WO2021127566A1 - Devices and methods for measuring physiological parameters - Google Patents

Abstract

Described herein are devices, systems, and methods for monitoring a physical parameter of an individual. A system as described herein may comprise a server and a patient monitoring device. The patient monitoring device may comprise a sensor comprised to sense a patient parameter, a memory, a processor and a non-transitory computer readable medium. The non-transitory computer readable medium may be encoded with instructions configured to cause the processor to receive said patient parameter, analyze the patient parameter by applying a rule to said patient parameter thereby generating an analysis result, and perform, based on said analysis result, one or more of transmitting an alarm signal, transmitting said patient parameter to said server, and storing said parameter in said memory.

Description

DEVICES AND METHODS FOR MEASURING PHYSIOLOGICAL PARAMETERS

CROSS-REFERENCE

[0001] This application claims the benefit of U.S. Provisional Application No. 62/951,919 filed on December 20, 2019, which is herein incorporated by reference in its entirety.

BACKGROUND

[0002] Constant monitoring of patients is a challenge in home environment due to lack of access to medical personnel as well as specialized medical devices accessible in hospital settings. On the other hand, alarm fatigue and false alarms in professional healthcare systems, can reduce the efficiency of patient care significantly.

SUMMARY

[0003] Various embodiments of the present disclosure address the demand for accessible, easy to use, noninvasive monitoring devices and systems that are capable of monitoring physiological parameters as well as accurate patient alarms in case of adverse events or other situations of interest in patient care.

According to some aspects of the disclosure, a computer-based patient monitoring system is provided. The computer-based patient monitoring system may comprise a server and a patient monitoring device. The patient monitoring device may comprise a sensor comprised to sense a patient parameter, a memory, a processor and a non-transitory computer readable medium. The non-transitory computer readable medium may be encoded with instructions configured to cause the processor to receive said patient parameter, analyze the patient parameter by applying a rule to said patient parameter thereby generating an analysis result, and perform, based on said analysis result, one or more of transmitting an alarm signal, transmitting said patient parameter to said server, and storing said parameter in said memory.

[0004] In some embodiments, the server is a cloud based server. In other instances, the server may be a component of a hospital network. The server may also be a component of the patient monitoring device.

[0005] In some embodiments, the sensor comprises a vital sign sensor. The sensor may be configured for continuous monitoring.

[0006] In some embodiments, the patient monitoring device is configured to communicate with the server wirelessly using a WiFi signal or an RF signal. In some instances, the patient monitoring device communicates with the server wirelessly using a cellular network.

In some embodiments, the patient monitoring device is configured to perform edge computing. In this case the rule may be a dynamically changing rule that determines at least one performance characteristic of said patient monitoring device. The rule may be generated by a machine learning algorithm trained with patient data. In some embodiments, the rule identifies a critically abnormal patient parameter. The rule may be updated with patient parameter values for a patient that is being monitored with the patient monitoring device. In some embodiments, the rule comprises a threshold value for a specific patient parameter.

[0007] In some embodiments, the performance characteristic comprises performance efficiency. In some instances, the performance characteristic comprises power utilization optimization. In other instances, performance characteristic may comprise a true alarm rate and the true alarm may comprise an alarm signal that corresponds to a critically abnormal patient parameter. The alarm signal may be transmitted to a mobile computing device of a healthcare provider.

[0008] Also disclosed herein is a computer-based portable monitoring system. The monitoring system may comprise a server and a wearable monitoring device. The monitoring device may comprise a first and a second ECG electrodes configured to sense a lead I ECG tracing. The first electrode may be positioned on the wearable monitoring device so that the first electrode contacts a first upper extremity of a user when the wearable monitoring device is worn by said user. The monitoring device may further comprise a PPG sensor configured to sense a PPG, a memory, a processor and a non-transitory computer readable medium. The non-transitory computer readable medium may be encoded with instructions configured to cause said processor to receive the PPG, identify a signal within the PPG and analyze the signal thereby generating an analysis result. The processor may further perform, based on analysis result, one or more of transmitting an indication to the user to contact said second ECG electrode with a second upper extremity thereby generating said lead I ECG tracing, transmitting said lead I ECG tracing to said server, transmitting an alarm signal, and storing said ECG tracing in said memory. The server may be a component of a network of a patient care provider or patient care provider facility. In some cases, the PPG sensor is configured for continuous monitoring.

[0009] In some embodiments, the wearable monitoring device comprises a vital sign sensor. The wearable monitoring device may comprise a lab-on-a-chip device. The wearable monitoring device may further be configured to communicate with said server wirelessly using a WiFi signal or an RF signal. The communication with server may be configured through a cellular network. The wearable monitoring device may further be configured perform edge computing in which case the analysis result may be generated using a machine learning algorithm.

[0010] In some embodiments, the machine learning algorithm may be trained with patient data. The training may be done using the data from memory. The machine learning algorithm may be configured to determine power utilization optimization of said wearable monitoring device. The machine learning algorithm may further be configured to optimize a rate of true indication comprising a correspondence of said indication with a true presence of an arrhythmia.

[0011] In some embodiments, the analysis result comprises a critically abnormal patient parameter. The analysis result may be based on a threshold value of PPG signal.

[0012] In some embodiments, the alarm signal is transmitted to a mobile computing device of a health care provider.

[0013] Also disclosed herein is a system configured to monitor and individual. The system may comprise a plurality of sensors, a processor and a non-transitory computer readable medium.

The non-transitory computer readable medium may be encoded with software configured to cause the processor to receive a first signal from a first sensor of the plurality of sensors at a first time and a second signal from a second sensor of the plurality of sensors at a second time and determine if the individual has experienced a critical event based on the first signal and the second signal. In some embodiments, the first sensor comprises a heart rate sensor and second sensor comprises a different sensor. At least one of the said sensors may comprise an implantable sensor or a wearable sensor. The first and the second time may be the same time.

The critical event may comprise a myocardial infarction or arrythmia.

[0014] Also disclosed herein is a non-transitory computer readable medium encoded with software configured to cause said processor to carry out a monitoring process. The monitoring process may comprise receiving a patient parameter from a sensor, analyzing the patient parameter by applying a rule to the patient parameter thereby generating an analysis result and performing, based on the analysis result, one or more of transmitting an alarm signal, transmitting patient parameter to a server, and storing the parameter in local memory. The server may be cloud based server. The server may also be a component of a hospital network. The sensor may comprise a vital sign sensor.

[0015] Also disclosed herein is a system configured to monitor an individual. The system may comprise a plurality of sensors including a PPG sensor and a blood glucose monitoring sensor, a processor and a non-transitory computer readable medium. The non-transitory computer readable medium may be encoded with software configured to cause the processor to receive a first signal from the PPG sensor at a first time and a second signal from the blood glucose monitoring sensor at a second time and determine if the individual has experienced a critical event based on the first signal and the second signal. The first and the second time may be the same time. The critical event may comprise a hypoglycemia or hyperglycemia event. INCORPORATION BY REFERENCE

[0016] All publications, patents, and patent applications mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated to be incorporated by reference.

BRIEF DESCRIPTION OF THE DRAWINGS [0017] The novel features of the invention are set forth with particularity in the appended claims. A better understanding of the features and advantages of the present invention will be obtained by reference to the following detailed description that sets forth illustrative embodiments, in which the principles of the invention are utilized, and the accompanying drawings of which:

[0018] FIG. 1A shows an example of a patient monitoring system in a hospital setting in accordance with some embodiments of the disclosure.

[0019] FIG. IB and FIG. 1C show examples of sub-system devices in accordance with some embodiments of the disclosure.

[0020] FIG. 2 shows an example of a patient monitoring system in a home setting in accordance with some embodiments of the disclosure.

[0021] FIG. 3 shows an exemplary embodiment of a method for real-time measurement of PPG signals and analyzing the results.

[0022] FIG. 4 shows an exemplary embodiment of a patient monitoring system collecting, storing and analyzing the collected data.

[0023] FIG. 5 is a perspective view of a block diagram of the artificial intelligence architecture according to some embodiments.

[0024] FIG. 6 shows an exemplary embodiment of a method for real-time measurement of PPG signals along with another sensor and analyzing the results.

[0025] FIG. 7 shows a computer system that is programmed or otherwise configured to implement methods provided herein.

[0026] FIG. 8 schematically illustrates data analytics platform architecture in accordance with some embodiments.

DETAILED DESCRIPTION

[0027] While various embodiments of the invention have been shown and described herein, it will be obvious to those skilled in the art that such embodiments are provided by way of example only. Numerous variations, changes, and substitutions may occur to those skilled in the art without departing from the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed.

[0028] Whenever the term “at least,” “greater than,” or “greater than or equal to” precedes the first numerical value in a series of two or more numerical values, the term “at least,” “greater than” or “greater than or equal to” applies to each of the numerical values in that series of numerical values. For example, greater than or equal to 1, 2, or 3 is equivalent to greater than or equal to 1, greater than or equal to 2, or greater than or equal to 3.

[0029] Whenever the term “no more than,” “less than,” or “less than or equal to” precedes the first numerical value in a series of two or more numerical values, the term “no more than,” “less than,” or “less than or equal to” applies to each of the numerical values in that series of numerical values. For example, less than or equal to 3, 2, or 1 is equivalent to less than or equal to 3, less than or equal to 2, or less than or equal to 1.

[0030] Described herein are devices, systems, and methods for sensing a physical parameter of an individual. Systems described herein may be capable of estimating the likelihood of adverse events or deterioration of conditions for patients. If the likelihood is above a predefined value, preventative actions may be taken by healthcare professionals and the patient. This may be beneficial to patient’s health as well as cost saving for patient and the healthcare system. As a nonlimiting example, cardiac patients may have high rate of re-hospitalization due to reoccurrence of conditions such as atrial fibrillation. With a monitoring system that continuously monitors parameters such as PPG, the re-hospitalization and the burden in terms of costs and adverse events may be reduced substantially. The devices and systems described herein may further be used in a hospital setting as well as for post-operative patients in remote settings such as home or home care settings. Physiological parameters pertaining to different disorders such as heart conditions, diabetes or pulmonary problems may be monitored simultaneously.

[0031] In some embodiments, a physical parameter of an individual comprises an ECG sensed from the individual. In some embodiments, a physical parameter comprises a heart rate of an individual. In some embodiments, a physical parameter comprises a blood pressure of an individual. Non-limiting examples of physical parameters sensed by embodiments of the devices, systems, and methods described herein include temperature, SP02, perspiration, EEG, and urine output. One having skill in the art will understand that many other physical parameters may be sensed with an appropriate sensor type that is integrated with or into the devices, systems, and methods described herein.

[0032] The embodiments of the present disclosure may enable the continuous measurement and analysis of physiological parameters such as but not limited to temperature, PPG, SP02, heart rate (HR), heart rate variability (HRY), etc. The systems and devices provided herein may further allow for receiving signals from other patient monitoring devices or systems such as continuous glucose monitoring (CGM) systems and may analyze the received signals.

[0033] In some embodiments, a group of physiological parameters may be measured and correlated or compared against each other, since the change of a single parameter, may not be sufficient to predict some adverse events. The sudden change of a physiological parameter may be indicative of a normal change in the body such as the sudden change of blood glucose due to consumption of food. The embodiments of present disclosure provide methods for analysis of a plurality of physiological paraments in real time to inform the patient or the health care provider if further steps of monitoring or any actions may be needed.

[0034] Machine learning algorithms can be trained with various datasets of physiological signals to allow for prediction of adverse events. The machine learning algorithms can further allow for personalized data training based on each individual’s data including the data from various patient monitoring devices, personal information, behavioral data and electronic health records data. The algorithms used herein, may utilize data from different resources such as sensor data or electronic health records (EHR) data. Data available from different resources can include vital signals, physiological data, chemical data such as laboratory results, image data from different imaging modalities, EHR annotations and exam notes from previous clinical visits, behavioral data, environmental data such as temperature or humidity at certain times, etc.

[0035] In some embodiments, certain events may be of interest to be monitored by the devices and systems describes herein. Non-limiting examples of the events are peaks in the ECG data, lasting of a feature longer than a predefined duration of time, patterns in data such as beat to beat variation, etc. In some embodiments, a set of parameters, events or features in the parameters from the plurality of parameters, events or features may be selected as tokens to be fed to the machine learning algorithms as inputs.

[0036] The output of the machine learning algorithms may be related to detecting an adverse event and raising an alarm such as code blue or sepsis or the general deterioration of a patient’s conditions. Token selection may be done by a health care professional based on condition of the patient or the parameters of interest to be monitored or the available data wherein features can be extracted and monitored. The output of the machine learning algorithm may include the likelihood of an adverse event or deterioration of a patient’s health condition. If the likelihood is above a predefined threshold, the patient or the healthcare professional may be prompted to take actions to prevent adverse events. This may be substantially beneficial to the patient and cost saving for patient and the healthcare system.

[0037] In some embodiments, a computer-based patient monitoring system is disclosed. The patient monitoring system (hereinafter referred to as “system” for simplicity) may operate under various operational modes. The modes of operation may depend on factors such as the environment the monitoring system will be operating in. Nonlimiting examples of operation modes may include hospital mode, transition mode and home mode. In some embodiments, there may be no distinguishable operation mode available for the monitoring system.

[0038] FIG. 1A shows an example of a monitoring system 100 as described herein, in a hospital setting.

[0039] The monitoring system 100 may comprise a server. As shown in the example of FIG. 1, the server may be part of a hospital network 110. The server may further be a cloud based server. In some embodiments, the server may be a part of a local network. The server may be located in a patient monitoring device or be a part of an edge processing hardware. In some embodiments, the system is configured to upload to and/or download data from the server. In some embodiments, the server is configured to store sensor data, and/or other information for the individual. In some embodiments, the server is configured to store historical data (e.g., past sensor data).

[0040] The monitoring system may further comprise a patient monitoring device. Nonlimiting examples of patient monitoring devices may be portable, wearable, ingestible or implantable monitoring devices. In some embodiments, the patient monitoring device may be wearable such as a wristband, watch, neckwear, anklet, headgear, patch, etc. In some embodiments, the patient monitoring device may comprise a housing configured to be worn by an individual such as a smartwatch band or a wristlet.

[0041] The patient monitoring device may comprise one or a plurality of sensors. In some embodiments, at least one of the plurality of sensors may measures a vital sign or a signal related to a vital sign such as but not limited to PPG, blood pressure, heart rate, heart rate variability, stroke volume, respiration, air flow volume, SP02, perspiration, galvanic skin response (GSR), audible body sounds, etc. In some embodiments, the patient monitoring device may further comprise a lab-on-a-chip (LOC) device for measuring variety of physiological parameters in a single device within the patient monitoring device. The patient monitoring device may also be able to measure non-biological parameters such as location, acceleration and ambient pressure.

[0042] The patient monitoring device may comprise a processor for controlling the operation of the device as well as executing instructions. The monitoring device may further comprise a memory sufficiently large to store some or all of the measurement data by the sensors. The patient monitoring device may be configured to communicate with the server 110 wirelessly using a WiFi signal, RF signal or a cellular network. Wearable wristwatch 120 and wearable patch 130 are shown in the example of FIG. 1A (as well as in FIG. IB and FIG. 1C respectively). In some embodiments, the wearable patient monitoring devices may comprise a transmitter to transmit data to and from other patient monitoring devices such as but not limited to 12 lead ECG monitoring device, Holter monitor or continuous glucose monitoring (CGM) device. The transmitter may use Bluetooth, Bluetooth Low Energy chip, WiFi or cellular data connection.

[0043] In some embodiments, the sensors may make continuous measurements. The measurements may be stored in the local memory of the monitoring device and/or transmitted to a sever such as the hospital server of FIG. 1A. In other cases, the measurements may be discrete such as periodic or nonperiodic or based on the health condition of the individual. The measurement signal (patient parameter such as PPG signal) may be analyzed by the processor of the monitoring device or at the server to determine any abnormalities by for example comparing various sections of the signal to various thresholds. If the results of the analysis indicate abnormalities the patient may be notified by for example a visual or audio or motion alarm on the patient monitoring device to take further steps such as starting ECG measurement. If the analysis of measurement signal is done locally on the monitoring device, the results may be transferred to the server for further training of an algorithm or for record update. In some cases, the analysis result may identify a critically abnormal patient parameter leading to an adverse event such as a myocardial infarction. In a hospital setting an alarm signal may be transmitted to the healthcare provider such as nursing station. In some cases, based on the severity of the abnormality of the patient parameter a code blue may be generated containing the location of the patient as well as other related data.

[0044] As described elsewhere, the patient monitoring system may be operated under different operational modes. FIG. 2 shows an example of a home operation mode. In this operation mode, the measurement signal data may be transferred to a remote server such as a cloud-based server. Various monitoring devices such as blood pressure and heart rate monitor, blood oxygen monitor, etc. may send their data to a patient monitoring device such as the watch 120. The data may then be transferred to the server wirelessly or trough cellular connection. In some embodiments, the data from various monitoring devices may be transmitted to the server directly through a wired or a wireless connection. In some embodiments, at least a portion of data processing and analysis such as training the algorithms may be done on the patient monitoring device using edge processing technologies. Another portion of data processing and analysis may be done on the server. The results may be transmitted to the patient monitoring device or be stored on the server database. If the patient monitoring device detects critically abnormal patient parameter, the patient may be notified by a visual or audio or motion alarm on a user interface of the patient monitoring system. In some embodiments, an alarm may be transmitted a smart device of the patient and/or authorized individuals and/or the healthcare provider of the patient. In some cases, the patient monitoring device and/or the patient’s smart device may automatically be connected to emergency services. In some embodiments, the patient may be prompted by for example a text message on the monitoring device or the smart device to contact the emergency services.

[0045] The patient monitoring system may also operate in a transition mode. This transition may be a transition from hospital to other care facility or home or may be in the outpatient section of a hospital. In some embodiments, cellular network may be used for data transfer to and from a remote server such as a cloud server.

[0046] In some embodiments, the patient monitoring device (e.g. smart watch) may continuously measure PPG signal. The PPG signal may be analyzed on the server using techniques such as machine learning, fuzzy logic techniques, Markov models, state machines, Bayesian logic techniques, evolutionary or genetic algorithms or a combination thereof. In some embodiments, the PPG signal may be analyzed locally using the computational resources of the patient monitoring device such as edge processing resources. In some embodiments, training of the algorithms may also be performed locally by the processors of the patient monitoring device. In some cases, at least a portion of the training may be done locally, and another portion of the training may be done on a server such as remote server or a cloud-based server.

[0047] FIG. 3 shows an exemplary embodiment of a method 300 for a physical parameter measurement such as a PPG signal. In a step 302 the patient receives a patient monitoring device comprising a sensor such as but not limited to PPG sensor, processor, a receiver and a transmitter. In some embodiments, the device may comprise electrodes for measuring ECG signals. In a step 304 patient monitoring device may measure signals such as PPG continuously. In a step 306 the PPG signal is analyzed using a machine learning algorithm. The analysis result may be based on a threshold value of the PPG signal. If the algorithm detects an abnormality in the PPG signal or predicts a likelihood higher than a threshold for symptoms such as atrial fibrillation or arrhythmia (step 308) the patient monitoring device may alert the patient or the healthcare provides.

[0048] The alert may be in the form of a visual or audio message on a user interface of the patient monitoring device. In some embodiments, the alert may be displayed in the form of a visual or audio representation on a personal smart device of the patient through a linked mobile application to the patient monitoring device. Nonlimiting examples of alerts are vibration of the device, audio chimps and flashing screen of the user device. In some embodiments, the wearable patient monitoring device (e.g. wristlet or smart watch) may comprise a fist and a second ECG electrode configured to sense a lead I ECG tracing, wherein the first electrode 121 is positioned on wearable monitoring device so that first electrode contacts a first upper extremity of the patient such as the wrist, when worn by the patient. The alarm may indicate to the patient to put a second body part such as a finger of the other hand on the second ECG electrode 122, so that a lead I ECG measurement can be done. The ECG data may be stored on the local memory of the wearable patient monitoring device and/or transferred to a server. In a hospital operation mode, the nursing staff or the health care provider may be notified for example through the user interface of a smart device that the patient has done an ECG measurement and/or ECG data has been uploaded to a hospital server.

[0049] In some embodiments, the wearable monitoring device may comprise two wristlets, worn on both hands of the patient. The first electrode 121 on each wristlet contacting the wrist of the patient may be used to make a lead I ECG measurement. In this case, the ECG measurement may start automatically after the alert on the device for the abnormality of PPG is received. [0050] In some embodiments, the alarm may automatically trigger other monitoring devices such as the wearable patch 130 to make ECG measurements. In some embodiments, at least 30 seconds of ECG data may be stored on the local device memory or be transferred to the server. The server may be cloud based server or a component of a patient care provider or a patient care provider facility. The wearable monitoring device may further comprise other vital sign sensors. The wearable monitoring device may communicate with server wirelessly using a WiFi signal or an RF signal or using a cellular network.

[0051] In various operation settings of the systems described herein, the patient may be alarmed or alerted at different levels. Some events or adverse events may only raise to a message or warning level of alert. In such instances, the user may receive a message or instructions to take an action. In some instances, factors such as improper positioning may cause weak signal in one or a plurality of sensors. In this case the patient may receive messages on the monitoring device or a smart device linked to the monitoring device to change the settings of the sensor such as location of the sensor. In some instances, an alarm may indicate an adverse event that requires the attention of the healthcare professional but may not raise to a critical level. In other instances, an alarm may raise to a critical level, hereinafter referred to a as “critical alarm”. [0052] Critical alarms may be predictive measures that raise the awareness of the healthcare providers and patients prior to an adverse event. Critical alarms may also be interpretable in that the message indicating the critical alarm may also provide some information as for the reason of the alarm so that the health care providers may act more efficiently by knowing how to respond the alarm. Depending on the condition of the patient and the mode of operation of the devices and systems described herein, critical alarms may indicate different conditions. For example, in an intensive care unit (ICU) setting in a hospital the critical alarm may be indicative of code blue, sepsis or flat line. In a less critical ambulatory environment such as a home or an assisted living facility, the monitoring may be adjusted to patient deterioration measures. These deterioration measure may include activity level, behavioral health, vitals (blood pressure, spirometry, SP02, heart rate, respiration rate, etc.), electrolyte and chemical changes and environmental conditions that are indicative of worsening patient conditions. The adjustment of monitoring may be of especial importance for patients in transition from a hospital setting to an ambulatory setting for example in the case of early discharge of patients from hospital.

[0053] Large volume of data such as electronic health records (EHR), various sensors data, behavioral data, imaging data, etc. may be available either offline or being gathered in real time. It may be crucial to select the suitable data based on patient’s condition and the operation setting such as critical hospital setting or ambulatory setting. The selected data ma then be monitored and used by the artificial intelligence algorithms to be able to detect warning signs or the critical alarms with high accuracy.

[0054] FIG. 8 shows an embodiment of the data analytics platform as described herein. The raw data may be collected by sensors or other mean as described above. In some embodiments, the data may be labeled for use in the machine learning algorithms and stored in databases. Prior to labeling or during the labeling, the data can be grouped in various categories. Nonlimiting examples of the groups of data are disease groups, sensor type groups and image groups. The data can also be categorized into subsets of groups belonging to each group. In some embodiments, the labeling of data may be done manually for example by healthcare professionals. In some embodiments, at least a portion of labeling may be done manually and another portion of labeling may be done using artificial intelligence techniques such as machine learning or deep learning methods. In some cases, labeling may be done using machine learning or deep learning methods. Data may further be annotated by healthcare professionals.

[0055] Data may be stored in one or a plurality of databases to be used for training the machine learning algorithms or to be used as input to the machine learning algorithms. Databases can be update and new data can be added to the dataset continuously or in predefined time intervals. In some embodiments, clinicians and other professionals may select the suitable data to be used for training of the algorithms and/or input to the machine learning algorithms for monitoring. In other instances, data selection may also be automated using artificial intelligence techniques. [0056] Algorithm pipelines may be used to handle large volume of raw and annotated data. An algorithm repository may also include all the algorithms used in the systems described herein. Nonlimiting examples of algorithms in the repository are the PPG tracking algorithm for detection of atrial fibrillation or other disorders and the algorithms for detecting patient’s deterioration score based on predefined condition values. More than one machine learning algorithm or deep learning algorithm at any given time may be used to classify the data and predict the state of the patient. Other algorithms may also be used for various operations in the system.

[0057] The algorithms may be accessed via well-defined application programming interfaces (APIs) or well-defined applications such as mobile apps. In some embodiments, the access to some or all of the APIs from some of the monitoring devices or equipment may be free of charge. In some cases, there may be a charge such as a one time fee or a subscription fee for accessing some or all of the APIs from some of the monitoring devices or equipment. Non limiting examples of APIs include API for analysis of PPG signal to predict the likelihood of atrial fibrillation (AF), API for analysis of PPG signal to alarm in the event of AF, API for analysis of PPG signal predict or alarm for other disorders such as myocardial infarction (MI), APIs for various critical alarms and APIs for analysis of ECG to predict any abnormalities such as abnormalities in electrolytes. The algorithm repository may be stored and updated on local servers or remote servers such as cloud-based servers. The algorithms may be called by any monitoring device through a remote procedure call, which will allow for transferring/downloading of any updates to the algorithms on the monitoring devices. The downloading of updates or other related algorithms may be done on predetermined intervals or when the updates are available. The utilization of the APIs makes the system scalable and device agnostic.

[0058] The algorithm hosting may be beneficial for stand alone monitoring devices to make them compatible with the devices and systems described herein.

[0059] FIG. 4 illustrates an embodiment of the patient monitoring system wherein the system may consist of one or more data acquisition and storage devices 410 for wired or wireless data transfer 420, including but not limited to Bluetooth, Bluetooth Low Energy (BTLE), cellular and/or wireless local area networking. The data acquisition device may be a wearable patient monitoring device such as a wristband. The data storage device 410 can include but is not limited to metadata. Data can be stored in one or a plurality of databases. The databases can be in local storage system or on remote servers such as cloud servers or on local servers such as hospital server system. The one or more databases 440 may utilize any suitable database techniques. For instance, structured query language (SQL) or “NoSQL” database may be utilized for signal data, raw collected data, training datasets, trained model (e.g., hyper parameters), weighting coefficients, etc. Some of the databases may be implemented using various standard data-structures, such as an array, hash, (linked) list, struct, structured text file (e.g., XML), table, JSON, NOSQL and/or the like. Such data-structures may be stored in memory and/or in (structured) files. In another alternative, an object-oriented database may be used.

[0060] The network 430 may establish connections among the components in the monitoring system and a connection to external systems. The network 430 may comprise any combination of local area and/or wide area networks using wireless and/or wired communication systems or cellular narrow band networks. For example, the network 430 may include the Internet, local area network (LAN), as well as mobile telephone networks. In one embodiment, the network 430 may use standard communications technologies and/or protocols. The data exchanged over the network can be represented using technologies and/or formats including data in binary form (e.g., Portable Networks Graphics (PNG)), the hypertext markup language (HTML), the extensible markup language (XML), etc. In addition, all or some of links can be encrypted using conventional encryption technologies such as secure sockets layers (SSL), transport layer security (TLS), Internet Protocol security (IPsec), etc. In another embodiment, the entities on the network can use custom and/or dedicated data communications technologies instead of, or in addition to, the ones described above.

[0061] Various data operations and manipulations may be applied at the signal processing unit 450 including but not limited to data filtering, denoising, DC filtering and data compression. Other mathematical functions may be applied to filtered data to obtain physiological parameters. Physiological parameters may be displayed in different forms such as tables or simple text or graphs on a graphical user interface (GUI). The user may be able to input data on the GUI such as personal information including age and gender. The GUI may be part of the patient monitoring device or on a separate device such as a smart device.

[0062] In some embodiments, the processor is configured to analyze the signals measured either by the patient monitoring device such as PPG signals or the signals received from other devices, using a machine learning algorithm. The machine learning algorithm may detect an abnormality in a signal or predict the occurrence of an adverse event such as a myocardial infarction from subtle changes in the data or from a trend of change in the data. The monitoring system may utilize machine learning algorithms for analysis of the data such as extracting certain features from an R-R interval data in a PPG signal or an inter-beat interval (IB I).

[0063] The example system of FIG. 4 may include a machine learning block 470. The machine learning block 470 may include a training module. The training module may be configured to obtain and manage training datasets. The training datasets may include historical data or human data from pre-clinical studies or simulated data. The training datasets may include data such as but not limited to physiological signals (parameters) as well as personal attributes such as age, height, weight, gender, skin melanin content, body mass index (BMI) or medical history. The training module may be configured to train a deep learning network for risk assessment and/or prediction. The training module may comprise a supervised or unsupervised learning method such as, for example, SVM, random forests, clustering algorithms, gradient boosting, logistic regression, or decision trees. The machine learning block may comprise a neural network comprising a convolutional neural network (CNN), perceptron neural network, perceptron residual neural network, multilayer perceptron neural network, recurrent neural network (RNN), long short-term memory RNN (LSTM RNN), dilated CNN, fully connected neural networks, deep-learning self-attention networks, deep generative models or deep restricted Boltzmann machines.

[0064] The training module may train a model off-line. Alternatively or additionally, the training module may use real-time data as feedback to refine the model for improvement or continual training.

[0065] A population-based prediction machine learning algorithm may be developed based on the aggregated data. The prediction algorithm may then be tailored towards individuals by updating database with individual’s data. The algorithm may then be trained based on each individual’s data and may be updated and improved as more real time data is added to the database.

FIG. 5 shows an example of artificial intelligence (AI) architecture. The data such as PPG time series received (measured) by the sensor may be denoised at a denoiser module. The denoising process may improve the signal to noise ratio of the raw PPG signal X(t). At each time step t, denoised data may be used as input and processed at an artificial neural network such as the multilayer perceptron classifier 500, as shown in FIG. 5.

[0066] Various data features may be calculated from the input data including but not limited to heart rate, respiration rate, blood pressure, heart rate variation, SP02, inter-beat interval (IB I), signals within R-R interval, P-wave features, QRS-wave features and amplitude of PPG signal. Data features may be extracted from PPG signal data. The extracted features may be processed by an activation neuron (not shown in the figure) through computing an activation function and a threshold. The processed data may further be fed to a prediction model, where the patient specific information such as age, gender, skin tone, etc. may be incorporated to the model to fine tune the performance of the model. The prediction model may be able to predict the probability or likelihood of an abnormality in the heart rhythm or an adverse event such as myocardial infarction. It will be understood that multilayer perceptron model 500 is provided as an illustrative but not limiting example. Any other suitable processing technique such as but not limited to fuzzy logic techniques, Markov models, state machines, Bayesian logic techniques, evolutionary or genetic algorithms or a combination thereof may be used. [0067] A single neural network architecture may be used to train the data and be used for detecting critical alarms and adverse events. Alternatively or in addition, a plurality of neural networks may be used to train and analyze features of large datasets that may have complex features. In some embodiments, a combination of the neural networks and other methods such as fuzzy logic techniques, Markov models, state machines, Bayesian logic techniques, evolutionary or genetic algorithms may be used for feature engineering and analysis. As described elsewhere, the system of present disclosure presents a continuous measurement of signals such as PPG. Atrial fibrillation (AF) burden is the percentage of time the patient’s heart rhythm is in AF. Continuous measurement of PPG signal may allow for not only predicting or identifying the presence or absence of AF but also for measuring the AF burden. AF burden can be used in quantification of likelihood of risk for heart attacks. The data on AF burden for each patient can be fed back and used in training of the neural network to improve the predication ability of the network for each individual, as well as overall improvement of neural network prediction capability.

[0068] The initial training of the artificial intelligence network may be done offline on a cloud based or remote or a hospital server to reduce the computation burden of the patient monitoring device and improve the power efficiency of the device. Once the neural network is trained, the rest of computations and data analysis may be done on the local processor of the patient monitoring device such as the wristlet or the wearable patch or the smart device. The patient monitoring device may be configured to perform edge computing using for example an edge tensor processing unit (TPU). All or parts of the data processing for the neural network may be done on locally on the wearable monitoring device. Edge processing may increase the power efficiency of the system and the device by reducing the energy needed to transmit data to a remote or local server. Edge processing may also increase the speed of data processing but accessing the data and other resources locally.

[0069] The machine learning algorithm may apply a plurality of rules for analyzing the received signals. The rules may be dynamically changing based on the development and improvement of the training of the algorithm with new data. In some embodiments, the dynamic rules may determine performance characteristics of the patient monitoring device. The performance characteristic may comprise performance efficiency for example the efficiency of the machine learning algorithm to identify the presence or absence of AF or the efficiency of the machine learning algorithm to predict the onset of a disorder or a syndrome based on continuous measurement of a physiological signal. The rule may define a threshold for the efficiency of the algorithm in identifying the presence or absence of a syndrome such as atrial fibrillation. [0070] In some embodiments, the performance characteristic may comprise power utilization optimization. The edge computing and local storage of the data may optimize the power management of the device. The improvement in power utilization may be quantified and calculated and optimized as a performance characteristic.

[0071] Further training of the machine learning algorithm based on the signal data and behavioral data of each patient may improve the performance of the algorithm. The improved algorithm may lead to a higher rate of true alarm. The true alarm may comprise an alarm signal that corresponds to a critically abnormal patient parameter such as in the case of the event of myocardial infarction. With the improvement of the algorithm the rules may also be updated since the performance may enhance. For example, the percentage threshold for of the rate of true alarm may increase from an initial value to a larger value since the true alarm rate may increase. In other word, the rule may be updated with patient parameter values for a patient that is being monitored with the patient monitoring device.

[0072] The machine learning algorithm may also be configured to optimize a rate of true indication comprising a correspondence of said indication with a true presence of an arrhythmia. [0073] Another aspect of the disclosure provides an individual monitoring system that comprises a plurality of sensors. All or a subset of sensors may measure signals continuously or non-continuously. The processor may receive a first signal from a first sensor of the plurality of sensors at a first time and a second signal from a second sensor of the plurality of sensors at a second time and determine if the individual has experienced a critical event based on the first signal and the second signal. The first time and the second time may be the same. In some embodiments, the first sensor may be a heart rate sensor the second sensor may be a different sensor. In some embodiments, at least one of the sensors may be wearable sensor or an implantable sensor. The critical event may be a myocardial infraction or an arrythmia.

[0074] FIG. 6 shows an example of a method for measuring more than one signal. In a step 610 the first signal may be received by the processor. The first signal may be measured continuously. The processor may be programmed to send command for periodic or nonperiodic measurements for the signal. In some embodiments, the measurements may be initiate by the individual using the patient monitoring device. The second signal from a second sensor 610a may be received continuously or at discrete times. All or a subset of measurement data may be transmitted to a server such as a remote server or a hospital server to be added to a training dataset. The updated data related to machine learning algorithm may be transmitted back to the local processor of the patient monitoring device to enable edge computing. In a step 630 the data from both sensors may be analyzed. If the data from both sensors shows normal state, in a step 640 the data may be time stamped and/or annotated and saved on the local memory or transmitted to a server. The data may be used for further personalized or general training of the machine learning algorithm. An abnormality may be detected in the data of one or both of the sensors. If the abnormality points to a heart related adverse event, a mandatory ECG recording may be initiated at a step 650. Other actions may further be taken at this step such as notifying the healthcare provider team or sending alarm to a patient’s smart device or sending an alarm to the smart device of authorized individuals. If the detected abnormalities do not rise to the level of critical alarm, the user may be prompted to initiate an ECG recording using the wearable monitoring device in a step 660. The ECG data may be stored on a local memory or transmitted to a server. In a hospital operation mode, the healthcare staff may be notified when an ECG data is uploaded on the server for further observation.

[0075] In some embodiments, the first sensor may be a PPG sensor and the second sensor may be blood glucose monitoring sensor such as a continuous glucose monitoring (CGM) sensor or a blood glucose monitoring (BGM) which can be used to monitor blood glucose as needed. Some sudden changes in the blood glucose may be due to normal changes in the body such as intake of food. Continuous measurement of PPG signals and training a machine learning algorithm may allow for distinguishing between normal and abnormal changes in the blood glucose more accurately. If abnormalities are detected in the PPG and blood glucose data, the patient may be notified through the user interface of the patient monitoring device or a smart device to take suitable action such as intake of food or medication or to initiate another blood glucose measurement.

[0076] In some embodiments, the algorithms may predict a likelihood of an abnormal change in the blood glucose. The likelihood may be larger than a predetermined threshold. In order to prevent the abnormal change or an adverse event, the user may be prompted to take suitable action by for an example an audio or visual or motion indication on the monitoring device or a smart device.

[0077] If the abnormality raises to the level of an adverse event such as acute hyperglycemia or acute hypoglycemia the patient as well as the healthcare providers may be alerted. In the cases leading to shock or seizure a code blue may initiated on the healthcare provider devices in a hospital operation mode or connection to emergency services may be initiated automatically in a home operation mode.

[0078] In some embodiments, a galvanic skin response (GSR) sensor may be integrated into the patient monitoring device or may be used separately as a part of the patient monitoring system. Variation in skin conductance through activation of sweat glands can be a measure of emotional and/or sympathetic responses. Using GSR data may add the sympathetic response in case of any abnormality detected in PPG data or ECG data or any other measurement. The added emotional and sympathetic response can improve the training of the artificial intelligence module and increase the accuracy of prediction of adverse events or abnormalities.

[0079] The present disclosure provides computer systems that are programmed to implement methods of the disclosure. FIG. 7 shows a computer system 701 that is programmed or otherwise configured to analyze the signals from one or a plurality of sensors such as a PPG sensor to determine or predict an abnormality in the signals. The computer system 701 can be configured to analyze data and use machine learning algorithms to assess the risk level for developing or progression of various disorders such as arrythmia and atrial fibrillation as described elsewhere herein. The computer system 701 can be an electronic device of a user or a computer system that is remotely located with respect to the electronic device. The electronic device can be a mobile electronic device or a patient monitoring device.

[0080] The computer system 701includes a central processing unit (CPU, also “processor” and “computer processor” herein) 705, which can be a single core or multi core processor, or a plurality of processors for parallel processing. The computer system 701 also includes memory or memory location 710 (e.g., random-access memory, read-only memory, flash memory), electronic storage unit 715 (e.g., hard disk), communication interface 720 (e.g., network adapter) for communicating with one or more other systems, and peripheral devices 725, such as cache, other memory, data storage and/or electronic display adapters. The memory 710, storage unit 715, interface 720 and peripheral devices 725 are in communication with the CPU 705 through a communication bus (solid lines), such as a motherboard. The storage unit 715 can be a data storage unit (or data repository) for storing data. The computer system 701 can be operatively coupled to a computer network (“network”) 730 with the aid of the communication interface 720. The network 730 can be the Internet, an internet and/or extranet, or an intranet and/or extranet that is in communication with the Internet. The network 730 in some cases is a telecommunication and/or data network. The network 730 can include one or more computer servers, which can enable distributed computing, such as cloud computing. The network 730, in some cases with the aid of the computer system 701, can implement a peer-to-peer network, which may enable devices coupled to the computer system 701 to behave as a client or a server. [0081] The CPU 705 can execute a sequence of machine-readable instructions, which can be embodied in a program or software. The instructions may be stored in a memory location, such as the memory 710. The instructions can be directed to the CPU 705, which can subsequently pro-gram or otherwise configure the CPU 705 to implement methods of the present disclosure. Examples of operations performed by the CPU 505 can include fetch, decode, execute, and writeback. [0082] The CPU 705 can be part of a circuit, such as an integrated circuit. One or more other components of the system 701 can be included in the circuit. In some cases, the circuit is an application specific integrated circuit (ASIC).

[0083] The storage unit 715 can store files, such as drivers, libraries and saved programs. The storage unit 715 can store user data, e.g., user preferences and user programs. The computer system 701 in some cases can include one or more additional data storage units that are external to the computer system 701, such as located on a remote server that is in communication with the computer system 701 through an intranet or the Internet.

[0084] The computer system 701 can communicate with one or more remote computer systems through the network 730. For instance, the computer system 701 can communicate with a remote computer system of a user (e.g., send information such as measurement data or other data). Examples of remote computer systems include personal computers (e.g., portable PC), slate or tablet PC’s (e.g., Apple® iPad, Samsung® Galaxy Tab), telephones, Smart phones (e.g., Apple® iPhone, Android-enabled device, Blackberry®), or personal digital assistants. The user can access the computer system 701 via the network 730.

[0085] Methods as described herein can be implemented by way of machine (e.g., computer processor) executable code stored on an electronic storage location of the computer system 701, such as, for example, on the memory 710 or electronic storage unit 715. The machine executable or machine readable code can be provided in the form of software. During use, the code can be executed by the processor 705. In some cases, the code can be retrieved from the storage unit 715 and stored on the memory 710 for ready access by the processor 705. In some situations, the electronic storage unit 715 can be precluded, and machine-executable instructions are stored on memory 710.

[0086] The code can be pre-compiled and configured for use with a machine having a processer adapted to execute the code, or can be compiled during runtime. The code can be supplied in a programming language that can be selected to enable the code to execute in a pre-compiled or as-compiled fashion.

[0087] Aspects of the systems and methods provided herein, such as the computer system 701, can be embodied in programming. Various aspects of the technology may be thought of as “products” or “articles of manufacture” typically in the form of machine (or processor) executable code and/or associated data that is carried on or embodied in a type of machine readable medium. Machine-executable code can be stored on an electronic storage unit, such as memory (e.g., read-only memory, random-access memory, flash memory) or a hard disk. “Storage” type media can include any or all of the tangible memory of the computers, processors or the like, or associated modules thereof, such as various semiconductor memories, tape drives, disk drives and the like, which may provide non-transitory storage at any time for the software programming. All or portions of the software may at times be communicated through the Internet or various other telecommunication networks. Such communications, for example, may enable loading of the software from one computer or processor into another, for example, from a management server or host computer into the computer platform of an application server. Thus, another type of media that may bear the software elements includes optical, electrical and electromagnetic waves, such as used across physical interfaces between local devices, through wired and optical landline networks and over various air-links. The physical elements that carry such waves, such as wired or wireless links, optical links or the like, also may be considered as media bearing the software. As used herein, unless restricted to non-transitory, tangible “storage” media, terms such as computer or machine “readable medium” refer to any medium that participates in providing instructions to a processor for execution.

[0088] Hence, a machine readable medium, such as computer-executable code, may take many forms, including but not limited to, a tangible storage medium, a carrier wave medium or physical transmission medium. Non-volatile storage media include, for example, optical or magnetic disks, such as any of the storage devices in any computer(s) or the like, such as may be used to implement the databases, etc. shown in the drawings. Volatile storage media include dynamic memory, such as main memory of such a computer platform. Tangible transmission media include coaxial cables; copper wire and fiber optics, including the wires that comprise a bus with-in a computer system. Carrier-wave transmission media may take the form of electric or electromagnetic signals, or acoustic or light waves such as those generated during radio frequency (RF) and infrared (IR) data communications. Common forms of computer-readable media therefore include for example: a floppy disk, a flexible disk, hard disk, magnetic tape, any other magnetic medium, a CD-ROM, DVD or DVD-ROM, any other optical medium, punch cards paper tape, any other physical storage medium with patterns of holes, a RAM, a ROM, a PROM and EPROM, a FLASH-EPROM, any other memory chip or cartridge, a carrier wave transporting data or instructions, cables or links transporting such a carrier wave, or any other medium from which a computer may read programming code and/or data. Many of these forms of coputer readable media may be involved in carrying one or more sequences of one or more instructions to a processor for execution.

[0089] The computer system 701 can include or be in communication with an electronic display 1535 that comprises a user interface (UI) 740 for providing, for example, the generated 3D proportionately scaled model of the user. Examples of UFs include, without limitation, a graphical user interface (GUI) and web-based user interface. [0090] Methods and systems of the present disclosure can be implemented by way of one or more algorithms. An algorithm can be implemented by way of software upon execution by the central processing unit 705. The algorithm can, for example, analyze the sensor signals such as PPG signals and measure various parameters in the signal. The algorithms may use machine learning or artificial intelligence methods for signals analysis and making the measurements .Example 1

As an example of the neural network used for the patient monitoring system of the disclosure, a 5 layer perceptron neural network such as the one in the example of FIG. 5 may be used. The Neural network may be used for optimizing the detection of abnormalities in PPG leading to atrial fibrillation or detection of adverse events. There may be 7200 input nodes feeding the recorded data to the neural network model. The data input as 30 second recordings at 240 Hz sampling rate. The processor used for the system may be chosen from ARM or Intel processor and may support the industrial standard TensorFlow Lite (from Google LLC). The model may also use an 18 layer Res 18 neural network for the implementation of the machine learning model. There may be two output nodes in the neural network model indicating the presence or absence of an abnormality such as atrial fibrillation.

[0091] While preferred embodiments of the present invention have been shown and described herein, it will be obvious to those skilled in the art that such embodiments are provided by way of example only. Numerous variations, changes, and substitutions will now occur to those skilled in the art without departing from the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in practicing the invention. It is intended that the following claims define the scope of the invention and that methods and structures within the scope of these claims and their equivalents be covered thereby.

Claims

CLAIMS WHAT IS CLAIMED IS:

1. A computer based patient monitoring system comprising:

(A) a server;

(B) a patient monitoring device comprising:

(i) a sensor configured to sense a patient parameter;

(ii) a memory;

(iii) a processor; and

(iv) a non-transitory computer readable medium encoded with instructions configured to cause said processor to:

(a) receive said patient parameter;

(b) analyze said patient parameter by applying a rule to said patient parameter thereby generating an analysis result; and

(c) perform, based on said analysis result, one or more of transmitting an alarm signal, transmitting said patient parameter to said server, and storing said parameter in said memory.

2. The system of claim 1, wherein said server comprises a cloud based server.

3. The system of claim 1, wherein said server is a component of a hospital network.

4. The system of claim 1, wherein said server is a component of said patient monitoring device.

5. The system of claim 1, wherein said sensor comprises a vital sign sensor.

6. The system of claim 1, wherein said sensor is configured for continuous monitoring.

7. The system of claim 1, wherein said patient monitoring device is configured to communicate with said server wirelessly using a WiFi signal or an RF signal.

8. The system of claim 1, wherein said patient monitoring device is configured to communicate with said server wirelessly using a cellular network.

9. The system of claim 1, wherein said patient monitoring device is configured to perform edge computing and wherein said rule is a dynamically changing rule that determines at least one performance characteristic of said patient monitoring device.

10. The system of claim 9, wherein said rule is generated by a machine learning algorithm trained with patient data.

11. The system of claim 9, wherein said performance characteristic comprises performance efficiency.

12. The system of claim 9, wherein said performance characteristic comprises power utilization optimization.

13. The system of claim 9, wherein said performance characteristic comprises a true alarm rate, wherein said true alarm comprises an alarm signal that corresponds to a critically abnormal patient parameter.

14. The system of claim 1, wherein said rule identifies a critically abnormal patient parameter.

15. The system of claim 14, wherein said rule is updated with patient parameter values for a patient that is being monitored with the patient monitoring device.

16. The system of claim 1, wherein said alarm signal is transmitted to a mobile computing device of a healthcare provider.

17. The system of claim 1, wherein said rule comprises a threshold value for a specific patient parameter.

18. A computer-based portable monitoring system comprising:

(A) a server

(B) a wearable monitoring device comprising:

(i) a first and a second ECG electrodes configured to sense a lead I ECG tracing, wherein said first electrode is positioned on said wearable monitoring device so that said first electrode contacts a first upper extremity of a user when said wearable monitoring device is worn by said user;

(ii) a PPG sensor configured to sense a PPG;

(iii) a memory;

(iv) a processor; and

(v) a non-transitory computer readable medium encoded with instructions configured to cause said processor to:

(a) receive said PPG;

(b) identify a signal within said PPG;

(c) analyze said signal thereby generating an analysis result; and

(d) perform, based on said analysis result, one or more of transmitting an indication to said user to contact said second ECG electrode with a second upper extremity thereby generating said lead I ECG tracing, transmitting said lead I ECG tracing to said server, transmitting an alarm signal, and storing said ECG tracing in said memory.

19. The system of claim 18, wherein said server comprises a cloud based server.

20. The system of claim 18, wherein said server is a component of a network of a patient care provider or patient care provider facility.

21. The system of claim 18, wherein said wearable monitoring device further comprises a vital sign sensor.

22. The system of claim 21, wherein the wearable monitoring device further comprise a lab-on- a-chip device.

23. The system of claim 18, wherein said PPG sensor is configured for continuous monitoring.

24. The system of claim 18, wherein said wearable monitoring device is configured to communicate with said server wirelessly using a WiFi signal or an RF signal.

25. The system of claim 18, wherein said wearable monitoring device is configured to communicate with said server wirelessly using a cellular network.

26. The system of claim 18, wherein said wearable monitoring device is configured to perform edge computing and wherein said analysis result is generated using a machine learning algorithm.

27. The system of claim 26, wherein said machine learning algorithm is trained with patient data.

28. The system of claim 26, wherein said machine learning algorithm is trained using data from said memory.

29. The system of claim 26, wherein said machine learning algorithm is configured to determine power utilization optimization of said wearable monitoring device.

30. The system of claim 18, wherein said machine learning algorithm is configured to optimize a rate of true indication comprising a correspondence of said indication with a true presence of an arrhythmia.

31. The system of claim 18, wherein said analysis result comprises a critically abnormal patient parameter.

32. The system of claim 18, wherein said alarm signal is transmitted to a mobile computing device of a healthcare provider.

33. The system of claim 18, wherein said analysis result is based on a threshold value for said PPG signal.

34. A system configured to monitor an individual comprising:

(a) a plurality of sensors;

(b) a processor; and

(c) a non-transitory computer readable medium encoded with software configured to cause said processor to:

(i) receive a first signal from a first sensor of the plurality of sensors at a first time and a second signal from a second sensor of the plurality of sensors at a second time; and

(ii) determine if said individual has experienced a critical event based on the first signal and the second signal.

35. The system of claim 34, wherein said first sensor comprises a heart rate sensor and said second sensor comprises a different sensor.

36. The system of claim 34, wherein at least one of said plurality of sensors comprises a wearable sensor.

37. The system of claim 34, wherein at least of said plurality of sensors comprises an implantable sensor.

38. The system of claim 34, wherein said first time and said second time are the same.

39. The system of claim 34, wherein said critical event comprises a myocardial infarction.

40. The system of claim 34, wherein said critical event comprises an arrhythmia.

41. A non-transitory computer readable medium encoded with software configured to cause said processor to carry out a monitoring process, said monitoring process comprising:

(a) receiving a patient parameter from a sensor;

(b) analyzing said patient parameter by applying a rule to said patient parameter thereby generating an analysis result; and

(c) performing, based on said analysis result, one or more of transmitting an alarm signal, transmitting said patient parameter to a server, and storing said parameter in local memory.

42. The non-transitory computer readable medium of claim 41, wherein said server comprises a cloud based server.

43. The non-transitory computer readable medium of claim 41, wherein said server is a component of a hospital network.

44. The non-transitory computer readable medium of claim 41, wherein said sensor comprises a vital sign sensor.

45. The non-transitory computer readable medium of claim 41, wherein said sensor is configured for continuous monitoring.

46. The non-transitory computer readable medium of claim 41, wherein said patient monitoring device is configured to communicate with said server wirelessly using a WiFi signal or an RF signal.

47. The non-transitory computer readable medium of claim 41, wherein said patient monitoring device is configured to communicate with said server wirelessly using a cellular network.

48. The non-transitory computer readable medium of claim 41, wherein said patient monitoring device is configured to perform edge computing and wherein said rule is a dynamically changing rule that determines at least one performance characteristic of said patient monitoring device.

49. The non-transitory computer readable medium of claim 48, wherein said rule is generated by a machine learning algorithm trained with patient data.

50. The non-transitory computer readable medium of claim 48, wherein said performance characteristic comprises performance efficiency.

51. The non-transitory computer readable medium of claim 48, wherein said performance characteristic comprises power utilization optimization.

52. The non-transitory computer readable medium of claim 48, wherein said performance characteristic comprises a true alarm rate, wherein said true alarm comprises an alarm signal that corresponds to a critically abnormal patient parameter.

53. The non-transitory computer readable medium of claim 41, wherein said rule identifies a critically abnormal patient parameter.

54. The non-transitory computer readable medium of claim 53, wherein said rule is updated with patient parameter values for a patient that is being monitored with the patient monitoring device.

55. The non-transitory computer readable medium of claim 41, wherein said alarm signal is transmitted to a mobile computing device of a healthcare provider.

56. The non-transitory computer readable medium of claim 41, wherein said rule comprises a threshold value for a specific patient parameter.

57. A system configured to monitor an individual comprising:

(a) a plurality of sensors including a PPG sensor and a glucose monitoring sensor;

(b) a processor; and

(c) a non-transitory computer readable medium encoded with software configured to cause said processor to:

(i) receive a first signal from a PPG sensor at a first time and a second signal from the glucose monitoring sensor at a second time; and

(ii) determine if said individual has experienced a critical event based on the first signal and the second signal.

58. The system of claim 57, wherein said first time and said second time are the same.

59. The system of claim 57, wherein said critical event comprises a hypoglycemia event.

60. The system of claim 57, wherein said critical event comprises a hyperglycemia event.