otherwise associated with the medication.

[0148] In certain aspects, input may include input relating to the patient’s L-DOPA delivery. In particular, input related to the patient’s L-DOPA delivery may be received from an L-DOPA pump. L-DOPA delivery information may include one or more of L-DOPA manufacturer, L-DOPA dosage, L-DOPA formulation, L-DOPA volume, basal vs bolus dose, number of units of L-DOPA delivered, time of delivery, etc. Other parameters, such as L-DOPA action time or duration of L-DOPA action, may also be received as inputs.

[0149] In certain aspects, time may also be provided as an input, such as time of day, UTC time or time from a real-time clock. Said real-time clock may be provided externally (synchronized to a server via a Wi-Fi, cellular, or Bluetooth wireless connection) or may be embedded as an integrated real-time clock (RTC) circuit within the wearable / sensor electronics. For example, in certain aspects, input L-DOPA data may be timestamped to indicate a date and time when the L- DOPA measurement was taken for the patient.

[0150] User input of any of the above-mentioned inputs 128 may be through a user interface, such a user interface of the display device 107 of FIG. 1.

[0151] As described above, in certain aspects, the DAM 116 and/or the therapy management engine 114 (e.g., using one or more trained models) determines or computes the patient’s metrics 130 based on the inputs 128. An example list of the metrics 130 is shown in FIG. 3.

[0152] In certain aspects, L-DOPA levels and L-DOPA rates of change may be determined from sensor data (e.g., L-DOPA measurements obtained from CLM sensor system 104). For example, an L-DOPA level rate of change refers to a rate that indicates how one or more time- stamped L-DOPA measurements or values change in relation to one or more other time-stamped L-DOPA measurements or values. L-DOPA level rates of change may be determined over one or more seconds, minutes, hours, days, etc. In some aspects, L-DOPA level rates of change may be affected by diet (e.g., temporal effects and/or sporadically different contradictory effects based on diet) as well as disease progression (e.g., chronic multi-day effect or building effect). [0153] In certain aspects, an L-DOPA trend may be determined based on L-DOPA levels over a certain period of time. In certain aspects, L-DOPA trends may be determined based on L-DOPA level rates of change over certain periods of time.

[0154] In certain aspects, an L-DOPA clearance rate may be determined from sensor data (e.g., L-DOPA levels obtained from the CLM sensor system 104) following the administration of a known, or estimated, amount of L-DOPA. L-DOPA trends may be indicative of an effectiveness of a medication type, dosage, and/or frequency.

[0155] In certain aspects, the L-DOPA clearance rate may be determined by calculating a slope between an initial high L-DOPA value (e.g., highest L-DOPA level during a period of 20-30 minutes after the administration of L-DOPA) at to and a subsequent low L-DOPA value at ti. The low L-DOPA value (LL) may be determined based on a patient’s initial high L-DOPA value (LH) and baseline L-DOPA value (LB) before the administration of L-DOPA. In certain aspects, LL can be an L-DOPA value between LH and LB, e.g., LL = LB + K*(LH - LB)/2, where K can be a percentage representing by how much a patient’s L-DOPA level returned to the patient’s baseline value. When K equals zero, the low L-DOPA value equals the baseline L-DOPA value. When K equals 0.5, the low L-DOPA value equals the mean L-DOPA value between the initial L-DOPA value and the baseline L-DOPA value.

[0156] In certain aspects, the L-DOPA clearance rate may be determined over one or more periods of time after the administration of L-DOPA. The L-DOPA clearance rate may be calculated for each time period to represent the dynamics of L-DOPA clearance rate after the administration of L-DOPA. These L-DOPA clearance rates calculated over time may be time-stamped and stored in the patient profile 118. Certain metrics may be derived from the time-stamped L-DOPA clearance rates, such as mean, median, standard deviation, percentile, etc.

[0157] In certain aspects, symptom levels may be determined from symptom information, for example, from the symptom feedback sensor(s) 206 discussed previously. Symptom levels can correspond to individual measurements from individual symptom feedback sensors of the symptom feedback sensor(s) 206. In addition, or alternatively, the symptom levels can include a calculated composite of symptom feedback information from multiple of the symptom feedback sensor(s) 206. In certain aspects, a symptom trend may be determined based on symptom levels over a certain period of time. In certain aspects, symptom trends may be determined based on L- DOPA level rates of change over certain periods of time.

[0158] In certain aspects, health and sickness metrics may be determined, for example, based on one or more of user input (e.g., pregnancy information or known sickness or disease information), from physiologic sensors (e.g., temperature), activity sensors, or a combination thereof. In certain aspects, based on the values of the health and sickness metrics, for example, a patient’s state may be defined as being one or more of healthy, ill, rested, or exhausted.

[0159] In certain aspects, the meal state metric may indicate the state the patient is in with respect to food consumption. For example, the meal state may indicate whether the patient is in one of a fasting state, pre-meal state, eating state, post-meal response state, or stable state. In certain aspects, the meal state may also indicate nourishment on board, e g., meals, snacks, or beverages consumed, and may be determined, for example, from food consumption information, time of meal information, and/or digestive rate information, which may be correlated to food type, quantity, and/or sequence (e.g., which food/beverage was eaten first).

[0160] In certain aspects, meal habits metrics are based on the content and the timing of a patient’s meals. For example, if a meal habit metric is on a scale of 0 to 1, the better/healthier meals the patient eats the higher the meal habit metric of the patient will be to 1, in an example. Also, the more the patient’s food consumption adheres to a certain time schedule or a recommended diet, the closer their meal habit metric will be to 1, in the example.

[0161] In certain aspects, medication compliance is measured by one or more metrics that are indicative of how committed the patient is towards their medication regimen. In certain aspects, medication compliance metrics are calculated based on one or more of the timing of when the patient takes medication (e.g., whether the patient is on time or on schedule), the type of medication (e.g., is the patient taking the right type of medication), and the dosage of the medication (e.g., is the patient taking the right dosage). In certain aspects, medication compliance of a patient may be determined in a clinical trial where medication consumption and timing of such medication consumption is monitored, through user input, and/or based on L-DOPA data received from the CLM sensor system 104. [0162] In certain aspects, the activity level metric may indicate the patient’s level of activity. In certain aspects, the activity level metric may be determined, for example, based on input from an activity sensor or other physiologic sensors, such as the symptom feedback sensor(s) 206. In certain aspects, the activity level metric may be calculated by the DAM 116 based on one or more of the inputs 128, such as one or more of exercise information, non-L-DOPA sensor data (e.g., accelerometer data), time, user input, etc. In certain aspects, the activity level may be expressed as a step rate of the patient. Activity level metrics may be time-stamped so that they can be correlated with the patient’s lactate levels at the same time.

[0163] FIG. 4 illustrates example operation of the continuous L-DOPA sensor 202 described relative to FIG. 2, in accordance with certain aspects of the present disclosure. In the example of FIG. 4, the continuous L-DOPA sensor 202 detects L-DOPA through an electrochemical and enzymatic method. In general, L-DOPA is oxidized via a catechol oxidase class of enzymes. In certain embodiments, the oxidized product can either undergo a redox reaction with a redox mediator, which in turn generates reductive current at the electrode, or be directly reduced at an electrode surface to generate a current. In various embodiments a resistance layer (RL) can help control or regulate a rate of analyte diffusion and also prevent interferents from reaching a working electrode (WE). Similarly, in various embodiments, an interferent layer (IL) can be a dedicated layer that further prevents specific interferents from reaching the WE.

[0164] FIG. 5 illustrates examples of data sources 550 that can provide, for example, at least a portion of the inputs 128 described relative to FIG. 3, in accordance with certain aspects of the present disclosure. The data sources 550 are shown to include a user interface 546, one or more motor function sensors 548, and a CLM sensor system 504. The CLM sensor system 504 can correspond, for example, to the CLM sensor system 104 described relative to FIGS. 1-3. The user interface 546 and each of the motor function sensor(s) 548 can each correspond, for example, to one of the symptom feedback sensor(s) 206 described relative to FIG. 2.

[0165] As shown in FIG. 5, CLM sensor system 504 can generate and provide L-DOPA measurement information 556. The L-DOPA measurement information 556 can include, for example, information related to L-DOPA levels over time (e.g., an L-DOPA versus time profile), peak concentration in interstitial fluid (Cmax), time to reach Cmax (Tmax), elimination half-life, rate of absorption, rate of clearance, rate of change, a therapeutic window, and/or the like. Examples of the L-DOPA measurement information 556 will be further described relative to FIG.

9.

[0166] The user interface 546 can be provided, for example, on one or more of the display devices 107, 210, 220, 230, and/or 240 discussed previously relative to FIGS. 1-3. As shown in FIG. 5, the user interface 546 can generate and provide user feedback 552. The user feedback 552 can include, for example, food consumption information, treatment information (e.g., medication information), symptom information (e.g., user-input events such as falls or episodes of severe muscle tremors), voice, and/or the like. Examples of the user feedback 552 will be further described relative to FIG. 10.

[0167] In general, the motor function sensor(s) 548 can include, for example, one or more sensors operable to provide motor function information such as information related to muscle tremors or muscle rigidity. The motor function sensor(s) 548 can include, for example, one or more sEMG sensors. In addition, or alternatively, the motor function sensor(s) 548 can include, for example, one or more inertial measurement units such as one or more accelerometers, gyroscopes, magnetometers, and/or the like. In addition, or alternatively, some or all of the motor function sensor(s) 548 can be an example of inertial measurement unit (e.g., accelerometer, gyroscope, and/or magnetometer) that may be provided by, or integrated with, for example, one or more of the display devices 107, 210, 220, 230, and/or 240 discussed previously relative to FIGS. 1-3.

[0168] As mentioned above, the motor function sensor(s) 548 can include, for example, one or more surface electromyography (sEMG) sensors as a source of symptom feedback. Each sEMG sensor can be placed on or over one or more muscles that are at or near a surface of the body, such as biceps or triceps, and can measure electrical activity produced by the muscle over which it is worn. In particular, each sEMG sensor can measure the electrical signals that cause the muscles to contract. In an example, in certain aspects, the biceps (flexor) and triceps (extensor) offer a combination of muscles that can benefit the detection of symptoms such as muscle rigidity.

[0169] In addition, or alternatively, the motor function sensor(s) 548 can include, for example, an inertial measurement unit. The inertial measurement unit can include, for example, an accelerometer, gyroscope (e.g., a rate gyroscope), and/or magnetometer, each of which can benefit measurement of motion pathologies involved in PD. The IMU may be included, for example, in a phone or a wearable such as a smartwatch. An accelerometer, for example, can measure acceleration. In some aspects, the accelerometer estimates angular data by comparison to the vector for acceleration due to gravity. A magnetometer, for example, can measure orientation in the Earth’s magnetic field, subject to forces which distort that field, including local magnets and structural ferromagnetic materials. In certain embodiments, the magnetometer provides measurement information usable to identify changes in device orientation, assuming a static local magnetic field. A rate gyroscope, for example, can indicate a rate of angular change. In certain aspects, combined with an accelerometer, the rate gyroscope can track a path of motion of the sensor. Examples of the motor function sensor(s) 548 will be described in greater detail relative to FIGS. 6B and 6C.

[0170] As shown in FIG. 5, the motor function sensor(s) 548 can generate and provide motor function feedback 554. The motor function feedback 554 can include symptom information such as, for example, information related to muscle tremors, gait, stride, balance, muscle rigidity, respiration, general activity and/or the like. In some aspects, the motor function feedback 554 can include a combination of different feedback provided by multiple distinct sensors (e.g., an sEMG sensor in combination with an accelerometer, gyroscope, and/or magnetometer). Examples of the motor function feedback 554 will be further described relative to FIGS. 11-20.

[0171] In certain implementations, power consumption can be considered and minimized relative the motor function sensor(s) 548. In certain aspects, accelerometers can be very low power, offering continuous measurements at single digit microwatt (pW) operating power levels. In various implementations, gyroscopes can burn on the order of single digit milliwatt (mW) power. In various implementations, magnetometers generally bum more power, in the 10-300 pW range and beyond. Therefore, in various embodiments, power efficiency can be improved by utilizing an IMU having an accelerometer and no gyroscope or magnetometer.

[0172] In further consideration of power consumption, in various aspects, the IMU and/or sEMG sensors can be rechargeable. In addition, or alternatively, some or all of the motor function sensor(s) can be selectively shut down. In an example, if it is detected that the patient is sleeping, some or all of the motor function sensor(s) 548 can be shut down or checked less frequently. In another example, if it is detected and/or predicted that the patient’s L-DOPA levels are optimal (e.g., within a target range), some or all of the motor function sensor(s) 548 can be shut down or checked less frequently. In such a scenario, the sensors can be periodically activated at or before a predetermined time when L-DOPA levels are detected or predicted to fall outside of a therapeutic window. In addition, or alternatively, for analysis in a power constrained environment, one could activate gyroscope measurements on any of the motor function sensor(s) 548 that are likely to show action from muscle measurements. For example, a watch IMU can be activated in response to muscle activity indicated by sEMG sensors on the patient’s biceps and/or triceps.

[0173] FIG. 6A illustrates an example of a sensor configuration 600 relative to a patient 601, in accordance with certain aspects of the present disclosure. In various embodiments, the sensor configuration 600 can be implemented, for example, by the therapy management system 100 of FIG. 1. The sensor configuration 600 includes an sEMG sensor set 625 (that includes, as shown, individual sEMG sensors 654 and 656), the CLM sensor system 504 of FIG. 5, an L-DOPA pump 608, and a smartwatch 640. For illustrative purposes, the sEMG sensor set 625 is shown to include two sEMG sensors, namely, an sEMG sensor 654 and an sEMG sensor 656. Of course, however, the sEMG sensor set 625 may include fewer sensors (such as one sEMG sensor), a greater amount of sensors (e.g., three or more sEMG sensors), etc.

[0174] In certain embodiments, the foregoing devices of the sensor configuration 600 are each operable to wirelessly communicate with a display device 620 which, in turn, is in network communication with cloud services 660. The display device 620 can correspond, for example, to any of the display devices 107, 210, 220, 230, and/or 240 of FIGS. 1 and 2. The cloud services 660 can include, for example, a cloud environment on which all or part of the therapy management engine 114 of FIG. 1 is implemented.

[0175] In the illustrated embodiment, the CLM sensor system 504 is shown to be positioned on a right arm of the patient 601, although one skilled in the art will appreciate that the CLM sensor system 504 can be similarly positioned relative to other locations on the patient 601 (e.g., left arm, left or right thigh, abdomen, etc.) without deviating from the principles described herein. The L- DOPA pump 608 is shown to be positioned on the abdomen of the patient 601. The L-DOPA pump 608 can operate, for example, as described relative to the L-DOPA delivery device 208 of FIG. 2.

[0176] In the illustrated embodiment, the sEMG sensor 656 is placed over a left biceps (flexor) of the patient 601, while an sEMG sensor 654 is placed over a left triceps (extensor) of the patient 601. In certain embodiments, the sEMG sensors 654 and 656 can collect complementary sEMG data, for example, to detect muscle rigidity, as further discussed below. In some embodiments, one or both of the sEMG sensors 654 and 656 can include an inertial measurement unit such as an accelerometer, gyroscope, and/or magnetometer. In certain of these embodiments, the smartwatch 640 can be omitted from the sensor configuration 600. In others of these embodiments, the IMUs of the sEMG sensors 654 and 656 and/or the smartwatch 640 can coexist to provide multiple data points relative to motion of the patient 601.

[0177] In some embodiments, the smartwatch 640 can include, for example, an inertial measurement unit such as an accelerometer, gyroscope, and/or magnetometer, as discussed previously relative to the symptom feedback sensor(s) 206 of FIG. 2 and the motor function sensor(s) 548 of FIG. 5. In this way, in these embodiments, the smartwatch 640 can detect motion caused by signals collected, for example, at the sEMG sensors 654 and 656.

[0178] In various embodiments, other components can be included in the sensor configuration 600. For example, as discussed previously relative to FIG. 1, a DBS component can be included. In certain embodiments, the DBS component produces electrical impulses that affect brain activity to treat PD and/or other medical conditions via, for example, electrodes that have been implanted within certain areas of the brain. In certain embodiments, a DBS waveform (e.g., frequency, repetition rate, magnitude of stimulus, etc.) produced by the DBS component is adjustable via, for example, communication with a user device such as the display device 620 and/or the smartwatch 640.

[0179] FIG. 6B illustrates an example architecture for an sEMG sensor 658 having IMU features, in accordance with certain aspects of the present disclosure. In certain aspects, the sEMG sensor 658 can serve as the sEMG sensor 654 and/or the sEMG sensor 656 of FIG. 6A. The sEMG sensor 658 includes sEMG sensor pads 652a and 652b, a reference electrode 776, an instrumentation amplifier 665, a band-pass filter (BPF) 668, an analog-to-digital converter (ADC) 670, a processor 672, and a wireless interface 674. For illustrative purposes, the sEMG sensor 658 is shown to include an accelerometer 662, a magnetometer 664, and a rate gyroscope 666.

[0180] According to the example circuitry shown in FIG. 6B, signals from the sEMG sensor pads 652a and 652b are amplified in instrumentation amplifier 665 to output an sEMG sensor signal. The sEMG sensor signal then passes through the BPF 668 and the ADC 670 before being fed to the processor 672. The reference electrode 776 provides a local ground for the sEMG sensor pads 652a and 652b to mitigate, for example, common mode issues. In some embodiments, the BPF 668, for example, can isolate and/or filter the sEMG sensor signal to remove signal portions not attributable to muscle activity (e.g., a 60 Hz background signal and/or other signals interfering with sEMG signals). In similar fashion, outputs of the accelerometer 662, the magnetometer 664, and/or the rate gyroscope 666 are routed to the processor 672. In this way, in various embodiments, the processor 672 can aggregate data from the sEMG sensor pads 652a and 652b, the accelerometer 662, the magnetometer 664, and/or the rate gyroscope 666 for processing.

[0181] In some embodiments, the processor 672 can correspond to, or reside on, for example, a smartphone (e.g., the display device 620 of FIG. 6A), the smartwatch 640, a cloud environment (e.g., the cloud services 660 of FIG. 6A), a combination of local (e.g., edge) processing and phone and/or cloud processing, and/or the like. In addition, or alternatively, as shown in FIG. 6B, the processor 672 can correspond to, or reside on, the sEMG sensor 658. In some of these embodiments, the processor 672 can serve as a main processor in a sensor configuration such as the sensor configuration 600 of FIG. 6A.

[0182] In some embodiments, as shown, the wireless interface 674 can be implemented on the sEMG sensor 658. In addition, or alternatively, the wireless interface 674 can be implemented on the display device 620 of FIG. 6A and/or the smartwatch 640 of FIG. 6A. The wireless interface 674 can provide, for example, an ability to communicate via Wi-Fi, cellular, NFC, Bluetooth and/or the like, as discussed relative to FIGS. 1 and 2. In similar fashion, as shown, the BPF 668 and the ADC 670 can be implemented on the sEMG sensor 658.

[0183] FIG. 6C illustrates an example communications framework 650 based on the sensor configuration 600 of FIG. 6A, in accordance with certain aspects of the present disclosure. The communications framework 650 includes the display device 620, the smartwatch 640, the sEMG sensor set 625, the CLM sensor system 504, and the L-DOPA pump 608.

[0184] The display device 620, which can be a smartphone, is shown to have an application 606 resident and executing thereon. The application 606 can generally operate, for example, as described relative to the application 106 of FIG. 1. In various aspects, the application 606 can be used to supply user input regarding, for example, symptom feedback. For example, the user input can indicate the presence or absence of symptoms, via oral feedback (e.g., via a microphone on the user device), textual feedback (e.g., via touchscreen, keyboard or the like), gestures (e.g., detected via a camera), and/or via other indications in a suitable interface of the user device. [0185] In the example of FIG. 6C, the display device 620 is operable to communicate with the sEMG sensors of the sEMG sensor set 625, the CLM sensor system 504 and the L-DOPA pump 608 via Bluetooth, and with the cloud services 660 via Wi-Fi. Further, according to the example of FIG. 6C, the display device 620 is operable to communicate with the smartwatch 640 via Bluetooth and/or Wi-Fi. Although FIG. 6C illustrates certain types of connectivity between devices, it should be appreciated that, in various embodiments, any suitable form of wired or wireless communication may be utilized to suit a given implementation (e.g., loT connectivity and/or NFC as discussed relative to FIGS. 1 and 2). Further, it should be noted that, in certain embodiments, any of the devices shown in FIG. 6C can be directly coupled (e.g., wired) to each other and/or integrated into a single device or wearable.

[0186] For example, in some embodiments, communication between the devices shown in FIG. 6C can be accomplished, at least in part, via an on-body or body area network that bridges to an off-body component. The body area network can include, for example, the sEMG sensor set 625, the CLM sensor 505, the L-DOPA pump 608, the smartwatch 640 and/or other components. In certain embodiments, the body-area network can minimize signals used in communication and minimize interference with or from other devices, for example, in household or medical settings. For example, a single node in the body area network can serve as a bridge (e.g., a Bluetooth Low Energy (BLE) bridge) to one or more off-body components, such as the display device 620, thereby minimizing the number of devices claiming airtime, for example, in the crowded 2.4 GHz band. Advantageously, in certain aspects, the body-area network reduces power and antennae requirements for individual sensors and other devices (e.g., the devices shown in FIG. 6C), thereby enabling such sensors and devices to be smaller.

[0187] FIG. 7 illustrates an example of a process 700 for establishing a sensor configuration such as the sensor configuration 600 of FIG. 6A, in accordance with certain aspects of the present disclosure. In some aspects, the process 700 can be executed, for example, by the therapy management engine 114 of FIG. 1. In addition, or alternatively, the process 700 can be executed, for example, by the CLM sensor system 104. In addition, or alternatively, the process 700 can be executed generally by any of the display devices 107, 210, 220, 230, and/or 240. In addition, or alternatively, the process 700 can be executed, for example, by the CLM sensor system 504, the display device 620, any of the sEMG sensors 654, 656 and 658, the smartwatch 640, and/or the cloud services 660 of FIG. 6A. In addition, or alternatively, the process 700 can be executed, for example, by the processor 672 of FIG. 6B. In addition, or alternatively, the process 700 can be executed, for example, by the application 606 of FIG. 6C. Although any number of systems, in whole or in part, can implement the process 700, to simplify discussion, the process 700 will be described in relation to the therapy management engine 114 of FIG. 1 and the data sources 550 of FIG. 5.

[0188] At block 702, the therapy management engine 114 receives patient information such as, for example, physiological and/or demographic characteristics like age, gender, weight, height and body mass index, lifestyle characteristics like diet, sleep patterns and exercise, and/or other information. In some embodiments, the patient information can be received, for example, via the user interface 546 discussed relative to FIG. 5.

[0189] At block 704, the motor function sensor(s) 548 are placed on the patient, for example, by the patient, a caregiver, a clinician and/or the like. In various implementations, the motor function sensor(s) 548 can be placed on the patient as discussed above relative to FIG. 6A. For example, as part of the block 704, the sEMG sensors 654 and 656 may be placed over the biceps and triceps, respectively, of the patient 601.

[0190] In another example, as part of the block 704, an IMU (e.g., an accelerometer, magnetometer and/or gyroscope) may be placed on the patient. In certain aspects, angular motions are generally more pronounced near extremities. According to these aspects, in some embodiments, an IMU containing an accelerometer can be positioned near extremities. For example, a smartwatch that includes an IMU, such as the smartwatch 640, can be worn by the patient (e.g., the patient 601), thereby offering inertial sensing at the end of the forearm. In some aspects, an IMU can be co-located with an sEMG sensor in order to leverage the fact that tremors are typically strongest near the tremulous muscle.

[0191] At block 706, the therapy management engine 114 calibrates the motor function sensor(s) 548. In some aspects, the calibration can occur individually for each sensor, for example, of the sensor configuration 600 of FIG. 6A. In addition, or alternatively, the calibration can occur collectively, for example, for the sensor configuration 600. [0192] In some aspects, the block 706 can include the therapy management engine 114 generating a baseline signature that represents a background signal not representative of muscle activity (e.g., an approximately 60Hz background signal that is commonly present). In various embodiments, the baseline signature can be used during monitoring to fdter the signal to remove a signal portion attributable to background noise, and to isolate a signal portion attributable to muscle activity.

[0193] In some aspects, the block 706 can include the therapy management engine 114 instructing the patient, for example, via the user interface 546, to perform a set of calibration activities. For example, the patient can be asked to perform daily activities such as sitting, lying in bed, lying on the ground, climbing up and down stairs, running, walking, etc. For each activity, the therapy management engine 114 can determine symptom feedback information produced by the motor function sensor(s) 548 (e.g., measurement information based on a type of sensor). Thereafter, for each activity, the therapy management engine 114 can generate a motion signature for each activity based on measured information from the symptom feedback sensors following the patient performing the activity. The motion signatures and/or baseline signatures can be individualized for a sensor and/or can be based on a combination of measurement information from a combination of sensors (e.g., a plurality of sEMG sensors, an accelerometer, a magnetometer, a gyroscope, and/or the like).

[0194] After block 706, the process 700 ends. In various aspects, the process 700 can be repeated for individual sensors such as, for example, individual sensors of the motor function sensor(s) 548). In addition, or alternatively, the process 700 can be repeated at various intervals to recalibrate responsive to a change in sensor configuration, change in sensor placement, disease progression, change in treatment, on-demand requests from the patient, and/or the like.

[0195] Table 5 below illustrates example motion signatures that are each defined as a combination of signals, for example, from the sEMG sensors 654 and 656 and the smartwatch 640 of FIG. 6A. In the example of Table 5, the sEMG sensors 654 and 656 each include an IMU, specifically, an accelerometer, gyroscope, and magnetometer. The example motion signatures include signatures for non-symptomatic motions such as lifting as well as for symptomatic motions such as stiffening, muscle tremors, muscle rigidity, and cog wheeling. It should be appreciated that each motion signature can include fewer or different signals, for example, depending on which sensors and signals are available in a given implementation. In various embodiments, Table 5 below assumes that the patient is stationary. In certain aspects, if the patient is moving (e.g., walking), motion from walking can be fdtered out. In certain embodiments, PD tremor bands are higher in frequency than most motions and can be isolated with signal processing techniques.

Table 5 [0196] FIG. 8 illustrates an example of a process 800 for monitoring and tracking PD symptoms, for example, using the sensor configuration 600 of FIG. 6A, in accordance with certain aspects of the present disclosure. In some aspects, the process 800 can be executed, for example, by the therapy management engine 114 of FIG. 1. In addition, or alternatively, the process 800 can be executed, for example, by the CLM sensor system 104. In addition, or alternatively, the process 800 can be executed generally by any of the display devices 107, 210, 220, 230, and/or 240. In addition, or alternatively, the process 800 can be executed, for example, by the CLM sensor system 504, the display device 620, any of the sEMG sensors 654 and 656, the smartwatch 640, and/or the cloud services 660 of FIG. 6A. In addition, or alternatively, the process 800 can be executed, for example, by the processor 672 of FIG. 6B. In addition, or alternatively, the process 800 can be executed, for example, by the application 606 of FIG. 6C. Although any number of systems, in whole or in part, can implement the process 800, to simplify discussion, the process 800 will be described in relation to the therapy management engine 114 of FIG. 1 and the data sources 550 of FIG. 5.

[0197] At block 802, the therapy management engine 114 receives inputs such as, for example, L-DOPA measurement information from the CLM sensor system 504 and symptom information for the patient from the motor function sensor(s) 548.

[0198] At block 804, the therapy management engine 114 analyzes the symptom information to detect a PD symptom or event, for example, based on motion signatures such as the motion signatures discussed above. The block 804 can include, for example, the therapy management engine 114 comparing the symptom information to the motion signatures. Accordingly, in certain aspects, the detection of the PD symptom or event can be based on the symptom information matching the corresponding motion signature.

[0199] At block 806, the therapy management engine 114 prompts the patient to validate the detected PD symptom or event (e.g., “are you experiencing tremors?” or “did you fall”?). In some embodiments, the prompt can further ask the patient to indicate a severity of the symptom or event (e.g., on a scale of 1 to 5 or 1 to 10, with higher numbers indicating greater severity). In various embodiments, the prompt can presented to the patient via the user interface 546. [0200] At block 808, the therapy management engine 114 receives user input confirming or denying the symptom or event and, if applicable, indicating severity. The user input can be received, for example, via the user interface 546.

[0201] At block 810, the therapy management engine 114 tags the L-DOPA measurement information based on a severity of the symptom or event. For example, the L-DOPA measurement information can be tagged based on severity (e.g., moderate or dangerous, optionally based in part on a user-indicated severity), based on a type of event (e.g., “fall” or “dyskinesia”), and/or as out- of-range.

[0202] At block 812, the therapy management engine 114 adjusts or recommends a treatment (e.g., L-DOPA dosage). In an example, if the analysis indicates that undesired side-effects are resulting from a current L-DOPA dose that is too high for the patient, a future L-DOPA dose may be reduced. In another example, if the analysis indicates that a current L-DOPA dose is too low for the user and is not controlling the user’s PD symptoms, a future L-DOPA dose may be increased. In some aspects, the block 812 can include generating and presenting actionable treatment data, as will be discussed in greater detail relative to FIG. 23. In addition, or alternatively, the block 812 can include automatically determining a treatment, as will be further discussed relative to FIG. 26.

[0203] At block 814, the therapy management engine 114 updates models and/or executes learning (e.g., via rule-based models and/or ML-based models) such that, as similar signatures are seen in the future, an alert can be issued in advance of the symptoms or event. In an example, the therapy management engine 114 can execute a retrospective analysis of signals from the motor function sensor(s) 548 during a lookback period prior to a start of an event or “severe” symptoms severe (e.g., 30 minutes prior). According to this example, the therapy management engine 114 can generate one or more predictive signatures based on the signals from the motor function sensor(s) 548 during the lookback period. In various embodiments, the predictive signatures can be structured as described relative to the motion signatures discussed above. After block 814, the process 800 ends.

[0204] In various embodiments, the predictive signatures discussed above can categorized, for example, into a one or more zones, such as a safe or “green” zone, a moderate zone, and a dangerous zone. The safe or “green” zone may indicate, for example, that L-DOPA levels are deemed optimal. The moderate zone may indicate that some danger may be present, such as dyskinesia. For example, the moderate zone may be appropriate for a situation in which levels of L-DOPA are above a threshold associated with dyskinesia but are rapidly falling. The dangerous zone may indicate, for example, that L-DOPA levels are too low and are not rising, such that a fall can occur.

[0205] In various embodiments, the therapy management engine 114 can anticipate and alert regarding a risk of severe PD symptoms or events, for example, based on the predictive signatures discussed above. For example, the therapy management engine 114 can receive inputs such as, for example, L-DOPA measurement information from the CLM sensor system 504 and symptom information for the patient from the motor function sensor(s) 548, in similar fashion to block 802 discussed above. Thereafter, the therapy management engine 114 can analyze the symptom information to detect a predicted PD symptom or event, for example, based on the predictive signatures, in similar fashion to the motion signature detection discussed relative to block 804 above. If the symptom information matches any of the predictive signatures, an alert regarding symptoms or events associated with the matching predictive signature can be presented to the patient (e.g., prompt the patient to sit down if the matching predictive signature is associated with a fall).

[0206] In various embodiments, the therapy management engine 114 can track disease and/or treatment progression over time. In certain aspects, the therapy management engine 114 can stage the disease by determining, based on the motor function sensor(s) 548 and the motion signatures discussed above, when symptoms appear and what types of symptoms they are relative to the concentration of L-DOPA interstitially measured, for example, by the CLM sensor system 504. For example, the motor function sensor(s) 548 (e.g., sEMG signals in combination with accelerometer signals) can be used to gauge and categorize physical signatures including muscle tremors, muscle rigidity and daily activity to gauge progression of disease and lifestyle, all which can inform L-DOPA dosing, in certain embodiments.

[0207] Table 6 below illustrates an example use case, for example, according to the process 800 of FIG. 8. More particularly, Table 6 illustrates an example of a signature corresponding to a fall. Table 6 further illustrates user input to validate the occurrence of the fall. In certain embodiments, the example of Table 6 corresponds to a dangerous zone, as discussed above.

Table 6

[0208] Table 7 below illustrates another example use case, for example, according to the process 800 of FIG. 8. More particularly, Table 7 illustrates an example of a signature corresponding to dyskinesia. Table 7 further illustrates user input to validate that the patient is experiencing dyskinesia. In certain embodiments, the example of Table 7 corresponds to a moderate zone, as discussed above.

Table 7

[0209] Table 8 below illustrates another example use case, for example, according to the process 800 of FIG. 8. More particularly, Table 8 illustrates an example of a signature corresponding to appropriate symptom control. Table 8 further illustrates user input to validate that symptom control. In certain embodiments, the example of Table 8 corresponds to a safe or “green” zone, as discussed above.

Table 8

[0210] FIG. 9 illustrates an example of L-DOPA measurement information 956 that can be generated, for example, by a CLM sensor system such as the CLM sensor system 104 of FIGS. 1- 3 and/or the CLM sensor system 504 of FIG. 5, in accordance with certain aspects of the present disclosure. As shown, the L-DOPA measurement information 956 can include profiles of L-DOPA levels versus time.

[0211] FIG. 10 illustrates an example of user feedback 1052 that can be provided, for example, via the user interface 546 of FIG. 5, in accordance with certain aspects of the present disclosure. The user feedback 1052 can include, for example, information related to demographics (e.g., age, gender, etc.), activity, treatment (e.g., drug dose, timing, and/or formulation, a Unified Parkinson’s Disease Rating Scale (UPDRS) score, etc.), food consumption, user-assigned tasks, and/or the like. In some aspects, at least a portion of the user feedback 1052 can be provided by the patient, caretaker, or other user in response to prompts such as, “Are you experiencing tremors?,” “Did you fall?” or “Did you eat dinner?”

[0212] In certain embodiments, user feedback similar to the user feedback 1052 can be captured at various intervals during use, such as upon onboarding and/or when an event (e.g., a patient fall or freeze) is predicted or detected. In certain embodiments, this feedback can be used to match with patient population data (e.g., other CLM users). For example, certain types of specific user feedback, such as demographics, activity, and/or known disease state, can be used to perform an initial mapping of a first-time user onto a patient disease-progression worldview. In certain embodiments, this user feedback can help, for example, the therapy management engine 114 generate an initial set of predictions for dose and undesirable physiological responses, which predictions can be fine-tuned as more user data becomes available.

[0213] FIG. 11 illustrates examples of motor function feedback 1154 that can be provided, for example, by the accelerometer 662 of FIG. 6, in accordance with certain aspects of the present disclosure. For example, the motor function feedback 1154 can include data related to magnitude (e.g., tremor magnitude), frequency, and/or direction. In some aspects, the motor function feedback 1154 can represent movement as frequencies. In addition, or alternatively, magnitude and direction vectors can be tracked over time to detect activity. Activity as movement may be seen as changes in acceleration superimposed on the acceleration due to gravity. The motions involved in walking and running, for example, show strong cyclic components correlated to each step (e.g., step rate), and each pair of steps (i.e., full cycle of movement of both legs).

[0214] For example, in certain aspects, the motor function feedback 1154 can include data in the time versus frequency domain. In these aspects, different activities or events can be associated with different frequency signatures (e.g., signatures associated with walking, sleeping, falls, etc.). According to these aspects, viewing the data in the time versus frequency domain enables the therapy management engine 114 to detect activities or events, for example, by matching the motor function feedback 1154 to one or more of the frequency signatures.

[0215] FIG. 12 illustrates an example of motor function feedback 1254 that can be provided, for example, by the sEMG sensors 654 and 656 of FIG. 6A, in accordance with certain aspects of the present disclosure. In particular, the motor function feedback 1254 illustrates differences in an sEMG signal for different activities or states such as, for example, pushing a button, resting, performing curls (e.g., with 10-pound weights), and air boxing. In the example of FIG. 12, the motor function feedback is captured at 12.5 kilo-samples per second (kSPS), the analog bandwidth is under 200 Hz, and a 60Hz notch filter is employed to minimize interference.

[0216] According to the motor function feedback 1254, the sEMG signal is approximately zero when the patient is at rest, while pushing a button registers a slight pulse. Certain activities, such as curls and air boxing, result in greater amplitudes in the sEMG signal. For example, in the motor function feedback 1254, faster motions (e.g., two curls with a 10-pound weight in rapid succession or air boxing) result in greater amplitudes in the sEMG signal than singular or slower motions (e.g., a single curl).

[0217] FIGS. 13A-B illustrate an example of motor function feedback that can be provided, for example, by the accelerometer 662 of FIG. 6B, in accordance with certain aspects of the present disclosure. Specifically, FIG. 13A illustrates an example of time-based accelerometer magnitude data 1354A indicating a sleeping patient. FIG. 13B illustrates a time-based spectrogram 1354B showing which frequencies have energy for the sleeping patient.

[0218] More particularly, FIGS. 13A-B collectively illustrate an example of identifying sleep activity using an accelerometer. The time-based accelerometer magnitude data 1354A shown in FIG. 13 A reveals an orientation, for example, of the accelerometer 662, as the sleeping patient moves through different sleep positions, typically abruptly. The time-based spectrogram 1354B shown in FIG. 13B reveals more subtle movements and also shows transitions from one position to another.

[0219] FIGS. 14A-B illustrate another example of motor function feedback that can be provided, for example, by the accelerometer 662 of FIG. 6B, in accordance with certain aspects of the present disclosure. Specifically, FIG. 14A illustrates an example of time-based accelerometer magnitude data 1454A indicating various activities of a patient. FIG. 14B illustrates a time-based spectrogram 1454B showing which frequencies have energy for the patient. More particularly, FIGS. 14A-B collectively illustrate the utilization of data from the accelerometer 662, for example, to identify physical activity. In certain aspects, detection of muscle tremors follows naturally from the detection of frequencies of interest, as shown in FIGS. 14A-B. [0220] FIG. 15A illustrates an example of motor function feedback 1554A that can be provided, for example, by the accelerometer 662 of FIG. 6B, in accordance with certain aspects of the present disclosure. In certain aspects, the motor function feedback can include accelerometer feedback, such as directional data (e.g., an orientation vector) as a function of time. In these aspects, different events can be associated with different time-based directional signatures. According to these aspects, the therapy management engine 114, for example, can detect movement or events (e.g., falls and/or gait) by matching the accelerometer feedback to the timebased directional signatures. In certain aspects, averaging the g vector during sleep can yield a direction indicative of recumbent position, while averaging the g vector during activity can yield a direction indicative of standing or walking erect. In various aspects, this information can be used to confirm falls.

[0221] FIG. 15B illustrates an example of motor function feedback 1554B that can be provided, for example, by the accelerometer 662 of FIG. 6B, in accordance with certain aspects of the present disclosure. The motor function feedback 1554B can indicate, for example, a patient getting up at a fast rate, the patient lying down, the patient falling down at a fast rate, the patient getting up slowly (e.g., after having fallen), and/or the like.

[0222] FIG. 16 illustrates an example of correlating user feedback, accelerometer feedback, and L-DOPA sensor feedback, for example, to generate or recommend L DOPA dosing and/or a personalized therapeutic window (e g., a target range of L-DOPA levels deemed effective for the patient), in accordance with certain aspects of the present disclosure. In the example of FIG. 16, different ranges of L-DOPA levels can be identified as effective or ineffective, for example, based on muscle tremor magnitude.

[0223] FIG. 17 illustrates an example of correlating motor function feedback 1754 from an sEMG sensor and/or an accelerometer with L-DOPA sensor feedback 1756, in accordance with certain aspects of the present disclosure. In the illustrated embodiment, the motor function feedback 1754 includes a tremor-filtered signal 1770 and a dyskinesia-filtered signal 1772, while the L-DOPA sensor feedback 1756 includes L-DOPA levels 1774. According to the example of FIG. 17, dyskinesia is directly proportional to the L-DOPA levels 1774 (e.g., dyskinesia increases with L-DOPA levels). Similarly, muscle tremors are inversely proportional to the L-DOPA levels 1774 (e.g., muscle tremors increase with decreasing L-DOPA levels). In the example of FIG. 17, different ranges of the L-DOPA levels 1774 can be identified as effective or ineffective, for example, based on dyskinesia and/or muscle tremor magnitude indicated by the dyskinesia-filtered signal 1772 and the tremor-filtered signal 1770, respectively.

[0224] FIG. 18 illustrates examples of tracking an effectiveness of L-DOPA treatment, for example, with reference to a personalized therapeutic window, in accordance with certain aspects of the present disclosure. In various embodiments, the personalized therapeutic window can be tracked, for example, to make inferences about disease progression.

[0225] FIG. 19 illustrates examples of tracking L-DOPA treatment in relation to tremor values (e.g., tremor magnitude), in accordance with certain aspects of the present disclosure. In particular, FIG. 19 shows examples for both oral and pump-based administration of L-DOPA.

[0226] FIG. 20 illustrates examples of tracking an effectiveness of L-DOPA treatment in relation to meals (e.g., protein intake), meal timing (e.g., time before L-DOPA administration), and activities, in accordance with certain aspects of the present disclosure. The tracking can be performed, for example, by the therapy management engine 114. In various aspects, the examples of FIG. 20 leverage that: (A) protein intake directly at the time of drug administration directly affects therapeutic efficacy and therefore symptom relief; (B) the time meals are taken before drug administration directly affect therapeutic efficacy; (C) L-DOPA therapy and exercise can work synergistically to benefit a patient.

[0227] FIG. 21 illustrates an example of a process 2100 for determining a personalized L- DOPA range and administering an L-DOPA dose, in accordance with certain aspects of the present disclosure. For illustrative purposes, the process 2100 is described relative to the user interface 546, the motor function sensor(s) 548, the CLM sensor system 504 and the L-DOPA pump 608, as described relative to FIGS. 5-8. In general, the user interface 546, the motor function sensor(s) 548, the CLM sensor system 504, and the L-DOPA pump 608 can operate within the therapy management system 100 described relative to FIGS. 1-3. For simplicity, the process 2100 will be described as being performed by the therapy management engine 114 of FIG. 1.

[0228] At block 2102, the therapy management engine 114, based on L-DOPA measurement information from the CLM sensor system 504, estimates a current L-DOPA level. At block 2104, the therapy management engine 114 causes the estimated current L-DOPA level to be displayed via the user interface 546.

[0229] At block 2106, the therapy management engine 114, based on feedback from the motor function sensor(s) 548, estimates tremor amplitude and a rate of change in tremor amplitude. At decision block 2108, the therapy management engine 114 determines whether the tremor amplitude exceeds a predetermined threshold. If the therapy management engine 114 determines, at the decision block 2108, that the tremor amplitude is not excess of the predetermined threshold, at block 2110, the therapy management engine 114 causes the user interface 546 to prompt a user to confirm that current L-DOPA levels are desired. If the current L-DOPA levels are desired, at block 2112, the therapy management engine 114 flags the current L-DOPA levels as desired. The flagged L-DOPA levels can correspond, for example, to a personalized L-DOPA range (e.g., as all or part of a personalized therapeutic window)

[0230] Returning to the decision block 2108, if the therapy management engine 114 determines that the tremor amplitude exceeds the predetermined threshold, at block 2114, the therapy management engine 114 recommends an L-DOPA dose (e.g., a predetermined incremental amount to be administered as a bolus). At block 2116, the therapy management engine 114 causes the L-DOPA pump 608 to administer the recommended L-DOPA dose to the patient. After block 2116, the process 2100 ends.

[0231] FIG. 22 illustrates an example of a scenario 2200 for a patient experiencing a fall, in accordance with certain aspects of the present disclosure. For illustrative purposes, the scenario 2200 is shown relative to the CLM sensor system 504, the user interface 546, the accelerometer 662, the sEMG sensor set 625, the patient 601, and the user interface 546, as described above relative to FIGS. 6A-C, 7-12, 13A-B, 14A-B, 15A-B, and 17-20.

[0232] FIG. 23 illustrates an example of a process 2300 for continuous L-DOPA monitoring and therapy management, in accordance with certain aspects of the present disclosure. In some aspects, the process 2300 can be executed, for example, by the therapy management engine 114 of FIG. 1. In addition, or alternatively, the process 2300 can be executed, for example, by the CLM sensor system 104. In addition, or alternatively, the process 2300 can be executed, for example, by the application 106 of FIGS. 1 and 3. In addition, or alternatively, the process 2300 can be executed generally by any of the display devices 107, 210, 220, 230, and/or 240. Although any number of systems, in whole or in part, can implement the process 2300, to simplify discussion, the process 2300 will be described in relation to the therapy management engine 114 of FIG. 1.

[0233] At block 2302, the therapy management engine 114 receives L-DOPA measurement information and symptom information for the patient. The L-DOPA measurement information and the symptom information can generated by the CLM sensor system 104 and the symptom feedback sensor(s) 206, respectively, as discussed relative to FIGS. 1-4.

[0234] At block 2304, the therapy management engine 114 receives L-DOPA dosing information for the patient. The L-DOPA dosing information can indicate, for example, a dose and a time of administration. In various aspects, the L-DOPA dosing information can be received from the L-DOPA delivery device 208 of FIG. 2 (e.g., a continuous L-DOPA pump). In addition, or alternatively, in various aspects, the L-DOPA dosing information can be received, for example, from the display devices 107, 210, 220, 230, and/or 240 discussed previously. In certain aspects in which L-DOPA is not administered via an L-DOPA delivery device such as the L-DOPA delivery device 208, the L-DOPA dosing information may result from user entry via, for example, the display devices 107, 210, 220, 230, and/or 240.

[0235] At block 2306, the therapy management engine 114 correlates the L-DOPA measurement information, the symptom information, and/or the L-DOPA dosing information based on time. An example of the time-based correlation will be described relative to FIGS. 24A-B.

[0236] At block 2308, the therapy management engine 114 characterizes the L-DOPA measurement information based on the symptom information. In certain aspects, the therapy management system can attribute “POOR” motor symptom control to relatively low L-DOPA levels or, conversely, to motor complications caused by high L-DOPA levels. For example, “POOR” motor symptom control may be attributed to motor complications based on data relationships or trends in the correlated data (e.g., motor symptom control is decreasing as L- DOPA levels are increasing, or motor symptom control is relatively low while L-DOPA levels are at or near a peak).

[0237] In certain aspects, the therapy management engine 114 can define degrees, or categories, of symptom control in a personalized or patient-centric way based on the symptom information. For example, the therapy management engine 114 can identify time intervals during which the patient experienced “low” symptoms (e.g., “low” tremor values, no falls and/or nearbaseline speech), where “low” is defined relative to the symptom information of the patient using any suitable statistical or other methodology. According to this example, the therapy management engine 114 can characterize the L-DOPA measurement information for such time intervals as providing “ACCEPTABLE” motor symptom control, such that the patient’s motor symptoms are deemed to be well controlled during those intervals.

[0238] In another example of personalized characterization of L-DOPA measurement information, the therapy management system can identify time intervals during which the patient experienced “high” physical symptoms (e.g., “high” motor function symptoms as indicate by “high” tremor values, at least one fall, speech sufficiently matching a signature associated with physical symptoms, and/or speech sufficiently deviating from a baseline), where “high” is defined relative to the symptom information of the patient using any suitable statistical or other methodology. According to this example, the therapy management engine 114 can characterize the L-DOPA measurement information for such time intervals as providing “POOR” motor symptom control, such that the patient’s motor symptoms are deemed to be not well controlled during those intervals.

[0239] In another example of personalized characterization of L-DOPA measurement information, the therapy management system can identify time intervals during which the patient experienced no severe symptoms (e.g., no severe motor function symptoms, such as no falls and/or no “high” motor function symptoms in the fashion described above). According to this example, the therapy management engine 114 can characterize the L-DOPA measurement information for such time intervals as providing “ACCEPTABLE” motor symptom control, such that the patient’s motor symptoms are deemed to be well controlled during those intervals.

[0240] In certain aspects, the therapy management engine 114 can define degrees, or categories, of motor symptom control in a non-patient-centric way using one or more standard definitions. For example, the therapy management engine 114 can identify time intervals during which the patient experienced “low” symptoms (e g., “low” motor function symptoms as indicated by “low” tremor values, no falls and/or near-baseline speech), where “low” is defined relative to any suitable standard definition of controlled motor symptoms. According to this example, the therapy management engine 114 can characterize the L-DOPA measurement information for such time intervals as providing “ACCEPTABLE” motor symptom control, such that the patient’s motor symptoms are deemed to be well controlled during those intervals.

[0241] In another example of standardized characterization of L-DOPA measurement information, the therapy management engine 114 can identify time intervals during which the patient experienced “high” symptoms (e.g., “high” motor function symptoms as indicated by “high” tremor values, at least one fall, speech sufficiently matching a signature associated with physical symptoms, and/or speech sufficiently deviating from a baseline), where “high” is defined relative to any suitable standard definition of uncontrolled motor symptoms. According to this example, the therapy management engine 114 can characterize the L-DOPA information for such time intervals as providing “POOR” motor symptom control, such that patient’s motor symptoms are deemed to be not well controlled during those intervals.

[0242] In another example of standardized characterization of L-DOPA measurement information, the therapy management engine 114 can identify time intervals during which the patient experienced no severe physical symptoms (e.g., no falls and/or no “high” motor function symptoms in the fashion described above). According to this example, the therapy management engine 114 can characterize the L-DOPA information for such time intervals as providing “ACCEPTABLE” motor symptom control, such that the patient’s motor symptoms are deemed to be well controlled during those intervals.

[0243] At block 2310, the therapy management engine 114 defines a personalized therapeutic window for the patient based on the characterizations of the L-DOPA measurement information. For example, the therapy management engine 114 can identify, as the therapeutic window, a range of L-DOPA levels that are deemed to correspond to “ACCEPTABLE” motor symptom control as described above. In some aspects, the therapy management system can define multiple windows corresponding to multiple ranges, where some windows relate to tighter control of L-DOPA levels than others. For example, one window could correspond to “ACCEPTABLE” motor symptom control as described above, while another, narrower window could correspond to “OPTIMAL” motor symptom control as a narrower subset of the window for “ACCEPTABLE” motor symptom control.

[0244] At block 2312, the therapy management engine 114 generates and presents actionable treatment data to a patient, caregiver, clinician, or other user based on the characterizations of the L-DOPA measurement information, the one or more therapeutic windows, and/or other available information. For example, the therapy management engine 114 can generate and present any of the metrics 130 described above relative to FIG. 3.

[0245] In an example, as part of generating and presenting actionable treatment data at block 2312, the therapy management engine 114 can monitor whether L-DOPA levels are in range relative to the therapeutic window (e.g., corresponding to one or more of the target ranges referenced above) and appropriately update or alert the patient, clinician, caregiver or other user when the L-DOPA levels deviate from the therapeutic window. In some aspects, the therapy management system can monitor time in range and report the time in range to the patient, clinician, caregiver or other user.

[0246] In another example, as part of generating and presenting actionable treatment data at block 2312, the therapy management engine 114 can automatically determine L-DOPA dosing information for the patient based on the characterizations of the L-DOPA measurement information, the personalized therapeutic window, and/or other available information. In certain aspects utilizing a continuous L-DOPA pump as discussed above, the therapy management engine 114 can determine the dosing information in the form of an adjustment to a basal rate or as a bolus. In certain aspects not utilizing a continuous L-DOPA pump, the therapy management engine 114 can determine the dosing information for patient administration (e.g., oral administration). An example of automatically determining L-DOPA dosing information will be described relative to FIG. 26.

[0247] In another example, as part of generating and presenting actionable treatment data at block 2312, the therapy management engine 114 can generate and present predictions. For example, the therapy management engine 114 can predict effects of medication dosage adjustment and/or lifestyle changes. In another example, as part of generating and presenting actionable treatment data at block 2312, the therapy management engine 114 can predict when the patient will be inside or outside a therapeutic window (e.g., a “GOOD” time range and a “BAD” time range, respectively).

[0248] In another example, as part of generating and presenting actionable treatment data at block 2312, the therapy management engine 114 can, periodically or through continuous data interpolation, determine a stage of disease, for example, by looking changes in a therapeutic window over time (e ., changes to a low end, a high end, and/or length of the window). In certain aspects, the therapy management engine 114 can identify a narrowing of the therapeutic window over the time as an indicator of disease progression. Conversely, the therapy management engine 114 can identify a widening of the therapeutic window over time as an indicator of disease regression. In various aspects, a current therapeutic window can be compared to historical therapeutic windows for the patient, or to therapeutic windows for a mean population set similar to that patient. Additionally, data, for example, from the other sensor(s) 209 and/or the symptom feedback sensor(s) 206, can be included to help rule out confounding situations that may negatively affect the accuracy of such an interpretation (e.g., temporal changes to diet, sickness, etc.).

[0249] In another example, as part of generating and presenting actionable treatment data at block 2312, the therapy management engine 114 can motivate the patient to take their medication and improve compliance via audio or visual display. The audio or visual display can motivate the patient, for example, by audibly and/or visually presenting an improvement in metrics over time.

[0250] In another example, as part of generating and presenting actionable treatment data at block 2312, the therapy management engine 114 can determine if symptom modifying therapy is effective at the current dosage and delivery regime prescribed. The concentration of medication therapy measured over time can be used in accordance with symptom presentation and severity over time to determine when and/or to what extent the medication regimen prescribed is effective or not effective enough.

[0251] In another example, as part of generating and presenting actionable treatment data at block 2312, the therapy management engine 114 can determine if symptom modifying therapy is effective at the current dosage and delivery regime prescribed. The concentration of medication therapy measured over time can be used in accordance with symptom presentation and severity over time to determine when and/or to what extent the medication regimen prescribed is effective or not effective enough. For example, if Patient A is well controlled then no therapy modification would be recommended. In another example, if Patient B is not well controlled and the medication dosage given is low, but concentration remains within an expected therapeutic window, then Patient B could be recommended to evaluate a higher dose being added.

[0252] By of further example, if Patient C is well controlled when the medication concentration is at least a specific concentration, but the medication clears too quickly due to that patient’s unique metabolism (e.g., falling below the threshold within X minutes), then Patient C could be recommended to modify behavior (e.g., adjust an amount or timing of protein consumption) and/or change to a more frequent dosing to accommodate for an expected more rapid clearance.

[0253] By way of further example, if Patient D is sometimes well controlled and sometimes not, when the medication (e.g., L-DOPA) drops below a specific concentration then patient D could be recommended to be on an active therapy management medication regimen which would include predictive recommendations as to when to take medication that could vary on a daily basis such that the concentration of medication would be maintained within a desired therapeutic window for that patient. This would be the case for patients that are on injectable, oral or other types of bolus therapy and/or for patients on pump therapy in which case the sensor system could recommend that the patient add a bolus or adjust dose if the concentration of L-DOPA is likely to shortly trend below the effective therapeutic window previously established by symptom tracking for that patient.

[0254] In another example, as part of generating and presenting actionable treatment data at block 2312, for patients that are on a secondary therapy meant to slow down the progression of PD, the therapy management engine 114 can use the symptom feedback sensor(s) 206 and the other sensor(s) 209 to determine which specific secondary therapy being tried is most effective for that specific patient in slowing, halting, and/or reversing disease progress. In certain aspects, monitoring L-DOPA concentration and/or other medication meant to modify symptoms but not reduce disease progression, can be used by the therapy management engine 114 to distinguish between symptom control due to effective medication versus symptom control due to reducing, halting, or reversing disease progression.

[0255] In another example, as part of generating and presenting actionable treatment data at block 2312, the therapy management engine 114 can provide L-DOPA dosage timing recommendations based on the patient’s expected and/or self-reported daily acts of living. At times, L-DOPA and other medications can be affected by the diet of a patient. For example, consuming protein can interfere with L-DOPA and so it may be advisable for the patient take L-DOPA a predetermined amount of time in advance of a meal (e.g., 1-2 hours before the meal) or, if symptoms are manageable, to wait a predetermined amount of time after a meal (e.g., 1-2 hours after the meal has concluded) to take L-DOPA. Taking L-DOPA in the wrong window dramatically changes the bioavailability in the patient and protein can be especially detrimental to the goal of improving L-DOPA concentration. In various aspects, this feature can be used in conjunction with a meal and dietary intake logging tool that can learn the effect that various food and beverage items have on uptake and absorption characteristics of L-DOPA.

[0256] Continuing the example of providing L-DOPA dosage timing recommendations, in certain aspects, the therapy management engine 114 can generate and present a historical determination of when a patient typically eats (e.g., based on user-entered meals) and/or when a patient is likely to feel hunger (e.g., based on outputs of the CLM sensor system 104 and/or the other sensor(s) 209). In this manner, the therapy management engine 114 can provide, to the patient, a recommended time to take medication such that an applicable therapeutic window is maintained and such that the medication does not interfere with the patient’s historical meal cycle. According to this example, the patient can be prompted to take medication prior to feeling hunger (e.g., 1-2 hours in advance). In this way, the patient would be able to eat when they feel hunger, instead of having to wait 1-2 hours due to L-DOPA administration. Additionally, the patient would not feel as tempted to be non-compliant with medication due to hunger, thereby improving the therapeutic effect of the L-DOPA medication. The therapy management engine 114 can recommend when to consume food (e.g., take proteins or meals), when to take medication such as L-DOPA. In addition, or alternatively, the therapy management engine 114 can provide active therapy management regarding when meals are off-limits, for example, due to medication.

[0257] In another example, as part of generating and presenting actionable treatment data at block 2312, the therapy management engine 114 can, via the CLM sensor system 104, the symptom feedback sensor(s) 206 and/or the other sensor(s) 209, track patient compliance with taking their medication. If the therapy management engine 114 determines that the patient is consuming medication and not waiting long enough before the meal has been consumed (e.g., as indicate by other analytes measured by the other sensor(s) 209), then that fact can be identified and recorded by the therapy management engine 114, for example, so the patient and/or their care team can determine behavioral and/or therapy modifications to improve medication efficacy.

[0258] In another example, as part of generating and presenting actionable treatment data at block 2312, the therapy management engine 114 can, via the CLM sensor system 104, the symptom feedback sensor(s) 206 and/or the other sensor(s) 209, identify which meals (e g., e.g., chicken versus whey protein versus cereal and milk) are most impactful to the patient’s therapeutic efficacy and symptom presentation. In addition, or alternatively, the therapy management engine 114 can recommend different meal types, recommend a timing of when to consume medication and/or meals, and/or make recommendations that would enable the patient to control symptoms and/or achieve other goals.

[0259] In another example, as part of generating and presenting actionable treatment data at block 2312, the therapy management engine 114 can capture an impact of meals on L-DOPA levels. For example, the therapy management engine 114 can notify the patient or other users if, for example, L-DOPA levels change in response to a meal (e g., change more than configurable predefined amount). In this way, the therapy management engine 114 can provide real-time information about L-DOPA bioavailability in response to meals.

[0260] In another example, as part of generating and presenting actionable treatment data at block 2312, the therapy management engine 114 can determine the effectiveness of combination therapy of L-DOPA and/other medications or treatments (e.g., MAO-B inhibitors, dopamine agonists, DBS, or TAPS). For example, for patients that have just received a DBS, the therapy management engine 114 can titrate to an appropriate dose of L-DOPA (e.g., a dose within a therapeutic window) using the CLM sensor system 104, the symptom feedback sensor(s) 206 and/or the other sensor(s) 209, as discussed above and below with respect to FIG. 26. In various aspects, adjustment of a DBS signal frequency and intensity can be done in accordance with the incorporation of monitoring of L-DOPA. Additionally, or alternatively, during periods of expected low L-DOPA concentration, such as when consuming a high protein meal, the therapy management engine 114 can recommend, or cause, DBS activity to be increased or modified such that the symptomatic control would be the most effective possible for that patient’s condition during periods when L-DOPA may or may not be as therapeutically available in the body. More generally, in certain aspects, the therapy management engine 114 can inform an appropriate DBS waveform (e.g., frequency, repetition rate, magnitude of stimulus, etc.). In some aspects, therapy using DBS can be implemented via closed-loop control.

[0261] In addition, or alternatively, the therapy management engine 114 can monitor both L-DOPA treatment and other therapies such as DBS, TAPS, or the like. In certain aspects, the therapy management engine 114 can determine or predict disease progression or regression based on multiple scenarios. For example, one scenario may involve L-DOPA treatment without any secondary therapies, and another scenario may involve L-DOPA treatment in combination with one or more secondary therapies that are meant to slow, halt, or reverse the progression of disease (e.g., DBS or TAPS).

[0262] In another example, as part of generating and presenting actionable treatment data at block 2312, the therapy management engine 114 can monitor gastric motility. In certain aspects, monitoring gastric motility can be advantageous for patients on oral L-DOPA therapy and/or other oral PD medications. In certain aspects, the therapy management engine 114 can monitor gastric motility as compared to bioavailability of L-DOPA measured, for example, by the CLM sensor system 104. For example, L-DOPA co-formulated with glucose or consumed orally with glucose could be monitored, for example, via the CLM sensor system 104 along with an interstitial sensor for glucose (e.g., among the other sensor(s) 209 of FIG. 2). In addition, or alternatively, the CLM sensor system 104 can include a sensor (or sensors) configured to monitor both L-DOPA and glucose.

[0263] Continuing the above example, if a glucose peak or an L-DOPA peak is sensed as delayed compared to a time of oral consumption, the therapy management engine 114 can determine if the time entry was off by the patient or if the patient has symptomatic gastric motility dysfunction. In this example, the therapy management engine 114 can generate glucose curves for gastric dysfunction that demonstrate a delayed gastric emptying as compared to controls. Based on the glucose curves for gastric dysfunction, the therapy management engine 114 can predict the bioavailability of L-DOPA throughout an increasing future concentration or decreasing future concentration.

[0264] In addition, or alternatively, the therapy management engine 114 can use information related to monitoring gastric motility to provide therapy recommendations to adjust dosage and/or provide a more personalized dosage recommendation as to when to consume L-DOPA orally with or without food, drink, or other medications that are known to impact gastric motility. For example, if the therapy management engine 114 detects that delayed gastric motility is affecting absorption of orally consumed L-DOPA and/or other medications, the therapy management engine 114 can instruct the patient not to consume the medication with certain types of foods (e.g., solid foods, foods high in fiber, foods high in protein, or foods high in fats). Instead, the therapy management engine 114 can recommend that the patient consume the medication with liquid water, and that the patient not consume food or other drink during a period spanning predetermined times before and after oral administration of the medication (e.g. approximately 30-45 minutes before the oral administration until approximately 30-45 minutes after the administration).

[0265] In certain aspects, the therapy management engine 114 can provide personalized recommendations related to monitoring gastric motility. For example, the therapy management engine 114 can recommend that the patient not consume L-DOPA unless the consumption is to occur 1 hour before, or two hours after, a meal (e.g., a detected, predicted, or user-indicated meal) that contains protein. In certain aspects, such a recommendation can minimize dietary effects on bioavailability of the compound.

[0266] In addition, or alternatively, if the therapy management engine 114 detects symptomatic gastroparesis or impaired gastric motility for a patient, the therapy management engine 114 can notify the patient, the care team, healthcare providers, family members, or other users to inform so that action or medication can be provided to the patient. In this way, the action or medication can serve to minimize resultant deleterious effects on quality of life of the patient and/or a bioavailability of the orally consumed therapy.

[0267] In addition, or alternatively, if the therapy management engine 114 detects no such beneficial improvements in bioavailability, gastric function, and/or parkinsonian akinesias or dyskinesia after adjustments to diet, lifestyle, and or medication, then the therapy management engine 114 can recommend that alternative or more advanced therapies be pursued due to the progression of the patient’s PD and/or gastric dysfunction. The recommendation can include, for example, a pump-based L-DOPA delivery system, DBS, TAPS, and/or other treatment options.

[000268] After block 2312, the process 2300 ends.

[000269] FIG. 24A is a graph illustrating an example time-based correlation of the L-DOPA measurement information, the symptom information (e.g., motor function information), and/or the L-DOPA dosing information, in accordance with certain aspects of the present disclosure. FIG 24A further illustrates example characterizations of L-DOPA information based on degree of motor symptom control, with “NOT WELL CONTROLLED” symptoms being attributed to either relatively low L-DOPA levels or motor complications caused by relatively high L-DOPA levels.

[0270] In certain aspects, FIG 24A additionally illustrates a therapeutic benefit that a CLM sensor system, such as the CLM sensor system 104, can have in adjusting L-DOPA dosages over time. In addition, or alternatively, in certain aspects, a time at which an L-DOPA concentration changes, and/or at time at which symptom onset occurs relative to such changes, can facilitate staging of PD.

[0271] FIG 24B is a graph illustrating an example time-based correlation of the L-DOPA measurement information, the symptom information (e.g., motor function information), and/or the L-DOPA dosing information, in accordance with certain aspects of the present disclosure. In particular, FIG 24B plots symptoms versus L-DOPA concentrations over time. In various aspects, the combination of a CLM sensor system (e.g., the CLM sensor system 104) and symptom feedback sensors (e.g., the symptom feedback sensor(s) 206) can facilitate staging of PD, for example, by tracking a size of a therapeutic window over time.

[0272] FIGS. 25A-B illustrate example user interface outputs to a patient or other user, in accordance with certain aspects of the present disclosure. In particular, FIG. 25A illustrates an example user interface output indicating a dangerous zone for L-DOPA levels, as generally discussed above. FIG. 25B illustrates another example user interface output indicating a dangerous zone for L-DOPA levels, as generally discussed above.

[0273] FIG. 26 illustrates an example of a process 2600 for automatic treatment determination, in accordance with certain aspects of the present disclosure. In some cases, the process 2600 can be performed as part of the block 2312 of the process 2300 of FIG. 23. In some aspects, the process 2600 can be executed, for example, by the therapy management engine 114 of FIG. 1. In addition, or alternatively, the process 2600 can be executed, for example, by the CLM sensor system 104. In addition, or alternatively, the process 2600 can be executed, for example, by the application 106 of FIGS. 1 and 3. In addition, or alternatively, the process 2600 can be executed generally by any of the display devices 107, 210, 220, 230, and/or 240. Although any number of systems, in whole or in part, can implement the process 2600, to simplify discussion, the process 2600 will be described in relation to the therapy management engine 114 of FIG. 1. [0274] At block 2602, the therapy management engine 114 identifies a treatment objective.

Examples of treatment objectives include:

(1) maximize overall time window (e.g., a predicted amount of time) in a therapeutic;

(2) maximize time (e.g., a predicted amount of time) in a therapeutic window during the day (e.g., 8 am - 6 pm), such that there may be a higher tolerance for “POOR” control at other times of the day;

(3) maximize time window (e.g., a predicted amount of time) in a therapeutic window at night (e.g., 11 pm - 7 am), such that there may be a higher tolerance for “POOR” control at other times of the day;

(4) minimize motor complications caused by relatively high L-DOPA levels (e.g., muscle rigidity), such that there may be a higher tolerance for “POOR” control due to relatively low L-DOPA levels;

(5) minimize overall medication, for example, such that there may be a higher tolerance for “POOR” motor control due to low L-DOPA levels (e.g., muscle tremor);

(6) minimize “POOR” control due to relatively low L-DOPA levels, such that there may be a higher tolerance for motor complications caused by relatively high L- DOPA levels; and

(7) achieve maximum symptom control during a certain future time interval (e.g., a future time interval indicated by the user, such as a wedding from 2 - 4 pm on Saturday), such that there may be a higher tolerance for “POOR” control at other times of the day (e.g., during other future time intervals).

(8) determining a recommended timing for a given L-DOPA dosage (e.g., an optimal timing for a dosage indicated by the patient).

[0275] At block 2604, the therapy management engine 114 automatically determines a treatment that achieves the treatment objective. For example, with respect to the process 2300 of FIG. 23, the therapy management engine 114 can determine L-DOPA dosing information for the patient based on the characterizations of the L-DOPA measurement information, the personalized therapeutic window, and/or other available information. In certain aspects utilizing a continuous L-DOPA pump as discussed above, the therapy management engine 114 can determine the dosing information in the form of an adjustment to a basal rate or as a bolus. In certain aspects not utilizing a continuous L-DOPA pump, the therapy management engine 114 can determine the dosing information for patient administration (e.g., oral administration).

[0276] In some aspects, the treatment can be determined, at least in part, using rules based on patient characteristics (e.g., age, state of disease progression, and demographics), the L-DOPA information (e.g., L-DOPA levels), the symptom information (e.g., motor function information such as tremor values), and/or the like. In some aspects, the rules enable incremental changes that result in gradually improved motor symptom control.

[0277] In some aspects, the treatment can be generated using a model at least partially based in machine learning (e.g., supervised learning). For example, the model can be trained on datasets for a large set of patients. According to this example, the datasets on which the model is trained can include records detailing sets of features such as patient characteristics (e.g., age, state of disease progression, and demographics), L-DOPA information (e.g., L-DOPA levels), symptom information (e.g., motor function information such as tremor value), and/or the like. Each record can further include, or be labeled with, dosing information (e.g., dosage and timing) given the features of the record. Therefore, according to this example, the therapy management engine 114 can use the aforementioned information and/or other available information to determine the dosing information given the example model described above. In some aspects, the model can be updated for the patient based on the patient’s L-DOPA information (e.g., L-DOPA levels) and symptom information (e.g., motor function information such as tremor values) for particular L-DOPA dosing information.

[0278] At block 2606, the therapy management engine 114 can facilitate or cause treatment based on the automatically determined treatment. For example, the therapy management engine 114 command an L-DOPA delivery device, such as the L-DOPA delivery device 208 of FIG. 2 or the L-DOPA pump 608 of FIG. 6A, to deliver L-DOPA based on an adjusted basal rate, a determined bolus, etc. In addition, or alternatively, therapy management engine 114 can facilitate or cause treatment by presenting the automatically determined treatment to the patient, caregiver, or other use. After block 2606, the process 2600 ends.

[0279] In some aspects, at least portions of the processes 2300 and/or 2600 described above relative to FIGS. 23 and 26, respectively, can be performed continuously (e.g., on a defined interval). For example, on a continuous basis (e.g., on a defined interval), the therapy management engine 114 can receive and correlate L-DOPA measurement information from the CLM and symptom information from the symptom feedback sensor(s) 206 (see, e.g., blocks 2302-2306 of FIG. 23) and, based thereon, generate and present actionable treatment data of any of the types described above, for example, relative the block 2312 of FIG. 23. For example, on a continuous basis (e.g., on a defined interval), the therapy management engine 114 can determine dosing information in the fashion described relative to FIG. 26. In certain aspects, the L-DOPA measurement information and the symptom information can serve as continuous feedback for updating rules and/or models for determining L-DOPA dosing information.

[0280] In some aspects, the therapy management engine 114 can determine certain actionable treatment data, such as dosing information for the patient, in response to a defined trigger related to symptom control. For example, the trigger can be detection that the patient is trending towards “POOR” symptom control (e.g., “POOR motor symptom control), as indicated by symptom information received from the symptom feedback sensor(s) 206. The dosing information can be determined, for example, in similar fashion to the L-DOPA dosing information described above relative to FIG. 26. For example, in aspects utilizing a continuous L-DOPA pump, the dosing information responsive to the trigger may be a bolus. By way of further example, in aspects not utilizing a continuous L-DOPA pump, the dosing information responsive to the trigger may be a recommended oral administration of L-DOPA. In certain aspects involving a continuous L-DOPA pump, the CLM sensor system 104, for example, can operate in either closed-loop or open-loop mode.

[0281] FIG. 27 is a block diagram depicting a computing device 2700 configured for continuous L-DOPA monitoring and therapy management, according to certain embodiments disclosed herein. Although depicted as a single physical device, in embodiments, the computing device 2700 may be implemented using virtual device(s), and/or across a number of devices, such as in a cloud environment. As illustrated, the computing device 2700 includes a processor 2705, a memory 2710, a storage 2715, a network interface 2725, and one or more I/O interfaces 2720. In the illustrated embodiment, the processor 2705 retrieves and executes programming instructions stored in the memory 2710, as well as stores and retrieves application data residing in the storage 2715. The processor 2705 is generally representative of a single CPU and/or GPU, multiple CPUs and/or GPUs, a single CPU and/or GPU having multiple processing cores, and the like.

[0282] The memory 2710 is generally included to be representative of a random access memory (RAM). The storage 2715 may be any combination of disk drives, flash-based storage devices, and the like, and may include fixed and/or removable storage devices, such as fixed disk drives, removable memory cards, caches, optical storage, network attached storage (NAS), or storage area networks (SAN).

[0283] In some embodiments, the I/O devices 2735 (such as keyboards, monitors, etc.) can be connected via the I/O interface(s) 2720. Further, via the network interface 2725, the computing device 2700 can be communicatively coupled with one or more other devices and components, such as the patient database 110 and/or the historical records database 112. In certain embodiments, the computing device 2700 is communicatively coupled with other devices via a network, which may include the Internet, local network(s), and the like. The network may include wired connections, wireless connections, or a combination of wired and wireless connections. As illustrated, the processor 2705, memory 2710, storage 2715, network interface(s) 2725, and the I/O interface(s) 2720 are communicatively coupled by one or more interconnects 2730. In certain embodiments, the computing device 2700 is representative of the display device 107 associated with the user. In certain embodiments, as discussed above, the display device 107 can include the user’s laptop, computer, smartphone, and the like. In another embodiment, the computing device 2700 is a server executing in a cloud environment.

[0284] In the illustrated embodiment, the storage 2715 includes the patient profile 118. The memory 2710 includes the therapy management engine 114, which itself includes the DAM 116. The therapy management engine 114 is executed by the computing device 2700 to perform therapy management operations as discussed relative to FIGS. 1-26. Example Embodiments

[0285] According to an embodiment, a method includes identifying a current levodopa (L-DOPA) level in a patient’s system and determining an impact of the current L-DOPA level on the patient. The method also includes adjusting a dosage of L-DOPA administered to the patient based on the current L-DOPA level and the impact. Other embodiments may include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of the methods.

[0286] The current L-DOPA level may be determined utilizing a continuous L-DOPA monitor.

[0287] The impact may be determined by identifying one or more current physical characteristics of the patient. The one or more current physical characteristics may include an assessment of the patient’ s current motor function. The one or more current physical characteristics may include an occurrence of one or more tremors by the patient. The one or more current physical characteristics may include an occurrence of a fall by the patient. The one or more current physical characteristics may be identified utilizing at least one of an accelerometer, gyroscope, inclinometer, or a magnetometer. The one or more current physical characteristics may include vocal biomarkers identified utilizing a microphone. The one or more current physical characteristics may be identified utilizing an electromyography (EMG) sensor. The one or more current physical characteristics may be identified utilizing a galvanic skin response (GSR) sensor.

[0288] L-DOPA may be administered to the patient utilizing a continuous L-DOPA pump.

[0289] According to another embodiment, a system includes a memory having executable instructions. The system also includes a processor in data communication with the memory. The processor is configured to execute the executable instructions to identify a current levodopa (L-DOPA) level in a patient’s system and to determine an impact of the current L-DOPA level on the patient. The processor is also configured to execute the executable instructions to adjust a dosage of L-DOPA administered to the patient based on the current L-DOPA level and the impact.

[0290] The current L-DOPA level may be determined utilizing a continuous L-DOPA monitor.

[0291] The impact may be determined by identifying one or more current physical characteristics of the patient. The one or more current physical characteristics may include an assessment of the patient’ s current motor function. The one or more current physical characteristics may include an occurrence of one or more tremors by the patient. The one or more current physical characteristics may include an occurrence of a fall by the patient. The one or more current physical characteristics may be identified utilizing at least one of an accelerometer, gyroscope, inclinometer, or a magnetometer. The one or more current physical characteristics may include vocal biomarkers identified utilizing a microphone.

[0292] According to another embodiment, a monitoring system includes a continuous levodopa (L-DOPA) sensor configured to generate L-DOPA measurements associated with L- DOPA levels of a patient, a symptom feedback sensor configured to generate symptom information measuring a physical response of the patient to the L-DOPA levels, a memory having executable instructions, and a processor in data communication with the memory, the continuous L-DOPA sensor, and the symptom feedback sensor. The processor is configured to execute the executable instructions to receive, from the continuous L-DOPA sensor, the L-DOPA measurements, and to receive, from the symptom feedback sensor, the symptom information. The processor is configured to execute the executable instructions to characterize the L-DOPA measurements in achieving symptom control for the patient based on the symptom information, and to define a personalized therapeutic window for the patient based on the characterization of the L-DOPA measurements. The personalized therapeutic window includes a target range of L- DOPA levels to minimize symptoms.

[0293] According to another embodiment, a method includes identifying a current levodopa (L-DOPA) level in a patient’s system and receiving an input from a surface electromyography (sEMG) sensor positioned in relation to a muscle of the patient. The method also includes determining an impact of the current L-DOPA level on the patient based on the input from the sEMG sensor. The method also includes adjusting a dosage of L-DOPA administered to the patient based on the current L-DOPA level and the impact. Other embodiments may include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of the methods.

[0294] The current L-DOPA level may be determined utilizing a continuous L-DOPA monitor.

[0295] The impact may be determined by identifying one or more current physical characteristics of the patient. The one or more current physical characteristics may include an assessment of the patient’ s current motor function. The one or more current physical characteristics may include an occurrence of one or more muscle tremors by the patient. The one or more current physical characteristics may include an occurrence of a fall by the patient. The one or more current physical characteristics may include an occurrence of muscle rigidity by the patient. The one or more current physical characteristics may include an occurrence of dyskinesia by the patient.

[0296] The method may further include receiving an input from an accelerometer, where the determined impact of the current L-DOPA level on the patient may be further based on the input from the accelerometer.

[0297] L-DOPA may be administered to the patient utilizing a continuous L-DOPA pump.

[0298] According to another embodiment, a system includes a memory having executable instructions and a processor in data communication with the memory. The processor is configured to execute the executable instructions to identify a current levodopa (L-DOPA) level in a patient’s system and receive an input from a surface electromyography (sEMG) sensor positioned in relation to a muscle of the patient. The processor is also configured to execute the executable instructions to determine an impact of the current L-DOPA level on the patient based on the input from the sEMG sensor. The processor is also configured to execute the executable instructions to adjust a dosage of L-DOPA administered to the patient based on the current L-DOPA level and the impact.

[0299] The current L-DOPA level may be determined utilizing a continuous L-DOPA monitor.

[0300] The impact may be determined by identifying one or more current physical characteristics of the patient. The one or more current physical characteristics may include an assessment of the patient’ s current motor function. The one or more current physical characteristics may include an occurrence of one or more muscle tremors by the patient. The one or more current physical characteristics may include an occurrence of a fall by the patient. The one or more current physical characteristics may include an occurrence of muscle rigidity by the patient. The one or more current physical characteristics may include an occurrence of dyskinesia by the patient.

[0301] The processor may be further configured to execute the executable instructions to receive an input from an accelerometer, where the determined impact of the current L-DOPA level on the patient may be further based on the input from the accelerometer.

[0302] According to another embodiment, a computer-program product includes a non-transitory computer-usable medium having computer-readable program code embodied therein. The computer-readable program code is adapted to be executed to implement a method. The method includes identifying a current levodopa (L-DOPA) level in a patient’s system and receiving an input from a surface electromyography (sEMG) sensor positioned in relation to a muscle of the patient. The method also includes determining an impact of the current L-DOPA level on the patient based on the input from the sEMG sensor. The method also includes adjusting a dosage of L-DOPA administered to the patient based on the current L-DOPA level and the impact.

[0303] The current L-DOPA level may be determined utilizing a continuous L-DOPA monitor.

[0304] The determined impact of the current L-DOPA level on the patient may be further based on the input from the accelerometer.

[0305] L-DOPA may be administered to the patient utilizing a continuous L-DOPA pump.

[0306] According to another embodiment, a computer-program product includes a non- transitory computer-usable medium having computer-readable program code embodied therein. The computer-readable program code is adapted to be executed to implement a method. The method includes identifying a current levodopa (L-DOPA) level in a patient’s system and determining an impact of the current L-DOPA level on the patient. The method also includes adjusting a dosage of L-DOPA administered to the patient based on the current L-DOPA level and the impact.

[0307] According to another embodiment, a monitoring system includes a continuous levodopa (L-DOPA) sensor configured to generate L-DOPA measurements associated with L- DOPA levels of a patient, a surface electromyography (sEMG) sensor positioned in relation to a muscle of the patient and configured to generate symptom information measuring activity of the muscle, a memory having executable instructions, and a processor in data communication with the memory, the continuous L-DOPA sensor, and the sEMG sensor. The processor is configured to execute the executable instructions to receive, from the continuous L-DOPA sensor, the L- DOPA measurements, and to receive, from the sEMG sensor, the symptom information measuring the activity of the muscle. The processor is also configured to execute the executable instructions to characterize the L-DOPA measurements in achieving symptom control for the patient based on the symptom information, and to define a personalized therapeutic window for the patient based on the characterization of the L-DOPA measurements. The personalized therapeutic window includes a target range of L-DOPA levels to minimize symptoms.

[0308] According to another embodiment, a method includes identifying a future time interval for Parkinson’s disease (PD) symptom control for a patient. The method also includes automatically determining levodopa (L-DOPA) dosing information that prioritizes PD symptom control during the future time interval over PD symptom control in at least one other future time interval. The method also includes facilitating treatment of the patient based on the automatically determined L-DOPA dosing information. Other embodiments may include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of the methods.

[0309] The automatically determining may include maximizing PD symptom control during the future time interval based on a personalized therapeutic window for the patient.

[0310] The automatically determining may include maximizing a predicted amount of time in a therapeutic window during the future time interval.

[0311] The L-DOPA dosing information may include information related to dosage and timing.

[0312] According to another embodiment, a method includes receiving a levodopa (L-DOPA) dosage for a patient. The method also includes automatically determining recommended timing for the L-DOPA dosage based on a therapeutic window for the patient. The method also includes facilitating treatment of the patient based on the recommended timing for the L-DOPA dosage. Other embodiments may include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of the methods.

[0313] The automatically determining may be further based on meal timing for the patient.

[0314] The automatically determining may be further based on an amount of protein intake by the patient.

[0315] According to another embodiment, a method includes identifying a current levodopa (L-DOPA) level in a patient’s system. The method also includes receiving a real-time recording of the patient’s voice. The method also includes determining an impact of the current L-DOPA level on the patient based on an analysis of the patient’s voice in the recording. The method also includes adjusting a dosage of L-DOPA administered to the patient based on the current L-DOPA level and the impact. Other embodiments may include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of the methods.

[0316] The determining may include generating time-indexed information related to a patient’s speech function based on the recording of the patient’s voice.

[0317] The determined impact may be based on a match between the time-indexed information and at least one predetermined speech function signature.

[0318] According to another embodiment, a method includes identifying a current levodopa (L-DOPA) level in a patient’s system. The method also includes identifying one or more current physical characteristics of the patient based on information received from an inertial measurement unit associated with the patient. The method also includes determining an impact of the current L-DOPA level on the patient based on the one or more current physical characteristics. The method also includes adjusting a dosage of L-DOPA administered to the patient based on the current L-DOPA level and the impact. Other embodiments may include corresponding computer systems, apparatus, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of the methods.

[0319] The one or more current physical characteristics include an assessment of the patient’s current motor function.

[0320] The one or more current physical characteristics include an occurrence of one or more tremors by the patient.

[0321] The one or more current physical characteristics may include an occurrence of a fall by the patient.

[0322] The methods disclosed herein comprise one or more steps or actions for achieving the methods. The method steps and/or actions may be interchanged with one another without departing from the scope of the claims. In other words, unless a specific order of steps or actions is specified, the order and/or use of specific steps and/or actions may be modified without departing from the scope of the claims. [0323] As used herein, a phrase referring to “at least one of’ a list of items refers to any combination of those items, including single members. As an example, “at least one of: a, b, or c” is intended to cover a, b, c, a-b, a-c, b-c, and a-b-c, as well as any combination with multiples of the same element (e.g., a-a, a-a-a, a-a-b, a-a-c, a-b-b, a-b-c, b-b, b-b-b, b-b-c, c-c, and c-c-c or any other ordering of a, b, and c).

[0324] The previous description is provided to enable any person skilled in the art to practice the various aspects described herein. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects. Thus, the claims are not intended to be limited to the aspects shown herein, but is to be accorded the full scope consistent with the language of the claims, wherein reference to an element in the singular is not intended to mean “one and only one” unless specifically so stated, but rather “one or more.” Unless specifically stated otherwise, the term “some” refers to one or more. All structural and functional equivalents to the elements of the various aspects described throughout this disclosure that are known or later come to be known to those of ordinary skill in the art are expressly incorporated herein by reference and are intended to be encompassed by the claims. Moreover, nothing disclosed herein is intended to be dedicated to the public regardless of whether such disclosure is explicitly recited in the claims. No claim element is to be construed under the provisions of 35 U.S.C. §112(f) unless the element is expressly recited using the phrase “means for” or, in the case of a method claim, the element is recited using the phrase “step for.”

[0325] While various examples of the invention have been described above, it should be understood that they have been presented by way of example only, and not by way of limitation. Likewise, the various diagrams may depict an example architectural or other configuration for the disclosure, which is done to aid in understanding the features and functionality that can be included in the disclosure. The disclosure is not restricted to the illustrated example architectures or configurations, but can be implemented using a variety of alternative architectures and configurations. Additionally, although the disclosure is described above in terms of various example examples and aspects, it should be understood that the various features and functionality described in one or more of the individual examples are not limited in their applicability to the particular example with which they are described. They instead can be applied, alone or in some combination, to one or more of the other examples of the disclosure, whether or not such examples are described, and whether or not such features are presented as being a part of a described example. Thus the breadth and scope of the present disclosure should not be limited by any of the above-described example examples.

[0326] All references cited herein are incorporated herein by reference in their entirety. To the extent publications and patents or patent applications incorporated by reference contradict the disclosure contained in the specification, the specification is intended to supersede and/or take precedence over any such contradictory material.

[0327] Unless otherwise defined, all terms (including technical and scientific terms) are to be given their ordinary and customary meaning to a person of ordinary skill in the art, and are not to be limited to a special or customized meaning unless expressly so defined herein.

[0328] Terms and phrases used in this application, and variations thereof, especially in the appended claims, unless otherwise expressly stated, should be construed as open ended as opposed to limiting. As examples of the foregoing, the term ‘including’ should be read to mean ‘including, without limitation,’ ‘including but not limited to,’ or the like; the term ‘comprising’ as used herein is synonymous with ‘including,’ ‘containing,’ or ‘characterized by,’ and is inclusive or open-ended and does not exclude additional, unrecited elements or method steps; the term ‘having’ should be interpreted as ‘having at least;’ the term ‘includes’ should be interpreted as ‘includes but is not limited to;’ the term ‘example’ is used to provide example instances of the item in discussion, not an exhaustive or limiting list thereof; adjectives such as ‘known’, ‘normal’, ‘standard’, and terms of similar meaning should not be construed as limiting the item described to a given time period or to an item available as of a given time, but instead should be read to encompass known, normal, or standard technologies that may be available or known now or at any time in the future; and use of terms like ‘preferably,’ ‘preferred,’ ‘desired,’ or ‘desirable,’ and words of similar meaning should not be understood as implying that certain features are critical, essential, or even important to the structure or function of the invention, but instead as merely intended to highlight alternative or additional features that may or may not be utilized in a particular example of the invention. Likewise, a group of items linked with the conjunction ‘and’ should not be read as requiring that each and every one of those items be present in the grouping, but rather should be read as ‘and/or’ unless expressly stated otherwise. Similarly, a group of items linked with the conjunction ‘or’ should not be read as requiring mutual exclusivity among that group, but rather should be read as ‘and/or’ unless expressly stated otherwise.

[0329] The term “comprising as used herein is synonymous with “including,” “containing,” or “characterized by” and is inclusive or open-ended and does not exclude additional, unrecited elements or method steps.

[0330] All numbers expressing quantities of ingredients, reaction conditions, and so forth used in the specification are to be understood as being modified in all instances by the term ‘about.’ Accordingly, unless indicated to the contrary, the numerical parameters set forth herein are approximations that may vary depending upon the desired properties sought to be obtained. At the very least, and not as an attempt to limit the application of the doctrine of equivalents to the scope of any claims in any application claiming priority to the present application, each numerical parameter should be construed in light of the number of significant digits and ordinary rounding approaches.

[0331] Furthermore, although the foregoing has been described in some detail by way of illustrations and examples for purposes of clarity and understanding, it is apparent to those skilled in the art that certain changes and modifications may be practiced. Therefore, the description and examples should not be construed as limiting the scope of the invention to the specific examples and examples described herein, but rather to also cover all modification and alternatives coming with the true scope and spirit of the invention.

Claims

1. A method comprising: identifying a current levodopa (L-DOPA) level in a patient’s system; determining an impact of the current L-DOPA level on the patient; and adjusting a dosage of L-DOPA administered to the patient based on the current L-DOPA level and the impact.

2. The method of claim 1, wherein the current L-DOPA level is determined utilizing a continuous L-DOPA monitor.

3. The method of claim 1, wherein the impact is determined by identifying one or more current physical characteristics of the patient.

4. The method of claim 3, wherein the one or more current physical characteristics include an assessment of the patient’s current motor function.

5. The method of claim 3, wherein the one or more current physical characteristics include an occurrence of one or more tremors by the patient.

6. The method of claim 3, wherein the one or more current physical characteristics include an occurrence of a fall by the patient.

7. The method of claim 3, wherein the one or more current physical characteristics are identified utilizing at least one of an accelerometer, gyroscope, inclinometer, or a magnetometer.

8. The method of claim 3, wherein the one or more current physical characteristics include vocal biomarkers identified utilizing a microphone.

9. The method of claim 3, wherein the one or more current physical characteristics are identified utilizing an electromyography (EMG) sensor.

10. The method of claim 3, wherein the one or more current physical characteristics are identified utilizing a galvanic skin response (GSR) sensor.

11. The method of claim 1, wherein L-DOPA is administered to the patient utilizing a continuous L-DOPA pump.

12. A system, comprising: a memory comprising executable instructions; and a processor in data communication with the memory, wherein the processor is configured to execute the executable instructions to: identify a current levodopa (L-DOPA) level in a patient’s system; determine an impact of the current L-DOPA level on the patient; and adjust a dosage of L-DOPA administered to the patient based on the current L- DOPA level and the impact.

13. The system of claim 12, wherein the current L-DOPA level is determined utilizing a continuous L-DOPA monitor.

14. The system of claim 12, wherein the impact is determined by identifying one or more current physical characteristics of the patient.

15. A computer-program product comprising a non-transitory computer-usable medium having computer-readable program code embodied therein, the computer-readable program code adapted to be executed to implement a method comprising: identifying a current levodopa (L-DOPA) level in a patient’s system; determining an impact of the current L-DOPA level on the patient; and adjusting a dosage of L-DOPA administered to the patient based on the current L-DOPA level and the impact.