WO2023117560A1

WO2023117560A1 - Tool for identifying measures against hypertension and for their monitoring

Info

Publication number: WO2023117560A1
Application number: PCT/EP2022/085486
Authority: WO
Inventors: Frank Kramer; Eren Metin ELCI; Daniel Franz FREITAG; Hans-Peter Podhaisky; Töresin KARAKOYUN; Daniel PAULSON; Michael Kremliovsky; Ying Chen
Original assignee: Bayer Aktiengesellschaft
Priority date: 2021-12-20
Filing date: 2022-12-13
Publication date: 2023-06-29

Abstract

Systems, methods, and computer programs disclosed herein relate to identifying and monitoring one or more measures for treating hypertension in a patient.

Description

Tool for identifying measures against hypertension and for their monitoring

FIELD

BACKGROUND

Hypertension, also known as high or raised blood pressure, is a condition in which the blood vessels have persistently raised pressure. Blood is carried from the heart to all parts of the body in the vessels. Each time the heart beats, it pumps blood into the vessels. Blood pressure is created by the force of blood pushing against the walls of blood vessels as it is pumped by the heart. The higher the pressure, the harder the heart has to pump.

Hypertension is a serious medical condition and can increase the risk of heart, brain, kidney, and other diseases. It is a major cause of premature death worldwide.

Changing lifestyle may help control and manage high blood pressure. The following measures are generally recommended to lower blood pressure: consuming less salt and alcohol, getting regular physical activity, maintaining a healthy weight, losing weight in case of overweight, and/or not smoking.

However, sometimes lifestyle changes are not enough, and medication is needed to lower blood pressure.

There are several types of drugs used to treat high blood pressure, including: angiotensin -converting enzyme (ACE) inhibitors, angiotensin II receptor blockers (ARBs), diuretics, beta-blockers, calcium channel blockers, alpha-blockers, alpha-agonists, renin inhibitors, and/or combination medications.

When treating hypertension, it is often necessary to identify appropriate measures for a patient. Appropriate measures may include the selection of an appropriate medication and/or dosing regimen. Blood pressure should be lowered to a range (target range) that is usually determined by the physician. It is important to identify an appropriate medication and/or dosing regimen for each individual patient. The physician usually uses an iterative procedure to slowly approach an appropriate dose.

There is a need for an efficient and effective solution to identify appropriate measures to treat hypertension in a patient.

SUMMARY

These and other tasks are solved by the subject matter of the independent patent claims. Preferred embodiments can be found in the dependent patent claims, the present description, and in the drawings.

In a first aspect, the present disclosure provides a computer-implemented method for identifying and monitoring measures for treating hypertension in a patient, the method comprising: receiving initial patient data, the initial patient data comprising data about a patient’s blood pressure, identifying, on the basis of the initial patient data, one or more measures for reducing the patient' s blood pressure, and/or receiving one or more measures for reducing the patient s blood pressure, determining a predicted blood pressure that applies in the event that the patient complies with the one or more measures, outputting the predicted blood pressure, receiving further patient data, wherein the further patient data comprises a current blood pressure and information about whether the patient complies with the one or more measures, updating the one or more measures and/or the predicted blood pressure one or more times until the patient’s blood pressure reaches a target range, outputting the updated measures and/or the updated predicted blood pressure.

In another aspect, the present disclosure provides a computer system, the computer system comprising a processor; and a memory storing an application program configured to perform, when executed by the processor, an operation, the operation comprising: receiving initial patient data, the initial patient data comprising data about a patient’s blood pressure, identifying, on the basis of the initial patient data, one or more measures for reducing the patient' s blood pressure, and/or receiving one or more measures for reducing the patient' s blood pressure, determining a predicted blood pressure that applies in the event that the patient complies with the one or more measures, outputting the predicted blood pressure, receiving further patient data, wherein the further patient data comprises a current blood pressure and information about whether the patient complies with the one or more measures, updating the one or more measures and/or the predicted blood pressure one or more times until the patient’s blood pressure reaches a target range, outputting the updated measures and/or the updated predicted blood pressure.

In another aspect, the present disclosure provides a non-transitory computer readable medium having stored thereon software instructions that, when executed by a processor of a computer system, cause the computer system to execute the following steps: receiving initial patient data, the initial patient data comprising data about a patient’s blood pressure, identifying, on the basis of the initial patient data, one or more measures for reducing the patient' s blood pressure, and/or receiving one or more measures for reducing the patient' s blood pressure, determining a predicted blood pressure that applies in the event that the patient complies with the one or more measures, outputting the predicted blood pressure, receiving further patient data, wherein the further patient data comprises a current blood pressure and information about whether the patient complies with the one or more measures, updating the one or more measures and/or the predicted blood pressure one or more times until the patient’s blood pressure reaches a target range, outputting the updated measures and/or the updated predicted blood pressure.

DETAILED DESCRIPTION

The invention will be more particularly elucidated below without distinguishing between the aspects of the invention (method, computer system, computer-readable storage medium). On the contrary, the following elucidations are intended to apply analogously to all the aspects of the invention, irrespective of in which context (method, computer system, computer-readable storage medium) they occur.

If steps are stated in an order in the present description or in the claims, this does not necessarily mean that the invention is restricted to the stated order. On the contrary, it is conceivable that the steps can also be executed in a different order or else in parallel to one another, unless one step builds upon another step, this absolutely requiring that the building step be executed subsequently (this being, however, clear in the individual case). The stated orders are thus preferred embodiments of the invention. As used herein, the articles “a” and “an” are intended to include one or more items and may be used interchangeably with “one or more” and “at least one.” As used in the specification and the claims, the singular form of “a”, “an”, and “the” include plural referents, unless the context clearly dictates otherwise. Where only one item is intended, the term “one” or similar language is used. Also, as used herein, the terms “has”, “have”, “having”, or the like are intended to be open-ended terms. Further, the phrase “based on” is intended to mean “based at least partially on” unless explicitly stated otherwise.

Some implementations will be described more fully hereinafter with reference to the accompanying drawings, in which some, but not all implementations are shown. Indeed, various implementations may be embodied in many different forms and should not be construed as limited to the implementations set forth herein; rather, these example implementations are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

The present disclosure provides means for identifying and monitoring one or more measures for treating hypertension in a patient.

The “patient” is usually a living being, preferably a mammal, particularly preferably a human. The patient is usually suffering from hypertension.

“Hypertension”, also known as high or raised or elevated blood pressure, is a condition in which the blood vessels have persistently raised pressure.

There are different definitions for the presence of hypertension in a person.

For example, in the 2017 Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults of the American College of Cardiology (ACC) and the American Heart Association (AHA) (D01: 10.1016/j.jacc.2017.11.006), five categories of blood pressure in adults are defined: normal (healthy): a healthy blood pressure reading is less than 120/80 millimeters of mercury (mm Hg); elevated: systolic blood pressure between 120-129 mmHg and diastolic blood pressure less than 80; stage 1: systolic blood pressure between 130-139 mmHg or diastolic blood pressure between 80-89 mmHg; stage 2: systolic blood pressure at least 140 mmHg or diastolic blood pressure at least 90 mmHg; hypertensive crisis: systolic blood pressure over 180 mmHg and/or diastolic blood pressure over 120 mmHg.

Another example are the 2018 Guidelines for the management of arterial hypertension of the European Society of Cardiology and the European Society of Hypertension. Here, hypertension is defined “as the level of blood pressure at which the benefits of treatment (either with lifestyle interventions or drugs) unequivocally outweigh the risks of treatment, as documented by clinical trials” (see, e.g.: https ://doi .org/ 10.1093/eurheartj/ehy339) .

It should be emphasized that the present disclosure is not limited to any specific definition of hypertension. The present disclosure is applicable to all reasonable definitions of hypertension, wherein "reasonable" means that a majority of physicians would assume the presence of hypertension if the criteria defined in the respective case were present.

According to the present disclosure, one or more measures to reduce blood pressure in a patient are selected based on initial patient data. Further patient data are collected to evaluate whether the patient complies with the one or more measures and whether the one or more measures are successful.

A "measure" for treating hypertension in a patient is any measure that can help bring the patient's blood pressure into a range that is considered normal (healthy) by a majority of physicians and/or that a physician has determined for the patient and/or a range at which the physician will not change a medicament and/or a dosage regimen for at least a period of several days, weeks, or months because, in the physician's opinion, a change would not improve the patient's health and/or would lower blood pressure too much and/or would result in undesirable side effects. Said range is also referred to as target range.

The target range can be, e.g., the range from 120/80 mmHg to 130/85 mmHg measured in the brachial artery with blood pressure cuff (peripheral blood pressure), wherein the first value (the value before the slash) indicates systolic blood pressure, and the second value (the value after the slash) indicates diastolic blood pressure.

The one or more measures for treating hypertension in a patient may include the administration of one or more drugs according to one or more dosage regimens, weight reduction, change of diet, alcohol reduction or abstinence, less smoking or stop smoking, salt reduction, and/or physical activity.

The one or more measures for treating hypertension in a patient are selected on the basis of initial patient data. In other words: initial patient data are the basis for selecting the one or more measures.

“Initial patient data” preferably means information about the patient's anatomical, physiological, physical and/or behavioral condition at the time the one or more measures are selected. The initial patient data are thus collected before the one or more measures are selected. In contrast, further patient data (the patient data on the basis of which compliance with and/or success of the one or more measures can be determined) are data collected after the one or more measures have been selected.

“Initial patient data” and “further patient data” are also referred to in this description, individually or jointly, as “patient data”. The term “patient data” can therefore mean “initial patient data or further patient data”, or “initial patient data and further patient data”, or “initial patient data and/or further patient data”.

Preferably, the measures are updated until the patient's blood pressure reaches a target range and/or the blood pressure remains stable within defined limits. Updates can be made based on at least a portion of the further patient data and optionally additionally based on at least a portion of the initial patient data.

Patient data can be provided by the patient or any other person such as a physician and/or a nurse . Patient data may be entered into one or more computer systems by said person or persons via input means (such as a keyboard, a touch-sensitive surface, a mouse, a microphone, and/or the like).

Patient data can be captured (e.g., automatically) by one or more sensors, e.g., blood pressure sensor, motion sensor, activity tracker, blood glucose meter, heart rate meter, thermometer, impedance sensor, microphone (e.g., for voice analysis) and/or others.

Patient data can be generated by a laboratory and stored in a data storage by laboratory personnel.

Patient data can be read from one or more data storages.

Patient data comprise data about a patient’s blood pressure.

"Blood pressure" is the force of circulating blood on blood vessels per unit area. Most of this pressure results from the heart pumping blood through the circulatory system. Blood pressure varies along the blood circulation and is highest in the aorta and continues to drop as blood travels through the circulatory system via arteries, capillaries, and veins. When used without qualification, the term "blood pressure" usually refers to the pressure in the large arteries (arterial blood pressure).

In addition, (arterial) blood pressure varies continuously over the cardiac cycle, but in clinical practice only systolic and diastolic pressures are routinely reported. The systolic pressure is the maximum pressure during one heartbeat and the diastolic pressure is the minimum pressure between two heartbeats in the cardiac cycle. Systolic and diastolic pressures are usually expressed in millimeters of mercury (mmHg) above ambient atmospheric pressure. For decades, systolic and diastolic pressures were measured almost exclusively in the brachial artery with blood pressure cuff (peripheral blood pressure). However, the shape of the pressure waveform changes continuously throughout the arterial tree. Although diastolic and mean arterial pressures are relatively constant, systolic pressure may be up to 40 mmHg higher in the brachial artery than in the aorta. This phenomenon of systolic pressure amplification arises principally because of an increase in arterial stiffness moving away from the heart: as the pressure wave travels from the highly elastic central arteries to the stiffer brachial artery, the upper portion of the wave becomes narrower, the systolic peak becomes more prominent, and systolic pressure increases.

Central blood pressure is the pressure in the aorta, which is the large artery into which the heart pumps and is considered a better indicator of the pressure the heart and other vital organs experience than the peripheral blood pressure. Furthermore, central blood pressure has been shown to be a better predictor of vascular disease when compared to peripheral blood pressure.

Mean arterial pressure is the average arterial pressure throughout one cardiac cycle, systole, and diastole.

It should be noted that the present disclosure is not limited to any particular definition or expression of blood pressure, nor is it limited to any particular method of measurement. It is only important that the measured blood pressure values allow a statement to be made about the patient's state of health and/or that the measured blood pressure values indicate the presence or absence of hypertension in a patient. The measured blood pressure should make it possible to determine whether one or more measures should be taken to lower the blood pressure so that the patient does not have to expect any adverse health consequences in the short, medium or long term as a result of the measured blood pressure.

Thus, when the present disclosure refers to blood pressure, it may mean (arterial) systolic and/or diastolic arterial blood pressure, peripheral blood pressure, central blood pressure, mean (arterial) blood pressure, blood pressure values over full cardiac cycles, and/or derivatives of the foregoing, e.g., integration of pressure over time, and/or others.

Furthermore, blood pressure is usually not a constant value over time but is subject to fluctuations that may be regular or irregular. Therefore, the term blood pressure in this disclosure can also be a mean value averaged over a defined period of time, e.g., over one hour or several hours or one day or several days. The mean value may be, for example, the arithmetic mean.

In addition, the term blood pressure should not be understood to mean that the measurand recorded is actually a pressure. The measurand may also be another parameter that correlates with blood pressure. For example, there are also optical methods for estimating blood pressure (see, e.g., P. Junyung et al.: Photoplethysmogram Analysis and Applications: An Integrative Review, Frontiers in Physiology, 12, 2022, DOI= 10.3389/fphys .2021.808451 ) .

Preferably, patient data comprise blood pressure values of the patient over a period of more than one day, preferably of at least one week.

Preferably, patient data include blood pressure values that reflect a time course of the blood pressure within a defined period of time, e.g., within one day or several days or one week or more than one week (time course data).

Such “time course data” are data that provide information about the development of a measurand (in this case blood pressure) over time. The time intervals between the points at which measurement points are captured are preferably selected in such a way that a measured value of the measurand captured at a point in time between two measuring points would not deviate from the measured values at the two measuring points to a greater extent than defined in advance. In other words: the points in time at which the measurand is recorded should be so close together that the measurand does not exhibit any unexpected peaks between two measurement points. In a preferred embodiment, the measured values are so close together that each blood pressure value between two immediately successive measured values would be not greater than the greater of the two measured values and not less than the smaller of the two measured values.

In a preferred embodiment, the rate at which the blood pressure is measured is greater than twice the maximum frequency of the blood pressure variations over time (in accordance with the Nyquist- Shannon sampling theorem). However, the term “rate” should not be understood to mean that the interval between two measurement points is constant. It is irrelevant whether the blood pressure is measured at constant or variable intervals. If the blood pressure is measured at variable intervals, the “sampling rate” preferably represents the largest distance between two consecutive measuring points.

In addition to blood pressure data, patient data may include other information about the patient's anatomical, physiological, physical and/or behavioral condition at the time the one or more measures are selected and/or updated.

Patient data may include, e.g.,: age, gender, body size, body weight, body mass index, ethnics, resting heart rate, heart rate variability, sugar concentration in urine, body temperature, impedance (e.g., thoracic impedance), lifestyle information about the life of the patient, such as consumption of alcohol, smoking, and/or exercise and/or the patient’s diet, medical intervention parameters such as regular medication, occasional medication, or other previous or current medical interventions and/or other information about the patient’s previous and/or current treatments and/or reported health conditions and/or combinations thereof.

Patient data may comprise information from an EMR (electronic medical record, also referred to as EHR (electronic health record)). The EMR may contain information about a hospital’s or physician's practice where certain treatments were performed and/or certain tests were performed, as well as various other (meta-)information about the patient's treatments, medications, tests, and physical and/or mental health records.

Patient data may comprise information about a person's condition obtained from the person himself/herself (self-assessment data, (electronic) patient reported outcome data (e)PRO)). Besides objectively acquired anatomical, physiological, physical and/or behavioral data, the well-being of the patient also plays an important role in the monitoring of health. Subjective feeling can also make a considerable contribution to the understanding of objectively acquired data and of the correlation between various data. If, for example, it is captured by sensors that a person has experienced a physical strain, for example because the respiratory rate and the heart rate have risen, this may be because just low levels of physical exertion in everyday life place a strain on the person; however, another possibility is that the person consciously and gladly brought about the situation of physical strain, for example as part of a sporting activity. A self-assessment can provide clarity here about the causes of physiological features.

Subjective feeling can be collected by use of a self-assessment unit, with the aid of which the patient can record information about subjective health status. Preference is given to a list of questions which are to be answered by a patient. Preferably, the questions are answered with the aid of a computer system (e.g., laptop computer, tablet computer and/or smartphone). One possibility is that the patient has questions displayed on a screen and/or read out via a speaker. One possibility is that the patient inputs information into a computer by, e.g., inputting text via an input device (e.g., keyboard, mouse, touchscreen and/or a microphone (by means of speech input)). It is conceivable that a chatbot is used in order to facilitate the input of all items of information for the patient. It is conceivable that the questions are recurring questions which are to be answered once or more than once a day or a week by a patient. It is conceivable that some of the questions are asked in response to a defined event. It is, for example, conceivable that it is captured by means of a sensor that a physiological parameter is outside a defined range (e.g., an increased respiratory rate is established and/or the blood pressure exceeds a pre-defined threshold). As a response to this event, the patient can, for example, receive a message via his/her smartphone or smartwatch or the like that a defined event has occurred and that said patient should please answer one or more questions, for example in order to find out the causes and/or the accompanying circumstances in relation to the event. Patient data provided orally by the patient via a microphone can also be analyzed. Voice analysis can be a powerful tool to determine the emotional state of patients.

Patient' s heart rate may correlate with emotional stress.

A sensor can measure catecholamine levels in the skin as a marker of stress (or at least sweat production, which increases after stress, see, e.g.: https://pubmed.ncbi.nlm.nih.gov/15191805/).

Once initial patient data have been collected for a patient suffering from hypertension, one or more measures for reducing the patient' s blood pressure are selected.

The one or more measures can be selected by a physician. The computer system according to the present disclosure may be configured to display a plurality of measures to a physician. The physician can then select one or more measures for the patient based on the initial patient data. The selection of the physician is received by the computer system.

The selection may also involve manual entry of one or more parameters by a physician (or another person, such as a physician’s assistant), such as the name of a drug and/or the specification of a dose and/or the specification of times at which one or more medications should be taken be a patient.

In an embodiment, the computer system is configured to identify one or more blood pressure reduction measures for the patient on the basis of at least a portion of the initial patient data.

The computer system may be configured, for example, as an expert system that identifies one or more measures based on criteria. Whether the criteria are fulfilled is checked on the basis of the (initial) patient data. This will be illustrated by means of the following example: It is conceivable, for example, that a first class of antihypertensive medication is recommended only for mildly elevated blood pressure, while a second class of medication is recommended for severely elevated blood pressure. The expert system compares the blood pressure of the initial patient data with blood pressure ranges for which the medications of the first and second classes are recommended and selects one or more drugs of the class in whose range the blood pressure of the initial patient data lies.

In an analogous manner, the computer system can be configured to determine an individual dosage regimen for a patient on the basis of the patient’s body weight and, optionally, other data such as body height, age, gender, current blood pressure, ethnics, other medications taken by the patient, and/or others.

Such a dosage regimen specifies when the patient should take what amounts of one or more drugs.

Expert systems are described, for example, in: WO2020/139771A1, US8301465, W02007/075451A2, WO20I9/099553AI.

It is also possible that, based on the initial patient data, one or more patients are identified who have a defined similarity to the patient under consideration. It is possible to identify those measures that have been performed on the similar patients and/or that have successfully led to a reduction in blood pressure in the similar patients. Those measures can then be outputted to a physician and the physician can select one or more measures that are appropriate for the patient under consideration.

For such a similarity search, a feature vector can be generated for the patient based on initial patient data. The distance of this feature vector to feature vectors of a number of other patients can be determined and those patients can be determined whose feature vector has the smallest distance to the feature vector of the patient under consideration and/or whose feature vector does not exceed a maximum distance to the feature vector of the patient under consideration.

A “feature vector” is an w-dimcnsional vector of numerical features that represent an object (in this case a patient), wherein n is an integer greater than 0. Many algorithms require a numerical representation of objects since such representations facilitate digital processing and statistical analysis. Feature vectors are equivalent to the vectors of explanatory variables used in statistical procedures such as linear regression. The vector space associated with these vectors is often called the feature space. In order to reduce the dimensionality of the feature space, a number of dimensionality reduction techniques can be employed. Examples for the generation of feature vectors can be found in, e.g., in J. Frochte: Maschinelles Lernen, 2. AufL, Hanser-Verlag 2019, ISBN: 978-3-446-45996-0.

The term “distance” refers to the distance of two feature vectors in the feature space. There are different ways to quantify the distance between two feature vectors, e.g., by calculating the following distances: Euclidean distance, Manhattan distance, Chebyshev distance, Minkowski distance, weighted Minkowski distance, Mahalanobis distance, Hamming distance, Canberra distance, Bray Curtis distance, or a combination thereof. Instead of a distance, another similarity value can also be calculated, which quantifies the similarity between two patients preferably on the basis of feature vectors.

It is also conceivable that a part of the measures is identified by the computer system and suggested to the physician and another part is determined by the physician. It is also conceivable that the physician makes one or more changes to one or more of the measures suggested by the computer system.

The one or more identified measures can be outputted to a physician, for example displayed on a monitor and/or printed on a printer. The identified measure(s) can also be stored in a data storage.

Preferably, the one or more measures comprise a dosing schedule for gradually increasing the dosage of one or more medications.

Such a dosing schedule may provide for a final dose to be reached in two or three or four or five or six or more increments.

The gradual increase of a dose has the advantage that a physician can check whether the identified medication is having the desired effect and/or whether undesirable side effects are occurring. The gradual increase enables a physician to take countermeasures at an early stage if undesirable side effects become apparent and/or the medication does not appear suitable for other reasons.

This step-by-step procedure is reminiscent of a chemical titration process and is often referred to as dose titration or drug titration.

For the one or more identified, determined, selected, and/or updated measures, a blood pressure is predicted. The predicted blood pressure should result if the patient complies with the respective measures. The predicted blood pressure is a value that likely will occur in the future if the patient follows the one or more measures.

The blood pressure is predicted for a defined point in time or for a defined period of time in the future, for example for a day in a week after the prediction or for a day in a month after the prediction.

Instead of one predicted blood pressure value, more than one predicted blood pressure value can be determined. For example, it is possible to predict a blood pressure range that can be defined, for example, by an upper limit and a lower limit. The expected blood pressure can then remain within the range, i.e., it usually does not exceed the upper limit and it usually does not fall below the lower limit.

Preferably, it is determined how patient’s blood pressure develops as a function of time if the patient complies with the one or more measures. In the event that a dosing schedule has been identified and/or specified for a patient, the changes in the dose over time can be considered and blood pressure values as a function of time can be predicted that result when the dosing schedule is adhered to.

Preferably, the time course of the blood pressure is predicted for a period of at least one week, more preferably for at least one month starting with the day of the prediction.

It should be noted that the predicted blood pressure is a probability value based on statistical analysis.

It is thus uncertain whether the respective predicted blood pressure will actually result. The predicted blood pressure only results with a defined probability, which can be determined (e.g., from the model used for prediction).

The predicted blood pressure can be outputted to a physician. Similarly, the probability of the predicted blood pressure value actually occurring can be outputted to the physician.

The predicted blood pressure is preferably determined using a trained machine learning model. The machine learning model may be trained to predict one or more blood pressure values based on the initial patient data and based on data about the one or more measures.

Such a “machine learning model”, as used herein, may be understood as a computer implemented data processing architecture. The machine learning model can receive input data (such as initial patient data and data about the one or more measures) and provide output data (such as a predicted blood pressure) based on that input data and the machine learning model, in particular parameters of the machine learning model. The machine learning model can learn a relation between input data and output data through training. In training, parameters of the machine learning model may be adjusted in order to provide a desired output for a given input.

The process of training a machine learning model involves providing a machine learning algorithm (that is the learning algorithm) with training data to learn from. The term “trained machine learning model” refers to the model artifact that is created by the training process. The training data must contain the correct answer, which is referred to as the target. The learning algorithm finds patterns in the training data that map input data to the target, and it outputs a trained machine learning model that captures these patterns.

The training data can comprise, for each reference patient of a multitude of reference patients, patient data, and data about one or more measure taken to lower blood pressure (input data) as well as one or more blood pressure values that have resulted from following the one or more measures (target data).

The term “multitude” as it is used herein means an integer greater than 1, usually greater than 10, preferably greater than 100.

In the training process, training data are inputted into the machine learning model and the machine learning model generates an output. The output is compared with the (known) target. Parameters of the machine learning model are modified in order to reduce the deviations between the output and the (known) target to a (defined) minimum.

In general, a loss function can be used fortraining to evaluate the machine learning model. For example, a loss function can include a metric of comparison of the output and the target. The loss function may be chosen in such a way that it rewards a wanted relation between output and target and/or penalizes an unwanted relation between an output and a target. Such a relation can be, e.g., a similarity, or a dissimilarity, or another relation.

A loss function can be used to calculate a loss value for a given pair of output and target. The aim of the training process can be to modify (adjust) parameters of the machine learning model in order to reduce the loss value to a (defined) minimum.

If, for example, the output and the target are numbers, the loss function can be the absolute value of the difference of the numbers. In this case, a high absolute value of the loss function can mean that a parameter of the model needs to undergo a strong change.

In the case of vector-valued outputs, for example, difference metrics between vectors such as the root mean square error, a cosine distance, a norm of the difference vector such as a Euclidean distance, a Chebyshev distance, an Lp-norm of a difference vector, a weighted norm or any other type of difference metric of two vectors can be chosen. These two vectors may for example be the desired output (target) and the actual output. In the case of higher dimensional outputs, such as two-dimensional, three-dimensional or higherdimensional outputs, for example an element-wise difference metric may be used. Alternatively or additionally, the output data may be transformed, for example to a one -dimensional vector, before computing a loss function.

Training data can be collected in clinical studies, for example.

Instead of the patient data, a feature vector, as described above, can also be used as input to a machine learning model. Usually, in a first step, a machine learning model generates a feature vector based on patient data, which can be a compressed representation of the patient data. In a second step, a regression analysis can be performed that establishes a mathematical relationship between the feature vector and one or more blood pressure values.

Once the machine learning model has been trained, it can be used to predict blood pressure values for new patients on the basis of patient data and data about one or more measures to be taken to lower blood pressure.

The computer system can be configured that for each measure identified, one or more predicted blood pressure values are determined and outputted. The computer system can be configured that for one or more identified measures selected/determined by a physician, one or more predicted blood pressure values are determined and outputted. The computer system can be configured to recalculate one or more predicted blood pressure values in the event that a physician changes one or more measures and/or parameters of one or more measures.

For example, the computer system can be configured to receive changes from, for example, a physician, wherein the changes represent a new measure and/or represent a changed measure . The computer system may be configured to predict and output one or more blood pressure values for each change based on the changed data.

In this way, a physician can determine how one or more measures will affect the patient's blood pressure. The physician can determine how changes in the measures affect the patient's blood pressure. The physician can choose which measure(s) to provide for the patient.

The computer system may be configured to receive feedback from the physician, the feedback indicating what measure(s) the physician intends for the patient to take. The respective measure(s) may then be stored in a data store and/or printed on a printer and/or transmitted to the patient and/or transmitted to a pharmacist for providing one or more medications according to the selected measure and/or used to write a prescription.

Of course, the physician may also communicate the prescribed measure(s) to the patient verbally in a consultation and/or over the phone.

After the physician has communicated the measure(s) to the patient, further patient data are collected to monitor compliance with the measure(s) and/or the success of the measure(s).

Further patient data comprise data about the patient’s current blood pressure. The current blood pressure is the blood pressure that is gradually established as a result of the full or partial implementation of the one or more measures. If the patient does not implement any measures, the current blood pressure can be expected to correspond to the initial blood pressure.

Preferably, the patient’s blood pressure (in particular the current blood pressure) is measured continuously (or quasi-continuously) using one or more sensors.

The term “continuous(ly)”, as it is used herein, means that blood pressure measurements are taken at defined intervals over a period of at least one day, even more preferably at least one week, even more preferably at least one month, most preferably about a period of more than one month. The term “continuous(ly)” does not necessarily mean that one measurement is taken immediately after the other. Instead, the blood pressure values are measured at defined time intervals, with these intervals being no greater than one day, preferably no greater than one hour.

Preferably, blood pressure (in particular the current blood pressure) is measured automatically, i.e., without the patient's intervention.

Blood pressure measurements can be regular or irregular, the intervals between two consecutive measurements can be constant or variable. Preferably, the intervals between two consecutive measurements are not greater than one hour, preferably not greater than 15 minutes, even more preferably not greater than 10 minutes.

Preferably, the rate at which blood pressure is measured, is specified individually for each patient. Preferably, the rate is determined (at least partially) based on patient's blood pressure variation over time. If a patient experiences greater variation, a higher rate may be set for that patient than for a patient who experiences less variation.

The rate can also change over the course of a day. For example, it is possible to measure the blood pressure less frequently during periods of time in which experience has shown that only minor fluctuations in blood pressure occur for many patients or for the patient under consideration than during periods of time in which the fluctuations are greater and/or in which greater deviations occur.

Preferably, the rate at which blood pressure measurements are taken is adjusted overtime to the degree of change of blood pressure. Preferably, the sensor (or a controller controlling the sensor) is configured to shorten the intervals between two successive measurements when the blood pressure increases and/or when the blood pressure increases more rapidly than a pre-defined threshold, and/or when the blood pressure exceeds a pre-defined threshold. If the blood pressure drops and/or the blood pressure falls below a defined value, the intervals between two successive measurements can be extended again.

Blood pressure can be measured with one or more sensors, which are preferably placed in or on the patient's body.

Examples of sensors worn by the patient on patient’s body that continuously measure blood pressure can be found, e.g., in US10568527B2, EP3609394A1, US10952623B2, and WO2019/172569A1.

Examples of sensors that continuously measure blood pressure in the patient's body can be found, e.g., in W02006/023786A2, EP2397185B1, US8602999B2, and W02004/014456A2.

Alternatively or additionally, contactless sensors can be used to measure or estimate blood pressure (see, e.g.: N. Sugita et al. '. “Contactless Technique for Measuring Blood-Pressure Variability from One Region in Video Plethysmography”, Journal of medical and Biological Engineering, 2019, 39, 76-85).

As described above, for blood pressure measurement all current or new methods can be considered. In other words: the present disclosure is not limited to any particular method of determining blood pressure or to any particular device or type of device for blood pressure measurement. It is also possible to use different methods and/or devices and/or types of devices.

The measured blood pressure values can be transmitted to a computer system used by a physician. The transmission can take place immediately after a new value has been recorded. However, the transmission can also take place at defined times (for example, once a day at 23:00) and/or at defined time intervals (for example, every full hour) and/or when defined events occur (for example, when there is a defined deviation of the blood pressure from one or more reference values). Furthermore, the patient and/or the physician and/or another/further person can initiate the transmission by a corresponding command.

The computer system is preferably configured to compare the measured blood pressure values with one or more predicted blood pressure values and/or with predefined reference values. The computer system is preferably configured to output a message to the physician about the deviation in the event of a defined deviation of the measured blood pressure values from the one or more predicted blood pressure values and/or predefined reference values.

For example, the physician can be informed if the measured blood pressure does not fall as predicted or even rises.

In this way, the physician can determine at an early stage whether the recommended measures are having the desired effect or whether one or more changes are necessary to combat the patient's high blood pressure.

However, the physician can also be informed if the measured blood pressure has reached a target area and a change in the measure(s) may be necessary. In the case of a gradually increasing dosage, for example, the achievement of a first target area can be taken as an opportunity to increase the dosage in order to reach a second target area. This can be repeated until the final target area is reached.

In a preferred embodiment of the present disclosure, in addition to the patient’s blood pressure, the intake of one or more medications for the treatment of hypertension is monitored.

The term “intake” should not be construed restrictively, such that it only means an oral administration of a drug. Rather, the term “intake” covers every conceivable form of application, such as aural, buccal, by inhalation, intra-arterial, intra-articular, intragluteal, intradermal, intramuscular, intraocular, intrauterine, intravenous, intravitreal, intranasal, subcutaneous, rectal, sublingual, subcutaneous, topical, transdermal, vaginal and the like. Preferably, the medication is taken by the patient him/herself; i.e., a medical specialist is not required to administer the medication to the patient. The medication is preferably one that the patient takes orally.

The medication whose intake is being monitored need not be exclusively a drug that directly or indirectly affects the patient's blood pressure. It may also be a drug that is intended by the physician to suppress and/or alleviate concomitant symptoms of hypertension and/or to combat side effects of other medications and/or to improve the patient's overall health.

The term “monitoring the intake” means that it is monitored whether the patient has at least made preparations to take a portion of medication, and/or whether he/she actually has taken a portion of the medication.

The monitoring of the intake is carried out in order to increase the adherence. “Adherence” refers to the extent to which the behavior of the patient in relation to the taking of a medication conforms to the recommendations agreed with the treating physician.

In one embodiment, the patient confirms that he/she has taken one or more doses of one or more medications by entering the respective information into a computer system (e.g., a smartphone or atablet computer or a laptop used by the patient) . If the intake is not consistent with the dosing regimen provided by the physician, the patient may indicate a time when the intake occurred.

In another embodiment, the intake of one or more medications is monitored automatically by means of a monitoring unit.

The term “automatic monitoring” means that the patient is supported in the repeated administration of a drug by a technical monitoring unit for treatment with medication. The monitoring unit may be an integral part of the system according to the present disclosure. The monitoring unit is preferably a portable device that a patient can carry with him/her (such as the patient’s smartphone). The term “monitoring unit” is understood to mean any electronic system that detects an attempt to take and/or the actual taking of a portion of medication by a person.

The term “attempt to take” is understood to mean measures taken by a person to prepare to take a portion of a medicine. A typical example is the removal of a portion of medicine from its packaging, e.g., the removal of a tablet from a blister pack.

It is conceivable that preparatory measures for taking a portion of medicine are not taken by the patient himself/herself but, for example, by a physician or a nurse or by a relative. For the purposes of the present disclosure, it is irrelevant whether the preparatory action is carried out by the patient himself or herself or by another person; the present disclosure is intended to capture all of these possibilities. For the sake of a simpler presentation the present disclosure will be described primarily based on the first option (that the patient takes the preparatory measures), without intending to restrict the disclosure to this option.

In one embodiment of the present disclosure the monitoring unit registers whether and when the patient has made preparations for removing a portion from a medication storage device for storing the medication and/or whether and when the patient has removed a portion of medication from the medication storage device. It is conceivable that before he/she can remove a portion of medication from the storage device, the patient must indicate, for example by pressing a button or by presenting a biometric characteristic (for example, the finger to record a fingerprint in the context of a fingerprint recognition) that he/she would like to remove a dose of medication. In such a case the monitoring unit registers the action of the patient, which is designed to result in a removal of a dose of medication. It is also conceivable, however, that the monitoring unit registers the actual removal of a dose of medication. For example, it is conceivable that by squeezing a portion of a medication out of a blister pack an electrically conductive conductor track is interrupted, for example; this interruption can be detected by an electronic circuit, for example (see, e.g., WO9604881A1 or DE19516076A1).

In another embodiment, the monitoring unit is designed in such a way that it registers the actual taking of a portion of medication by the patient. As an example, US 10929983 discloses a system that tracks the actual taking of a portion of medication by a smartphone app using the smartphone's camera. Image analysis and image recognition algorithms ensure that the portion of medication and the face of the patient are detected. It is also detected that the patient puts the medication dose into his/her mouth and swallows.

In a preferred embodiment, the monitoring unit is designed in such a way that it reminds the patient (or else the nursing staff or another person) that they are due to take a portion of medication; for example acoustically (e.g. by means of a signal tone or a voice message), visually (e.g. by a flashing light or a text message) and/or by tactile means (for example by means of vibration).

Information about patient adherence can be transmitted to the physician. The transmission can take place immediately after a dose has been taken. However, the transmission can also take place at defined times (for example, once a day at 23:00) and/or at defined time intervals (for example, every full hour) and/or when defined events occur (for example, when a time limit for taking a medication has passed). Furthermore, the patient and/or the physician and/or another/further person can initiate the transmission by a corresponding command.

Preferably, the physician is informed if a patient does not take the recommended dose on time and/or at all one or more times.

In addition to the blood pressure values and optionally the information on the patient's adherence, additional further patient data (as described above) can be collected and transmitted, e.g., to the physician’s computer system.

The further patient data collected and transmitted can be used to determine whether the one or more identified measures are having their desired effect. A patient’s blood pressure that does not drop as predicted is an indication that the one or more measures are not having the desired effect.

The further patient data can be used to identify one or more causes for the absence of the desired effect.

For example, the further patient data can be used to determine whether the patient has complied with the one or more measures.

If non-compliance with the one or more measures can be ruled out as the cause, it is conceivable that one or more measures will have to be changed to achieve the desired effect.

But it is also conceivable that a sub-goal has been achieved and one or more measures need to be changed to achieve the next sub-goal. For example, it may be that an initially comparatively low dosage of a drug shows the expected effect, but in a further step must be increased to finally bring the blood pressure into the (final) target range.

At least a portion of the further patient data can also be used to provide an update on the one or more measure(s). Optionally, the update can additionally be performed on at least part of the initial patient data, in particular those patient data that typically do not change much, if at all, over time, such as gender, ethnics, height, age, and/or others; or those patient data collected during the collection of initial patient data but not during the collection of further patient data.

"Update" means to identify one or more new and/or modified measures based, at least partially on at least a portion of the further patient data.

The one or more updated measures can be outputted to a physician. For the one or more updated measures, one or more blood pressure values can be predicted to occur if the patient complies with the updated measure(s). The trained machine learning model as described above can again be used for prediction.

After the physician has communicated the updated measure(s) to the patient, further patient data may again be collected to monitor the success of the measure(s) and/or compliance with the measure(s) as described.

One or more updates of the measures may occur. Preferably, the measures are updated until the patient's blood pressure reaches a target range and/or the blood pressure remains stable within defined limits.

The invention is explained in more detail below with reference to the drawings, without limiting the invention to the features and combinations of features shown in the drawings.

Fig. 1 shows schematically by way of example, the process of identifying measures, monitoring measures, and updating measures for the treatment of hypertension in a patient.

The process is described using a timeline (t = time).

In a period of time prior to the time point tl, initial patient data PDI are collected that provide information about the health status of a patient P. The initial patient data PDI comprise information about the (initial) blood pressure BPI of the patient. At the time point tl, one or more measures for reducing the blood pressure of the patient P are identified. The identified measures comprise the intake of a first medication Ml according to a first dosage regimen DR1. The identified measures are selected and/or acknowledged by a physician D and communicated to the patient P.

In a time period between time point tl and time point t2, further patient data PDF1 are collected. The further patient data PDF 1 comprise information about the blood pressure BP 1 in the time period between time points tl and t2, and information about whether the patient P complies with the one or more measures communicated by the physician D. At time point t2, the one or more measures are updated. This means that one or more measures for reducing the blood pressure of the patient P are identified, but now based, at least partially on at least a portion of the further patient data PDF1, in particular at least partially on the basis of the blood pressure BP1. It is conceivable that some or all of the initial patient data PDI are included in the identification of updated measures. This is illustrated by the dashed arrow. The updated measures comprise the intake of a second medication M2 according to a second dosage regimen DR2. The second medication M2 can be the same medication as the first medication Ml, but it can also be a different medication or an additional medication. Similarly, the second dosage regimen DR2 can be the same dosage regimen as the first dosage regimen DR1, or it can be a different dosage regimen. The updated medication can be selected and/or acknowledged by the physician D and communicated to the patient P.

In a time period between time points t2 and time t3, further patient data PDF2 are collected. The further patient data PDF2 comprise information about the blood pressure BP2 in the time period between time points t2 and t3, and information about whether the patient P complies with the one or more measures communicated by the physician D. At time point t3, the one or more measures are updated a second time. This means that one or more measures for reducing the blood pressure of the patient P are identified, but now at least partially based on at least a portion of the further patient data PDF2. It is conceivable that some or all of the initial patient data PDI and/or of the further patient data PDF 1 are included in the identification of the renewed update of the measures. This is illustrated by the dashed arrows. The newly updated measures comprise the intake of a third medication M3 according to a third dosage regimen DR3. The third medication M3 can be the same medication as the second medication M2, but it can also be a different medication or an additional medication. Similarly, the third dosage regimen DR3 can be the same dosage regimen as the second dosage regimen DR2, or it can be a different dosage regimen. The newly updated medication can be selected and/or acknowledged by the physician D and communicated to the patient P.

After time point t3, further patient data PDF3 may be collected in order to monitor patient’ s health status . The further patient data PDF3 may comprise information about the blood pressure BP3 after time point t3.

Fig. 2 shows schematically by way of example, the identification of one or more measures for reducing a patient’s blood pressure on the basis of patient data. Patient data PD are inputted into a computer system CS. The computer system CS is configured to determine, on the basis of the patient data, one or more measures MS for reducing patient’s blood pressure. The one or more measures MS may comprise the intake of a medication M by the patient according to a dosage regimen DR.

Fig. 3 shows schematically by way of example, the identification of one or more measures on the basis of a similarity search. Patient data PD are inputted into a computer system which comprises a feature generation unit FGU. The feature generation unit FGU is configured to generate a feature vector FV on the basis of the patient data PD. The feature vector FV is a compressed representation of the patient data PD. On the basis of the feature vector FV, a similarity retrieval SR is performed. From a data storage in which a plurality m of reference feature vectors RFVm are stored, one or more reference feature vectors RFVi having a defined similarity to the feature vector FV are identified (m is an integer preferably greater than 100). For the identified reference feature vectors RFVi, those measures MS are identified that resulted in a reduction in blood pressure in the corresponding reference patients.

Fig. 4 shows schematically by way of example, the gradual increase of the dose of a medication in order to bring a patient’s blood pressure into a target region (drug dose titration process). Fig. 4 shows the time course (t = time) of a patient's blood pressure BP.

In the period between time point tO and time point tl, the mean blood pressure is BP1. The mean blood pressure BP1 or the time course data in the period between time point tO and time point tl , and optionally further initial patient data can be used to identify one or more measures for reducing the blood pressure BP1. The one or more measures comprise the intake of a first medication Ml by the patient according to a first dosage regimen DR1. The one or more measures are intended to bring the blood pressure into a first target range TRI. The first target range TRI is defined by an upper limit and a lower limit. The first target range TRI and/or the mean blood pressure BP2 can be a predicted blood pressure.

The patient follows the one or more measures in the time period between time points tl and t2. As a result, the mean blood pressure drops to blood pressure BP2. At the time t2 the one or more measures are updated. The updated measures comprise the intake of a second medication M2 by the patient in accordance with a second dosage regimen DR2. The updated measures are intended to bring the blood pressure into a second target range TR2. The second target range TR2 is defined by an upper limit and a lower limit. The second target range TR2 and/or the mean blood pressure BP3 can be a predicted blood pressure. The second medication MR2 can be the same medication as the first medication MR1 or it can be different. The second dosage regimen DR2 can be the same dosage regimen as the first dosage regimen or it can be different.

At time point t3 the mean blood pressure has dropped to the blood pressure BP 3, and the measured blood pressure values remain within the target range TR2. At time point t3, a final update of the one or more measures occurs. The final update of the one or more measures comprises the intake of a third medication M3 by the patient according to a third dosage regimen DR3. The third medication MR3 can be the same medication as the second medication MR2 or it can be different. The third dosage regimen DR3 can be the same dosage regimen as the second dosage regimen DR2 or it can be different.

Fig. 5 shows schematically by way of example the process of training a machine learning model. The machine learning model MLM is trained on the basis of training data. The training data comprise a multitude of data sets, each data set comprising input data and target data. In the example shown in Fig. 5, only one training data set TD comprising input data ID and target data T is shown. The input data ID is inputted into the machine learning model MLM. The machine learning model is configured to generate, at least partially on the basis of the input data ID and model parameters MP, an output O. The output O is compared with the target T. This is done by using a loss function LF, the loss function quantifying the deviations between the output O and the target T. For each pair of an output O and the respective target T, a loss value is computed. During training the model parameters are modified in a way that reduces the loss values to a defined minimum. The aim of the training is to let the machine learning model generate for each input data an output which comes as close to the corresponding target as possible. Once the defined minimum is reached, the (now fully trained) machine learning model can be used to predict an output for new input data (input data which have not been used during training and for which the target is usually not (yet) known). Such a machine learning model can be trained to predict blood pressure on the basis of patient data.

Fig. 6 shows schematically by way of example how a trained machine learning model can be used for making predictions. The trained machine learning model (MLM^T) can be the machine learning model described with reference to Fig. 5. New input data ID* are inputted into the trained machine learning model (MLM^T). The trained machine learning model (MLM^T) is configured and trained to generate, at least partially on the basis of the new input data ID* and the model parameters MD, an output O* . Such a trained machine learning model can used to one or more predict blood pressure values on the basis of patient data.

Fig. 7 shows schematically by way of example a preferred embodiment of a system according to the present disclosure.

The system comprises a first computer system CS1 and a second computer system CS2. The first computer system CS1 is usually used by a physician, the second computer system CS2 is usually used by a patient suffering from hypertension. The first computer system CS1 and the second computer system CS2 can be connected via one or more networks, such as a mobile phone network and/or the internet. The first computer system CS1 may be designed as a tablet computer, a laptop computer, a desktop computer, or a computer tower. The second computer system CS2 may be designed as a smartwatch, a mobile phone (smartphone), a tablet computer, a desktop computer, or a computer tower.

The first computer system CS1 can comprise a trained machine learning model MLM^T. The trained machine learning model MLM^T can be configured to predict one or more blood pressure values on the basis of patient data.

The first computer system CS1 can comprise a measure identification unit MIU. The measure identification unit MIU can be configured to identify one or more measures for reducing blood pressure on the basis of patient data.

The first computer system CS1 can comprise a feature generation unit FGU. The feature generation unit FGU can be configured to generate a compressed representation of a patient in the form of a feature vector on the basis of patient data. Such a feature vector can be used to identify similar patients and/or measures that have helped lower blood pressure in similar patients in the past. Such feature vector can be used as an input to the trained machine learning model MLM^T and/or to the measure identification unit MIU.

The first computer system CS1 can be connected to a data storage DB, e.g., via cable and/or wirelessly via one or more network connections. The data storage can also be an integral part of the first computer system CS1. In the data storage DB, one or more trained machine learning model MUM^T, reference patient data (such as feature vectors of reference patients), a feature generation unit, a medication identification unit and/or the like can be stored.

The second computer system CS2 is connected to a blood pressure sensor BPS. The blood pressure sensor BPS can be connected to the second computer system CS2 via cable or via a short-range communications technology such as NFC, RFID, Bluetooth, Bluetooth UE, ZigBee, infrared and/or the like. The blood pressure sensor BPS can also be part of the second computer system CS2. The blood pressure sensor BPS is configured to (preferably continuously) measure the blood pressure of a patient (and optionally additional patient data). The measured blood pressure data (and optionally additional patient data) are transmitted to the second computer system CS2.

The second computer system CS2 can be connected to one or more further sensors FS via cable or via a short-range communications technology such as NFC, RFID, Bluetooth, Bluetooth EE, ZigBee, infrared and/or the like. The one or more further sensors FS can also be part of the second computer system CS2. The one or more further sensors FS are configured to (preferably continuously) measure further parameters related to a patient's anatomical, physiological, physical and/or behavioral condition. The one or more further sensors FS can comprise a motion sensor, an activity tracker, a blood glucose meter, a heart rate meter, a thermometer, an impedance sensor, a microphone (e.g., for voice analysis) and/or others.

The second computer system CS2 can be connected to medication intake monitoring unit MU via cable or via a short-range communications technology such as NFC, RFID, Bluetooth, Bluetooth LE, ZigBee, infrared and/or the like. The medication intake monitoring unit MU can also be part of the second computer system CS2. The medication intake monitoring unit MU is configured to detect an attempt to take and/or the actual taking of a portion of medication by a person.

The second computer system CS2 is configured to collect patient data and transmit patient data to the first computer system CS 1. The first computer system CS 1 is configured to receive patient data, identify, on the basis of patient data, one or more measures for reducing a patient' s blood pressure, and determine predicted blood pressure values.

Fig. 8 shows schematically by way of example one embodiment of a computer system. Generally, a computer system of exemplary implementations of the present disclosure may be referred to as a computer and may comprise, include, or be embodied in one or more fixed or portable electronic devices.

The computer system CS comprises an input unit (2), a processing unit (3), a memory (4), an output unit (5), and communication interface(s) (6).

Data (e.g., patient data, and or control commands) can be entered into the computer system CS via the input unit (2). The input unit (2) can act as an interface to a user by accepting commands to control the computer system CS. User input interface(s) may be wired or wireless, and may be configured to receive information from a user into the computer system CS, such as for processing, storage and/or display. Suitable examples of user input interfaces include a microphone, image or video capture device (e.g., a camera), keyboard or keypad, joystick, touch-sensitive surface (separate from or integrated into a touchscreen) and/or the like. In some examples, the user interfaces may include automatic identification and data capture (AIDC) technology for machine -readable information. This may include barcode, radio frequency identification (RFID), magnetic stripes, optical character recognition (OCR), integrated circuit card (ICC), and the like. The user interfaces may further include one or more interfaces for communicating with peripherals such as sensors, printers, and/or the like.

The processing unit (3) may be composed of one or more processors alone or in combination with one or more memories. The processing unit (3) is generally any piece of computer hardware that is capable of processing information such as, for example, data, computer programs and/or other suitable electronic information. The processing unit (3) is composed of a collection of electronic circuits some of which may be packaged as an integrated circuit or multiple interconnected integrated circuits (an integrated circuit at times more commonly referred to as a “chip”). The processing unit (3) may be configured to execute computer programs, which may be stored onboard the processing unit (3) or otherwise stored in the memory (4). The processing unit (3) may be a number of processors, a multi -core processor or some other type of processor, depending on the particular implementation. For example, it may be a central processing unit (CPU), a field programmable gate array (FPGA), a graphics processing unit (GPU) and/or a tensor processing unit (TPU). Further, the processing unit (3) may be implemented using a number of heterogeneous processor systems in which a main processor is present with one or more secondary processors on a single chip. As another illustrative example, the processing unit (3) may be a symmetric multi -processor system containing multiple processors of the same type. In yet another example, the processing unit (3) may be embodied as or otherwise include one or more ASICs, FPGAs or the like. Thus, although the processing unit (3) may be capable of executing a computer program to perform one or more functions, the processing unit (3) of various examples may be capable of performing one or more functions without the aid of a computer program. In either instance, the processing unit (3) may be appropriately programmed to perform functions or operations according to example implementations of the present disclosure.

The memory (4) is generally any piece of computer hardware that is capable of storing information such as, for example, data, computer programs (computer-readable program code) and/or other suitable information either on a temporary basis and/or a permanent basis. The memory (4) may include volatile and/or non-volatile memory and may be fixed or removable. Examples of suitable memory include random access memory (RAM), read-only memory (ROM), a hard drive, a flash memory, a thumb drive, a removable computer diskette, an optical disk, a magnetic tape, or some combination of the above. Optical disks may include compact disk - read only memory (CD-ROM), compact disk - read/write (CD-R/W), DVD, Blu-ray disk or the like. In various instances, the memory may be referred to as a computer-readable storage medium. The computer-readable storage medium is a non-transitory device capable of storing information and is distinguishable from computer-readable transmission media such as electronic transitory signals capable of carrying information from one location to another. Computer- readable medium as described herein may generally refer to a computer-readable storage medium or computer-readable transmission medium. Information and/or data can be output from the computer system CS via the output unit (5). The output unit (5) may be or comprise one or more component(s) that provide (s) output information from the computer system CS, e.g., a display, a speaker, one or more light-emitting diodes (LEDs), a printer, a vibration unit (e.g., forthe generation of vibration alerts) and/or the like. The display may be configured to present or otherwise display information to a user, suitable examples of which include a liquid crystal display (LCD), light-emitting diode display (LED), plasma display panel (PDP) or the like.

The computer system CS further comprises one or more interfaces for transmitting and/or receiving information from or to other devices. Such communications interface(s) (6) may be configured to transmit and/or receive information, such as to and/or from one or more sensors, other computer(s), network(s), database(s), medication dispensing devices, or the like. The communications interface(s) (6) may be configured to transmit and/or receive information by physical (wired) and/or wireless communications links. The communications interface(s) (6) may include interface(s) to connect to a network, such as using technologies such as cellular telephone, Wi-Fi, satellite, cable, digital subscriber line (DSL), fiber optics and the like. In some examples, the communications interface(s) may include one or more short-range communications interfaces configured to connect devices using short-range communications technologies such as NFC, RFID, Bluetooth, Bluetooth LE, ZigBee, infrared (e.g., IrDA) or the like.

As indicated above, program code instructions may be stored in memory (4), and executed by processing unit (3) that is thereby programmed, to implement functions of the computer system CS, subsystems, tools and their respective elements described herein. As will be appreciated, any suitable program code instructions may be loaded onto a computer system or other programmable apparatus from a computer- readable storage medium to produce a particular machine, such that the particular machine becomes a means for implementing the functions specified herein. These program code instructions may also be stored in a computer-readable storage medium that can direct a computer, processing unit or other programmable apparatus to function in a particular manner to thereby generate a particular machine or particular article of manufacture. The instructions stored in the computer-readable storage medium may produce an article of manufacture, where the article of manufacture becomes a means for implementing functions described herein. The program code instructions may be retrieved from a computer-readable storage medium and loaded into a computer, processing unit or other programmable apparatus to configure the computer, processing unit or other programmable apparatus to execute operations to be performed on or by the computer, processing unit or other programmable apparatus.

Retrieval, loading and execution of the program code instructions may be performed sequentially such that one instruction is retrieved, loaded and executed at a time. In some example implementations, retrieval, loading and/or execution may be performed in parallel such that multiple instructions are retrieved, loaded, and/or executed together. Execution of the program code instructions may produce a computer-implemented process such that the instructions executed by the computer, processing circuitry or other programmable apparatus provide operations for implementing functions described herein.

Claims

1. A computer-implemented method comprising: receiving initial patient data, the initial patient data comprising data about a patient’s blood pressure, identifying, on the basis of the initial patient data, one or more measures for reducing the patient' s blood pressure, and/or receiving one or more measures for reducing the patient' s blood pressure, determining a predicted blood pressure that applies in the event that the patient complies with the one or more measures, outputting the predicted blood pressure, receiving further patient data, wherein the further patient data comprise a current blood pressure and information about whether the patient complies with the one or more measures, updating the one or more measures and/or the predicted blood pressure one or more times until the patient’s blood pressure reaches a target range, outputting the updated measures and/or the updated predicted blood pressure.

2. Method according to claim 1, wherein the one or more measures comprise (s) the administration of one or more medications according to one or more dosage regimens.

3. Method according to claim 1 or 2, wherein the further patient data comprise time course data of the patient' s blood pressure.

4. Method according to any one of claims 1 to 3, wherein the one or more measures are identified by a computer system on the basis of at least a portion of the initial patient data.

5. Method according to any one of claims 1 to 4, wherein identifying one or more measures for reducing the patient' s blood pressure comprises: generating a feature vector on the basis of at least a portion of the initial patient data, identifying one or more other feature vectors of other patients, the one or more other feature vectors having a defined similarity to the feature vector, identifying one or more measures that led to a reduction in blood pressure in the one or more other patients, selecting one or more of the measures that led to the reduction in blood pressure in the one or more other patients.

6. Method according to any one of claims 1 to 5, wherein the one or more measures comprise a dosing schedule for gradually increasing the dosage of one or more medications.

7. Method according to any one of claims 1 to 6, wherein the predicted blood pressure is determined by using a trained machine learning model.

8. Method according to any one of claims 1 to 7, further comprising: comparing the current blood pressure with the predicted blood pressure, notifying a physician in case the current blood pressure deviates in a pre-defined way from the previous blood pressure.

9. A computer system comprising: a processor; and a memory storing an application program configured to perform, when executed by the processor, an operation, the operation comprising: receiving initial patient data, the initial patient data comprising data about a patient’s blood pressure, identifying, on the basis of the initial patient data, one or more measures for reducing the patient' s blood pressure, and/or receiving one or more measures for reducing the patient' s blood pressure, determining a predicted blood pressure that applies in the event that the patient complies with the one or more measures, outputting the predicted blood pressure, receiving further patient data, wherein the further patient data comprises a current blood pressure and information about whether the patient complies with the one or more measures, updating the one or more measures and/or the predicted blood pressure one or more times until the patient’s blood pressure reaches a target range, outputting the updated measures and/or the updated predicted blood pressure.

10. A non-transitory computer readable medium having stored thereon software instructions that, when executed by a processor of a computer system, cause the computer system to execute the following steps: receiving initial patient data, the initial patient data comprising data about a patient’s blood pressure, identifying, on the basis of the initial patient data, one or more measures for reducing the patient' s blood pressure, and/or receiving one or more measures for reducing the patient' s blood pressure, determining a predicted blood pressure that applies in the event that the patient complies with the one or more measures, outputting the predicted blood pressure, receiving further patient data, wherein the further patient data comprises a current blood pressure and information about whether the patient complies with the one or more measures, updating the one or more measures and/or the predicted blood pressure one or more times until the patient’s blood pressure reaches a target range, outputting the updated measures and/or the updated predicted blood pressure.