Systems, devices, and methods for evaluation of heart beat parameters involving adjustments based on movement

Abstract

An apparatus includes a photoplethysmography (PPG) sensor configured to measure a PPG signal associated with a user, an accelerometer configured to measure an acceleration signal associated with the user, and a processor operatively coupled to the PPG sensor and the accelerometer. The processor is configured to determine whether the user is engaging in movement based on the acceleration signal, in response to determining that the user is engaging in movement, determine a type of movement of the user based on the acceleration signal, generate a probability distribution associated with a pulse rate of the user based on the type of movement, and determine the pulse rate of the user based on the PPG signal, the acceleration signal, and the probability distribution.

Description

SYSTEMS, DEVICES, AND METHODS FOR EVALUATION OF HEART BEAT PARAMETERS INVOLVING AD JUSTMENTS BASED ON MOVEMENTCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to and benefit of U.S. Provisional Patent Application No. 63 / 667,572, filed on July 3, 2024, and titled, “SYSTEMS, DEVICES, AND METHODS FOR EVALUATION OF HEART BEAT PARAMETERS INVOLVING ADJUSTMENTS BASED ON MOVEMENT,” the disclosure of which is incorporated herein by reference in its entirety.TECHNICAL FIELD

[0002] Devices, systems, and methods herein relate to non-invasive user monitoring.BACKGROUND

[0003] Vital signs such as heart rate are commonly used to indicate the status and health of a user. For example, vital signs may be used to monitor health conditions and for early detection of conditions and / or diseases. Therefore, additional systems, devices, and methods for non- invasive user monitoring may be desirable.SUMMARY(0004] In one embodiment, an apparatus includes a photoplethysmography (PPG) sensor configured to measure a PPG signal associated with a user. The apparatus includes an accelerometer configured to measure an acceleration signal associated with the user. The apparatus includes a processor operatively coupled to the PPG sensor and the accelerometer. The processor is configured to determine whether the user is engaging in movement based on the acceleration signal, in response to determining that the user is engaging in movement, determine a type of movement of the user based on the acceleration signal, generate a probability distribution associated with a pulse rate of the user based on the type of movement, and determine the pulse rate of the user based on the PPG signal, the acceleration signal, and the probability distribution.

[0005] In one embodiment, a method includes receiving a photoplethysmography (PPG) signal associated with a user from a PPG sensor. The method includes receiving an accelerationsignal associated with the user from an accelerometer. The method includes determining whether the user is engaging in movement based on the acceleration signal. The method includes, in response to determining that the user is engaging in movement, determining a type of movement of the user based on the acceleration signal. The method includes determining, using a machine learning model, a pulse rate of the user based on a power spectrum analysis of the PPG signal and the acceleration signal.

[0006] In one embodiment, a method includes receiving a photoplethysmography (PPG) signal associated with a user from a PPG sensor, the PPG signal including data measured over a plurality of time windows. The method includes receiving an acceleration signal associated with the user from an accelerometer, the PPG signal including data measured over the plurality of time windows. The method includes determining whether the user is engaging in movement based on the acceleration signal. The method includes, in response to determining that the user is engaging in movement, determining a type of movement of the user based on the acceleration signal. The method includes determining, for each time window of the plurality of time windows, a pulse rate of the user based on the PPG signal and the acceleration signal. The method includes determining, based on the pulse rate for each time window of the plurality of time windows, a pulse rate biomarker.BRIEF DESCRIPTION OF THE DRAWINGS

[0007] FIG. 1 A is a block diagram of a sensing system, according to an embodiment. FIG. IB is a block diagram of a compute device of the sensing system, according to an embodiment.

[0008] FIG. 2 is a block diagram of a sensing system, according to an embodiment.

[0009] FIGS. 3A-3C are schematic diagrams of respective sensing devices, according to embodiments.

[0010] FIG. 4 is a flow chart illustrating a method of estimating a heart rate biomarker, according to an embodiment.

[0011] FIG. 5A is an example of a probability distribution, according to an embodiment. FIG. 5B is the probability distribution of FIG. 5A after thresholding, according to an embodiment.

[0012] FIGS. 5C-5D show examples of probability distributions, according to embodiments.

[0013] FIG. 6 is a perspective view of a sensing device, according to an embodiment.

[0014] FIG. 7A is a perspective view of a sensing device, according to an embodiment. FIG. 7B is a schematic diagram of a light source of the sensing device, according to an embodiment.

[0015] FIG. 8 is a schematic diagram of a sensor, according to an embodiment.DETAILED DESCRIPTION|0016| Described here are devices, systems, and methods for providing real-time, non- invasive monitoring of one or more physiological parameters, which may be used to estimate one or more vital signs of a user (e.g., patient). These systems and methods may, for example, receive physiological data of a patient, and process that data to determine a biomarker (e.g., a pulse rate (PR) biomarker or other cardiac parameter). In some embodiments, systems, devices, and methods described herein may receive accelerometer data and determine movement characteristics of a patient. Such movement characteristics (or data indicative thereof) can also be used to process other physiological data (e.g., photoplethysmography (PPG) data). This may, for example, allow insight into a patient’s health status on a continuous or semi- continuous, real-time basis. The devices described herein for use in, for example, estimating a cardiac parameter may be compact and portable such that they may allow for continuous or semi-continuous, real-time monitoring without restricting the day-to-day activities of a patient. For example, the device may be a wearable device worn on a patient’s wrist.[001.7] Additionally or alternatively, health care professionals may remotely monitor patients using the systems, devices, and methods described herein on a more frequent basis than intermittent clinic visits, thereby reducing costs. Current standard of care often relies on a combination of patient visual appearance, blood tests (e.g., serologic test), and vital measurements. The data provided from the monitoring system can supplement these traditional data sets and provide additional insights into patient status, and account for a patient’s historical (e.g., reference baseline) status. Continuous (or periodic) patient monitoring also allows health care professionals to address complications and / or poor treatment efficacy in real-time before issues exacerbate.I. Systems

[0018] Generally, the systems described here may include a sensing device (e.g., patient monitor including one or more devices with sensors) and one or more other compute devices, e.g., a partner device, a network, a server, and a database. The sensing device may measure patient data, and may, in some embodiments, transmit the patient data to another compute device, partner device, remote server, and / or database for processing and analysis. In other embodiments, the patient data may be processed and analyzed by the sensing device itself. As mentioned above, the patient data may include one or more of physiological data and demographic data. The physiological data may be measured using one or more sensors of one or more devices (e.g., wearable device, smartphone, portable device). The patient data may be processed and analyzed to determine a biomarker (e.g., a PR biomarker or some measure associated therewith). The measurement of patient data may be performed for predetermined intervals or continuously. The results of the parameter estimation may be output to one or more of the sensing device, compute device, partner device, network, server, database, combinations thereof, and the like. Additionally (e.g., concurrently or subsequently) or alternatively, the estimated parameter(s) may be output to one or more of a health care professional and designated users (e.g., partner, family, support group, researchers).

[0019] A patient monitoring system (e.g., sensing system) may include one or more of the components necessary to measure and / or generate physiological data using the devices as described herein. FIG. 1 A is a block diagram of an embodiment of a patient monitoring system 100. As shown there, the system 100 may include a sensing device (e.g., user device, patient monitor) 110 and a compute device 120.

[0020] In some embodiments, measured patient data may be processed and / or analyzed (e.g., for parameter estimation) on any one of the devices of the system 100 (e.g., sensing device 110, compute device 120), while in other embodiments, the processing may be distributed throughout a plurality of devices. In some embodiments, patient data processing may include filtering data (e.g., reject data, remove artifacts), determining key events (e.g., extract features), calculating a metric, and / or calculating a confidence score associated therewith. In some embodiments, one or more of patient data and patient identifying information may be encrypted and stored in memory 114, 124 according to Health Insurance Portability and Accountability Act (HIPAA) regulations.

[0021] The sensing device 110 can be a compute device that is associated with a user. The sensing device 110 can be configured to removably attach to a patient and measure patient data (e.g., physiological data). As described in more detail herein, the sensing device 110 may include one or more of an optical sensor, PPG sensor, cardiac sensor, accelerometer, electrodermal activity sensor, temperature sensor, magnetometer, altimeter, electrocardiogram (ECG) sensor, electromyography (EMG) sensor, or ambient light sensor. In some embodiments, the sensing device 110 may be configured to be a wearable device. For example, the sensing device 110 may be configured to be worn on a patient’s limb (e.g., wrist, arm, calf).

[0022] The sensing device 110 may further include a communications interface (e.g., communication device) configured to establish a communication channel with one or more other devices or systems. For example, the sensing device 110 may be coupled to compute device 120 through one or more wired or wireless communication channels. The sensing device 110 may be operatively coupled one or more compute devices 120, networks, servers, databases, and other devices, as further detailed below.

[0023] The sensing device 110 may be configured to measure patient data (e.g., accelerometer data, cardiac data, physiological data) during a plurality of time periods. In some embodiments, the measured patient data may be transmitted to a compute device 120 for data processing and metric calculations as described herein. In some embodiments, the sensing device 110 may be controlled from one or more other compute devices. In some embodiments, the sensing device 110 described herein may be configured to perform a subset of the measurement, processing, and calculation steps described herein.

[0024] In some embodiments, the sensing device 110 may include one or more of a processor 112, a memory 114, one or more sensors 116, an input / output (I / O) device 118, and a communications interface 119.

[0025] The processor, memory, and input / output device are described in more detail herein. The one or more sensors 116 may be configured to measure data associated with one or more physiological signals and / or parameters. In some embodiments, the physiological parameters may include a cardiac parameter (e.g., blood volume changes, heart rate, heart rate index, heart rate variability, blood pressure, blood-oxygen level), patient activity (e.g., accelerometer data, movement), patient position, sleep / wake status, sleep quality, NREM / REM staging, electrodermal activity (e.g., skin conductance, skin resistance, skin potential, motion, etc.), respiration, respiration rate index, metabolic equivalent of task (MET), quantity and type ofmotion (MOT), stress level, relaxation level, temperature, skin temperature, heat flux, autonomic nervous system (ANS) activity (e.g., indicating levels of sympathetic, parasympathetic, and enteric nervous system arousal or activation), muscle electrical potential, nerve electrical potential, brain waves, steps taken, pace, distance, altitude, direction, velocity, speed, time elapsed, time left, proteins (e.g., cytokines, Interleukin-6 (IL-6) and Interleukin- 12 (IL- 12), corresponding physiologically relevant indexes, combinations thereof, and the like.

[0026] In some embodiments, one or more metrics (e.g., parameters) may be derived from a set of other metrics. For example, heart rate can be derived or determined through analysis of one or more PPG signal waveforms. In some embodiments, heart rate variability (HRV) may be derived or determined as well. HRV may be defined as the beat-to-beat variations in heart rate. The larger the alterations, the larger the HRV.

[0027] In some embodiments, the sensor 116 may include an optical sensor, PPG sensor, cardiac sensor (e.g., electrocardiography), blood oxygen sensor, accelerometer, electrodermal activity sensor, gyroscope, geolocation sensor (e.g., GPS), barometer, pressure sensor, temperature sensor (e.g., skin temperature sensor, ambient temperature sensor), glucose sensor, barometer, electrodes, AC current sensor, DC current sensor, light emitter, magnetometer, capacitive sensor, humidity sensor, altimeter, 1-lead electrocardiogram (ECG) sensor, electromyography (EMG) sensor, ambient light sensor, cytokine sensor, protein sensor, combinations thereof, and the like. In some embodiments, the accelerometer may include one or more of a 3-axis accelerometer and gyroscope configured to measure one or more of acceleration, movement, and position. For example, the accelerometer or gyroscope may be configured to quantify the duration a patient is lying down. In some embodiments, the total daily duration that the patient is not lying down can be monitored as a metric for patient activity. A predetermined deviation from the reference baseline duration may be configured to output a notification (e.g., alert, instruction).

[0028] In some embodiments, the processor 112 of the sensing device 110 may incorporate data received from memory 114 of the sensing device 110 and over a communication channel to control one or more components, e.g., of the sensing device 110. The memory 114 may further store instructions to cause the processor 112 to execute modules, processes and / or functions associated with the methods described herein. In some embodiments, the memory 114 and processor 112 may be implemented on a single chip. In other embodiments, they can be implemented on separate chips.

[0029] In some embodiments, the sensing device 110 may be coupled directly to the compute device 120. In some embodiments, the sensing device 110 may be coupled to an external device (not shown) for storage, to recharge, to transfer data, combinations thereof, and the like. In some embodiments, the external device can be a charging device, a dock, a connector, etc. The sensing device 110 may include a power source (e.g., battery) configured to provide electrical power to the sensing device 110 and / or the sensing device 110 can be couplable to an external power source (e.g., via the external device). In some embodiments, the wearable device can include a wrist band (e.g., strap) or other attachment mechanism (e.g., magnet, adhesive, etc.) for releasable attachment to a user.

[0030] Suitable examples of devices for measuring physiological parameters of a user include, for example, devices such as those described in U.S. Patent No. 10,506,944, titled “Apparatus for electrodermal activity measurement with current compensation,” issued on December 17, 2019, U.S. Patent No. 9,833,155, titled “Device, system and method for detection and processing of heartbeat signals,” issued on December 5, 2017, U.S. Patent No. 10,134,378, titled “Systems, apparatuses, and methods for adaptive noise reduction,” issued on November 20, 2018, U.S. Patent Publication No. US2022 / 0265214, titled “Wearable biosensing device,” published on August 25, 2022, and U.S. Patent Publication No. US2023 / 0039091, titled “Methods and systems for non-invasive forecasting, detection and monitoring of viral infections,” published on February 9, 2023, the contents of each of which are incorporated herein by reference.

[0031] In some embodiments, the compute device 120 may include one or more of a processor 122, a memory 124, an input / output device 128, and a communications interface 129. Alternatively, the compute device 120 can be another sensing device. For example, the compute device 120 can be a cellular telephone (e.g., smartphone), tablet computer, laptop computer, desktop computer, portable media player, and the like.

[0032] The compute device 120 may be configured to receive various types of data. For example, the compute device 120 may be configured to receive demographic data (e.g., gender, weight, Body Mass Index, height, birthday, age, genetic information, diagnosis date, anniversary date using the device, pre-existing conditions, medication history, medical history), patient data (e.g., blood pressure data, heart rate data, electrodermal activity data), general health information of other similarly situated patients (e.g., cohort patient data), template waveforms, or any other relevant information. For example, medical history mayinclude a history of one or more of respiratory illness, immune-system disease, hypertension, cardiovascular disease, diabetes, and the like.

[0033] In some embodiments, the compute device 120 may be configured to create, receive, and / or store patient profiles. A patient profile may contain any of the patient demographic data previously described. While the above-mentioned information may be received by the compute device, in some embodiments, the compute device may be configured to process any of the above data from information it has received using software stored on the device itself, or externally.100341 The processor 122 of the compute device 120 can be configured to receive patient data from the sensing device 110 and other data (e.g., demographic data) from other sources (e.g., network, database, server). The processor 122 may be configured to receive, process, analyze, compile, store, and access data. The processor 122 may be configured to receive data directly input and / or measured from a patient. The processor 122 may receive the data through a network connection, as discussed in more detail herein, or through a physical connection with the device or storage medium (e.g. through Universal Serial Bus (USB) or any other type of port).

[0035] Processors 112, 122 may be any suitable processing device configured to run and / or execute a set of instructions or code and may include one or more data processors, image processors, graphics processing units, physics processing units, digital signal processors, and / or central processing units. More specifically, processor 112, 122 may be configured to execute instructions associated with modules (e.g., pre-processing 124a, motion artifact removal 124b, movement classification 124c, transition probability estimation 124d, threshold 124e, physiological parameter model 124f), functions, and / or processes. Each processor 112, 122 may be, for example, a general purpose processor, Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), and / or the like. The underlying device technologies may be provided in a variety of component types (e.g., metal-oxide semiconductor field-effect transistor (MOSFET) technologies like complementary metal-oxide semiconductor (CMOS), bipolar technologies like emitter-coupled logic (ECL), polymer technologies (e.g., silicon-conjugated polymer and metal-conjugated polymer-metal structures), mixed analog and digital, and / or the like.

[0036] Each memory 114, 124 may include a database (not shown) and may be, for example, a random access memory (RAM), a memory buffer, a hard drive, an erasableprogrammable read-only memory (EPROM), an electrically erasable read-only memory (EEPROM), a read-only memory (ROM), Flash memory, and the like. Each memory 114, 124 may store instructions to cause the processor to execute modules, processes, and / or functions associated with the communication device, such as patient data processing, sensor measurement, parameter estimation, sensing device or patient monitoring control, authentication, encryption, and / or communication. Some embodiments described herein relate to a computer storage product with a non-transitory computer-readable medium (also may be referred to as a non-transitory processor-readable medium) having instructions or computer code thereon for performing various computer-implemented operations. The computer- readable medium (or processor-readable medium) is non-transitory in the sense that it does not include transitory propagating signals per se (e.g., a propagating electromagnetic wave carrying information on a transmission medium such as space or a cable). The media and computer code (also may be referred to as code or algorithm) may be those designed and constructed for the specific purpose or purposes.(0037] Examples of non-transitory computer-readable media include, but are not limited to, magnetic storage media such as hard disks, floppy disks, and magnetic tape; optical storage media such as Compact Disc / Digital Video Discs (CD / DVDs); Compact Disc-Read Only Memories (CD-ROMs), and holographic devices; magneto-optical storage media such as optical disks; solid state storage devices such as a solid state drive (SSD) and a solid state hybrid drive (SSHD); carrier wave signal processing modules; and hardware devices that are specially configured to store and execute program code, such as Application-Specific Integrated Circuits (ASICs), Programmable Logic Devices (PLDs), Read-Only Memory (ROM), and Random-Access Memory (RAM) devices. Other embodiments described herein relate to a computer program product, which may include, for example, the instructions and / or computer code disclosed herein.

[0038] FIG. IB is a block diagram of the compute device 120 of the sensing system 100, according to embodiments. In some embodiments, memory 124 stores instructions that cause processor 122 to execute modules, processes, and / or functions associated with processing or analyzing sensor data and / or determining physiological information of a user. In some embodiments, the memory 124 may store instructions, algorithms, and / or models associated with one or more of pre-processing 124a, motion artifact removal 124b, movement classification 124c, transition probability estimation 124d, thresholds 124e, and physiological parameter model 124f. In some embodiments, memory 124 may optionally include patient data.

[0039] The pre-processing 124a is optional. In some embodiments where pre-processing 124a is implemented, it can include processing data received by the compute device, e.g., from one or more sensors. The data received at the compute device can include, for example, accelerometer data and / or PPG data. The processing can include filtering, smoothing, normalization, removing noise, amplifying, and / or the like. In some embodiments, the preprocessing 124a can be unique to each type of data. In other words, different pre-processing 124a can be applied to different data. For example, in some embodiments, the pre-processing 124a may include applying a first band-pass filter to the PPG data and a second band-pass filter to the accelerometer data, where the first and second band-pass filters may have different bandpass ranges. In some embodiments, the pro-processing 124a may include applying a band-pass filter to the PPG data but not to the accelerometer data, and vice versa. In some embodiments, the pre-processing 124a may include applying the same band-pass filter to the PPG data and the accelerometer data. In some embodiments, the pre-processing 124a may not be applied to certain data types while it is applied to other data types. For example, no pre-processing 124a may be applied to accelerometer data, while pre-processing may be applied to PPG data. In some embodiments, pre-processing 124a may include segmenting the sensor data or splitting the data into windows, e.g., based on predetermined criteria. For example, the pre-processing 124a may include splitting data into n-second (e.g., 0.1 second, 0.5 second, 1 second, 5 second, 10 second, 30 second, etc.) segments or time periods. In some embodiments, the pre-processing 124a may include data rejection algorithms. For example, data pre-processing 124a may include determining whether a portion of data is not suitable for further processing and / or analysis and may reject that portion of the data. For example, if the data has a predetermined amount of noise, has gaps, has unexpected values, is saturated too high or too low, etc., the data may be rejected.

[0040] Optionally, motion artifact removal 124b may be applied to the sensor data (e.g., accelerometer data, PPG data). In some embodiments, the motion artifact removal 124b may include applying adaptive filtering algorithms, e.g., such as those described in P.C.T. Patent Application No. PCT / US2024 / 039838, title “Systems, Devices, and Methods for Reducing Signal Noise,” filed July 26, 2024, the content of which is incorporated herein by reference. In some embodiments, motion artifact removal 124b may be applied to sensor data associated with a predetermined degree of movement. For example, motion artifact removal 124b may be applied to sensor data when a predetermined amount of movement or noise is expected in the sensor data, such as, for example, when a user is engaging in an activity that involves apredetermined degree of movement, when the sensor data has a predetermined amount of noise or artifacts associated with it, etc. In some embodiments, if the sensor data is not associated with a predetermined degree of movement, noise, artifacts, etc., then the motion artifact removal 124b may not be applied to the sensor data. For example, when sensor data is being collected during rest or sleep, or during other activities that are associated with less movement, then the motion artifact removal 124b may not be applied. In such instances, applying the motion artifact removal 124b may remove useful information from the raw or pre-processing sensor data and / or cause over-filtering of the sensor data and introduce inaccuracies. In some embodiments, the motion artifact removal 124b that is applied may depend on type of movement or activity, e.g., as provided by a user and / or determined by the compute device (e.g., in implementing movement classification 124c). For example, a different adaptive filtering model may be applied to each type of data, such as, for example, a first model for running, a second model for walking, a third model for gesturing, etc.[00411 In some embodiments, the movement classification 124c may include applying one or more models or algorithms to detect that movement and / or activity is occurring and / or determine that a type of movement and / or activity is occurring. In some embodiments, the movement classification 124c may include applying a hierarchical model such as a cascading decision tree, e.g., to determining one or more of (1) whether movement is occurring, (2) the intensity of the movement, and / or (3) the type of movement or activity. For example, the movement classification 124c may include determining, at a first decision node of a cascading tree, whether movement is occurring or not. If movement is determined, then the movement classification 124c may include determining, at a second decision node of the cascading tree, whether the movement is of a first type or second type. For example, the movement classification 124c may determine whether the movement is associated with locomotion (e.g., walking, running, etc.) or gesturing (e.g., stationary hand movements, etc.). If the movement is determined to be locomotion, the movement classification 124c may include determining, at a subsequent decision node of the cascading tree, the type of movement. For example, the type of movement may include walking, running, cycling, swimming, team sports (e.g., soccer, football, ice hockey, etc.), individual sports (e.g., golf, tennis, rock climbing, etc.), and / or the like. In some embodiments, movement classification 124c may be based on acceleration data, e.g., captured by one or more accelerometers. In some embodiments, movement classification 124c may be based on PPG data, e.g., captured by one or more PPG sensors. In some embodiments, movement classification 124c may be based on a combination of sensor data,including, for example, one or more of, acceleration data, PPG data, temperature data, respiratory data, gyroscopic data, etc. In some embodiments, the movement classification 124c may include requesting an input from a user, e.g., such as an input indicating movement and / or the type of physical activity. For example, a user may input that the user is running. When a user inputs his movement and / or type of physical activity, then the movement classification 124c may determine the user’s movement and / or activity based on the user’s input.

[0042] In some embodiments, an optional transition probability estimation 124d may be performed. In some embodiments, the transition probability estimation 124d may include generating a probability distribution associated with a biomarker or physiological parameter of interest. The probability distribution can be used to limit the amount of change that may be determined for a particular biomarker or physiological parameter. The probability distribution can indicate the probability that a biomarker or physiological parameter is a certain value and / or within a certain range based on predetermined inputs, including, for example, the movement classification, acceleration data, PPG data, or other sensor data. In some embodiments, the probability distribution may also be generated based on the signal quality of one or more inputs, e.g., the amount of noise in an input signal. For example, if one or more input signals exhibit a high degree of noise, then the probability distribution may have a narrow profile centered around the value that was determined for a particular biomarker or physiological parameter during a previous period. This can avoid the value for the biomarker or physiological parameter changing significantly due to noisy data.

[0043] In some embodiments, the probability distribution may take on different profiles based on one or more inputs. For example, different inputs from the movement classification 124c may impact whether the probability that a biomarker or physiological parameter may change in different ways. In some embodiments, when evaluating heart rate or pulse rate, the probability distribution may change depending on movement and / or activity of the user. For example, if the movement classification 124c indicates that the user is exercising, then the probability distribution may have a positive skew such that higher heart rate values may be associated with higher probabilities relative to lower heart rate values. During exercise, changes in heart rate in a positive direction may be expected, while changes in heart rate in a negative direction may be less expected and attributable to noise and / or erroneous sensor data. In some embodiments, the degree of skew of the probability distribution may be dependent on an intensity of the physical activity and / or exercise.

[0044] FIGS. 5A-5D depict four different examples of probability distributions, according to embodiments. The probability distributions may indicate the probability that a biomarker or physiological parameter is a particular value for a range of values during a time period n (i.e., t=ri). As shown in FIG. 5A, the probability distribution 500 is symmetrical with a peak 502. The peak 502 may be set at a value that corresponds to a value determined for the biomarker or physiological parameter during a previous or last time period (e.g., t=n-l The likelihood that a user’s biomarker or physiological parameter does not change from one period to the next may be most likely, and therefore the value of the probability distribution may be highest at this value. With a symmetrical distribution as shown in FIG. 5A, there is an equal probability that the value for the biomarker or physiological parameter may increase or decrease. The distribution may have a predetermined standard deviation, which may be different for different biomarkers or physiological parameters, according to some embodiments. Additionally or alternatively, the standard deviation of the distribution may change depending on whether one or more sensor data inputs are noisy and / or whether the user is engaging in activity with less or more movement where more movement. In some embodiments, the probability distribution may have a narrow profile, e.g., as shown with the probability distribution 540 in FIG. 5D, when input data may be noisy, the user may be engaging in high movement activity, and / or the like. The user’s movement or activity at high levels may affect sensor capture, and therefore a narrower distribution may be used to limit changes in the value of the biomarker or physiological parameter. Conversely, if the input data is cleaner and / or there is less movement, then the probability distribution may have a higher standard deviation, e.g., to allow for greater changes in the determined value of the biomarker or physiological parameter.

[0045] In some embodiments, depending on a user’s movement and / or activity, a skewed distribution may also be used. As described above, a probability distribution for determining heart rate can be skewed negative when a user is engaging in high intensity activity or exercise, as it can be expected that heart rate may increase during exercise. This is shown by the probability distribution 520 in FIG. 5C. In other instances, e.g., when a user is resting, the probability distribution may be skewed negative. This can attribute higher probabilities to lower heart rates.

[0046] Optionally, in some embodiments, the processor 122 may also implement thresholds 124e, whereby thresholds may be applied to limit the lower or upper bound of the determined value for a biomarker or physiological parameter. For example, the threshold 124e may be applied to limit the range of values determined for a biomarker or physiologicalparameter to a predetermined range of values that are commonly associated with the biomarker or physiological parameter, e.g., based on literature and / or physiological possibility, user history and / or condition, etc. The thresholds 124e therefore prevent unexpected spikes or changes in the value of the biomarker or physiological parameter, e.g., due to inaccurate sensor readings that may result from a sensor becoming loose, getting wet, being worn improperly, etc. In some embodiments, if the determined value for the biomarker or physiological parameter were to fall outside of the expected range of values set for the biomarker or physiological parameter (e.g., based on the thresholds 124e), then a visual or audio alert, or other type of message and / or alert, may be provided to a user to confirm accurate sensor placement.

[0047] In some embodiments, the thresholds 124e can be applied to the probability distribution. As shown in FIG. 5B, the probability distribution 500’ may be cutoff at a lower threshold 504 and at an upper threshold 506. This adjusted probability distribution 500’ filters out unlikely values of the biomarker or physiological parameter.

[0048] In some embodiments, the physiological parameter model 124f may be configured to determine a biomarker or physiological parameter based on one or more inputs, e.g., the probability distribution, the movement classification, the sensor data (raw, pre-processed, or after motion artifact removal), etc. The physiological parameter model 124f may receive these inputs and generate an output that is indicative of the biomarker or physiological parameter. In an embodiment, the physiological parameter is heart rate or pulse rate. In some embodiments, one or more inputs such as the sensor data (raw, pre-processed, or after motion artifact removal) may be transformed into a suitable format for further processing by the physiological parameter model 124f. For example, a Fourier transform may be used to transform the sensor data (e.g., acceleration data, PPG data) into spectral domain data before the data is used to determine the physiological parameter by the physiological parameter model 124f. In some embodiments, the physiological parameter model 124f includes at least one of a recurrent neural network (RNN), a support vector model (SVM), a hidden Markov model (HMM), and / or the like. In some embodiments, the physiological parameter model 124f may generate an output indicative of the physiological parameter for each window of data (e.g., associated with between about 0.1 seconds and about 30 seconds, inclusive of all ranges and values therebetween), and then average the values determined for the physiological parameter over a larger period of time (e.g., between about 30 seconds and 5 minutes, inclusive of all ranges and values therebetween).

[0049] While particular algorithms and / or methods for determining a physiological parameter such as heart rate are described herein, it can be appreciated that other algorithms and / or methods for determining heart rate can be used. The algorithms and / or methods described herein may be suitable for determining heart rate in sensor data associated with greater movement and / or motion artifacts. In some embodiments, it may also be beneficial to compare and / or average the heart rate determined using different algorithms and / or methods, e.g., to improve accuracy.

[0050] Referring back to FIGS. 1A-1B, input / output devices 118, 128 may be configured to permit a user to control one or more of the devices of the system. For example, an input / output device 128 of the compute device 120 may include an input device for a user to input commands and an output device for a user to receive output (e.g., heart rate readings on a display device). Each input / output device 118, 128 may include a network interface configured to connect the compute device to another system (e.g., Internet, remote server, database) by wired or wireless connection (e.g., via network). In some embodiments, the network interface may include a radiofrequency (RF) receiver, transmitter, and / or optical (e.g., infrared) receiver and transmitter configured to communicate with one or more devices and / or networks. The network interface may communicate by wires and / or wirelessly with one or more of the sensing device 110, compute device 120, network, database, servers, and other devices.[0051 [ In some embodiments, an output device of a compute device (e.g., incorporated in an input / output device 118, 128) may output calculated metrics (e.g., cardiac parameter). Data analysis may be displayed by the output device (e.g., display). Data including a calculated metric or metric derived therefrom may be output visually and / or audibly through one or more output devices. In some embodiments, an output device may include a display device including at least one of a light emitting diode (LED), liquid crystal display (LCD), electroluminescent display (ELD), plasma display panel (PDP), thin film transistor (TFT), organic light emitting diodes (OLED), electronic paper / e-ink display, laser display, and / or holographic display. In some embodiments, the output device can include an audio device. The audio device may audibly output patient data, physiological data, system data, alarms and / or notifications. For example, the audio device may output an audible alarm when a calculated metric reaches a predetermined threshold or when a malfunction in the sensing device 110 is detected. In some embodiments, an audio device may include at least one of a speaker, piezoelectric audio device, magnetostrictive speaker, and / or digital speaker. In some embodiments, a patient maycommunicate with other users using the audio device and a communication channel. For example, a patient may form an audio communication channel (e.g., VoIP call) with a remote health care professional.

[0052] In some embodiments, an input device of a compute device (e.g., input / output device 118, 128) may include at least one switch configured to generate a control signal. For example, an input device may include a touch surface for a user to provide input (e.g., finger contact to the touch surface) corresponding to a control signal. An input device comprising a touch surface may be configured to detect contact and movement on the touch surface using any of a plurality of touch sensitivity technologies including capacitive, resistive, infrared, optical imaging, dispersive signal, acoustic pulse recognition, and surface acoustic wave technologies. In embodiments of an input device comprising at least one switch, a switch may include, for example, at least one of a button (e.g., hard key, soft key), touch surface, keyboard, analog stick (e.g., joystick), directional pad, mouse, trackball, jog dial, step switch, rocker switch, pointer device (e.g., stylus), motion sensor, image sensor, and microphone. A motion sensor may receive patient movement data from an optical sensor and classify a user gesture as a control signal. A microphone may receive audio data and recognize a patient voice as a control signal.

[0053] In some embodiments, a haptic device may be incorporated into an input / output device (e.g., input / output device 118, 128) to provide additional sensory output (e.g., force feedback) to the user. For example, a haptic device may generate a tactile response (e.g., vibration) to confirm user input to an input device (e.g., touch surface). As another example, haptic feedback may notify that user input is overridden by the compute device. In some embodiments, a haptic device can be used to send alerts to a user (e.g., when a user has an abnormal heart rate, when a sensor reading is abnormal, etc.).

[0054] FIG. 2 is a block diagram of an embodiment of a sensing system 200. As shown there, the system 200 may include a sensing device (e.g., user device, patient monitor) 210 similar to sensing device 110 described herein, a compute device 220, and a network 202, and optionally a server 250, a database 260, and other device(s) 290. In some embodiments, the sensing device 210 may be in communication with one or more of the compute device 220 and the network(s) 202. In some embodiments, the sensing device 210 may be coupled directly to any of the compute device 220, network(s) 202, server 250, database 260, and other device(s) 290.

[0055] The compute device 220 may be an external device that is operatively coupled to or integrated with the sensing device 210, a server 250, database 260, and / or other device 290 e.g., via network 202. In some embodiments, the compute device 220 may be a user device, e.g., a mobile device, a cellphone, a tablet, a personal computer, etc. In some embodiments, the compute device 220 can be associated with a server 250. The server 250 can be a dedicated server that receives data and signals from and sends data and signals to one or more sensing device(s) 110 and / or compute device(s) 220. In some embodiments, the server 250 can be configured to receive raw and / or processed data associated with a user (e.g., physiological data associated with a user) from the sensing device 210, and to process and / or analyze the data to determine other information about the user (e.g., prediction of user health, conditions, and / or other physiological data, characteristics, and / or information). The database 260 can be any type of data storage device or collection of data storage devices, including, for example, one or more databases associated with a hospital (e.g., for storing patient information), one or more databases associated with a health monitoring application (e.g., for storing physiological data of users), etc.

[0056] The network 202 can be any type of network (e.g., a local area network (LAN), a wide area network (WAN), a virtual network, a telecommunications network) implemented as a wired network and / or wireless network and used to operatively couple compute devices, including sensing devices 110, compute devices 120, and / or partner devices 130. The communication may or may not be encrypted. A wireless network may refer to any type of digital network that is not connected by cables of any kind. Examples of wireless communication in a wireless network include, but are not limited to cellular, radio, satellite, and microwave communication. However, a wireless network may connect to a wired network in order to interface with the Internet, other carrier voice and data networks, business networks, and personal networks. A wired network is typically carried over copper twisted pair, coaxial cable and / or fiber optic cables. There are many different types of wired networks including wide area networks (WAN), metropolitan area networks (MAN), local area networks (LAN), Internet area networks (IAN), campus area networks (CAN), global area networks (GAN), like the Internet, and virtual private networks (VPN). The network may include or be coupled to one or more databases and servers for processing and / or storage.

[0057] FIGS. 3A-3C are schematic diagrams of respective sensing devices. FIG. 3A depicts a sensing device 310, similar to sensing devices 110, 210 described herein, coupled to a tissue T (e.g., skin) having a vessel V (e.g., blood vessel). In some embodiments, the vesselV includes one or more vessels. As shown in FIG. 3 A, the sensing device 310 may be placed near the vessel V (e.g., adjacent to and / or engaged with a surface of tissue T near vessel V). For example, the sensing device 310 may be placed against a skin surface that is above the vessel V.[0058| In some embodiments, the sensing device 310 may include a processor 312, an accelerometer 318, an emitter 316 (e.g., light emitter, electromagnetic radiation (EMR) emitter, LED, diode), and a detector 317 (e.g., light detector, EMR detector, photodiode, phototransistor). The sensing device 310 may be configured to measure accelerometer data and cardiac data such as a PPG signal. For example, the emitter 316 may be configured to output a light emission 320 into the tissue T towards the vessel V. The detector 317 may be configured to receive a reflection 322 (e.g., light reflection) from the vessel V and through the tissue T. The processor may be configured to process the received reflection 322 as measured cardiac data (e.g., PPG signal) as described in more detail herein. The emitter 316 and detector 317 may be configured for predetermined intensities and / or wavelengths of light. For example, the emitter 316 may be configured to emit any wavelength of light that is safe and suitable for capturing the pulsatile movement of the vessel V, and the detector 317 may be configured to detect the wavelength of light emitted by the emitter 316 (or a range of wavelengths including the wavelength of light emitted by emitter 316). In some embodiments, the sensing device 310 may include a plurality of emitter / detector pairs.

[0059] FIG. 3B depicts a sensing device 310', similar to sensing devices 110, 210, 310 described herein, coupled to a tissue T (e.g., skin) having a vessel V (e.g., blood vessel). In some embodiments, the sensing device 310' may include a processor 312, an accelerometer 318, an emitter 316, and a detector 317. The sensing device 310' may be configured to measure accelerometer data and cardiac data such as a PPG signal. A first actuator 319a may be coupled to the emitter 316 and configured to actuate (e.g., displace) the emitter 316. Actuation of the emitter 316 by the first actuator 319a may modify a path length and an angle of incidence of light emission 320 relative to the tissue T. A second actuator 319b may be coupled to the detector 317 and configured to actuate (e.g., displace) the detector 317. Actuation of the detector 317 by the second actuator 319b may modify a path length and an angle of reflection 322 relative to the tissue T. Actuation of the emitter 316 and / or detector 317 enables a sensing distance / location to be adjusted. In some embodiments, the first actuator 319a and the second actuator 319b may be configured to actuate a respective emitter 316 and detector 317 with up to six degrees of freedom (e.g., translation along one or more axes and / or rotation about one ormore axes). In some embodiments, the sensing device 310' may include a plurality of emitter / detector pairs with corresponding actuators.

[0060] The emitter 316 may be configured to output a light emission 320 into the tissue T towards the vessel V. The detector 317 may be configured to receive a reflection 322 (e.g., light reflection) from the vessel V and through the tissue T. The processor may be configured to process the received reflection 322 as measured cardiac data (e.g., PPG signal). In some embodiments, a plurality of signals corresponding to a plurality of path lengths may be measured to obtain a stronger signal and / or provided for signal validation. Factors including skin pigmentation (e.g., skin tone) and other skin characteristics (e.g., hair, scars, moles) may affect light emission and reflection through skin such that different sensing distances, locations, and angles may facilitate accurate cardiac data measurement for different demographic groups.

[0061] FIG. 3C depicts a sensing device 310", similar to sensing devices 110, 210, 310, 310' described herein, coupled to a tissue T (e.g., skin) having a vessel V (e.g., blood vessel). In some embodiments, the sensing device 310" may include a processor 312, an accelerometer 318, a first emitter 316a, a second emitter 316b, a first detector 317a, and a second detector 317b. The sensing device 310" may be configured to measure accelerometer data and cardiac data such as a PPG signal. In some embodiments, the processor 312 may be configured to select between the emitters 316a, 316b and detectors 317a, 317b to provide for different sensing path lengths, intensities, wavelengths, etc. For example, selection of the first emitter 316a and first detector 317a as a first emitter / detector pair corresponds to a first sensing distance at a first location, selection of the second emitter 316b and the second detector 317b as a second emitter / detector pair corresponds to a second sensing distance at a second location, and selection of the first emitter 316a and the second detector 317b as a third emitter / detector pair corresponds to a third sensing distance at a third location where the third sensing distance is longer than the first sensing distance. Accordingly, emitter / detector pair selection may modify a path length and an angle of incidence / reflection relative to the tissue T. In some embodiments, a longer sensing distance may be configured for thicker tissue and a shorter sensing distance may be configured for thinner tissue. As another example, in some embodiments, the first emitter 316a and the second emitter 316b may be configured to emit different wavelengths and / or intensities of light. Therefore, selecting between the two emitters may enable different sensing capabilities. For example, a first wavelength of light may be better suited for use with different thicknesses of skin and / or skin tone compared to a second wavelength of light.

[0062] The emitters 316a, 316b may be configured to output respective light emissions 320a, 320b into the tissue T towards the vessel V. The detectors 317a, 317b may be configured to receive respective reflections 322a, 322b (e.g., light reflection) from the vessel V and through the tissue T. The processor may be configured to process one or more of the received reflections 322a, 322b as measured cardiac data (e.g., PPG signal). In some embodiments, a plurality of signals may be measured to obtain a stronger signal and / or provided for signal validation.10063] FIG. 6 depicts a sensing device 800, similar to sensing devices 110, 210, 310, 310', 310" described herein. In some embodiments, the sensing device 810 may include a housing 810 (e.g., enclosure, pod), a first emitter 816a, a second emitter 816b, a third emitter 816c, a fourth emitter 816d, a first detector 817a, a second detector 817b, and a light barrier 818. In some embodiments, the emitters 816a-816d may comprise one or more of a light emitter, LED, and diode. For example, the first emitter 816a and the fourth emitter 816d may be configured to output a first light having a first wavelength (e.g., red or near red), and the second emitter 816b and the third emitter 816c may be configured to output a second light having a second wavelength (e.g., green or near green). That is, the first emitter 816a and the fourth emitter 816d may comprise a first pair, and the second emitter 816b and the third emitter 816c may comprise a second pair.

[0064] In some embodiments, the detectors 817a-817b may comprise one or more of a light detector, photodiode, and phototransistor. In some embodiments, the light barrier 818 may be configured to reduce light leakage from the emitters 816a-816d and / or ambient or external light from affecting the detectors 817a-817b. That is, the light barrier 818 can be configured to allow light from the emitters 816a-816d to travel into the skin without leaking out, and to be reflected and captured by the detectors 817a-817b. The light barrier 818 can also prevent external light from affecting the reflected signal captured by the detectors 817a-817b. In some embodiments, the light barrier 818 may protrude from a surface of the housing 810 further than the emitters 816a-816d and the detectors 817a-817b. In other embodiments, the light barrier 818 may be flush with a surface of the housing that is designed to be placed against a patient’s skin surface.

[0065] In some embodiments, each of the emitters 816a-816d, detectors 817a-817b, and light barrier 818 may be disposed on a surface of the housing 810 configured to be placed over tissue (e.g., skin) of a user. For example, the sensing device 800 may be a wearable device worn around a wrist.

[0066] FIG. 7 A depicts a sensing device 900, similar to sensing devices 110, 210, 310, 310', 310", 800 described herein. In some embodiments, the sensing device 900 may include a housing 910 (e.g., enclosure, pod), a first emitter 919a, a second emitter 919b, a third emitter 919c, a first detector 917a, a second detector 917b, a third detector 917c, a fourth detector 917d, and a light barrier 918.

[0067] In some embodiments, each emitter 919a-919c may comprise one or more of a light source, LED, and / or diode, e.g., as depicted in FIG. 7B. For example, each emitter 919a-919c may include a plurality of light sources DI, D2, D3, where each of DI, D2, D3 emits a different type of light (e.g., a different wavelength of light). In use, the emitters 919a-919c may be configured to output light having the same or different wavelength. For example, the first emitter 919a may be configured to output a first light having a first wavelength, the second emitter 919b may be configured to output a second light having a second wavelength, and the third emitter 919c may be configured to output a third light having a third wavelength, the three wavelengths being the same or different. Additionally or alternatively, the emitters 919a-919c may be configured to emit light together or independently. FIG. 7B is a schematic diagram of light emitters 919a-919c of the sensing device 900. Each emitter 919a-919c may comprise a corresponding lens 916a-916c configured to focus light emitted by the emitter 919a-919c.10068 [ In some embodiments, the detectors 917a-917c may comprise one or more of a light detector, photodiode, phototransistor. In some embodiments, the light barrier 918 may be configured to reduce light leakage from the emitters 917a-917c and / or external light from impacting the readings of the detectors 917a-917d. That is, the light barrier 918 can be configured to limit light leakage that may impact the accuracy of the light detected by the detectors 917a-917d. In some embodiments, the light barrier 918 may be disposed between adjacent emitters and detectors. Light emission from the first emitter 919a to the first detector 917a may have a first path length and a first sensing location while light emission from the third emitter 919c to the third detector 917c may have a second path length (longer than the first path length) and a second sensing location different from the first sensing location. In this manner, a plurality of signals corresponding to the plurality of path lengths may be measured to obtain a stronger signal and / or provided for signal validation. In some embodiments, the plurality of signals can be averaged and / or selected with a smart channel selection algorithm (e.g., channel selection based on one or more inputs), e.g., to reduce noise and / or other artifacts.|0069| In some embodiments, each of the emitters 919a-919c, detectors 917a-917b, and light barrier 918 may be disposed on a surface of the housing 910 configured to be placed overtissue (e.g., skin) of a user. For example, the sensing device 900 may be a wearable device worn around a wrist.

[0070] FIG. 8 is a schematic diagram of a sensing device 1000 similar to sensing device 900 described herein. In some embodiments, the sensing device 1000 may include a first emitter 1016a, a second emitter 1016b, a third emitter 1016c, a first detector 1017a, a second detector 1017b, a third detector 1017c, and a fourth detector 1017d. The first emitter 1016a paired with the first detector 1017a may correspond to a first path length P1002, the first emitter 1016a paired with the second detector 1017b may correspond to a second path length Pl 004, the first emitter 1016a paired with the third detector 1017c may correspond to a third path length Pl 006, and the first emitter 1016a paired with the fourth detector 1017d may correspond to a fourth path length Pl 008. Each of the path lengths P1002-P1008 may be independent in that they correspond to different sensing distances and / or locations. For example, all four path lengths P1002-P1008 may correspond to different sensing locations. The first and third path lengths may be about equal to each other, the second and fourth path lengths may be about equal to each other, and the second and fourth path lengths may be greater than the first and third path lengths. Accordingly, the position of the emitters 1016a-1016c relative to the detectors 1017a-1017d may correspond to a plurality of path lengths that enables a sensing distance / location to be adjusted. As described above, this may be beneficial for capturing PPG signals at different locations (e.g., to avoid scars or blocked areas that may prevent accurate PPG signal capture) and / or capture PPG signals at different tissue depths (e.g., with longer path lengths).II. Methods

[0071] Also described here are methods for non-invasively monitoring a patient using the systems and devices described herein (e.g., sensing device 110, compute device 120, etc.). In particular, the systems, devices, and methods described herein may be used to calculate a metric such as a cardiac parameter. For example, the sensing systems and devices described herein enable non-medical professionals (e.g., patients) to continuously or semi-continuously, non- invasively, and remotely determine one or more metrics without interrupting their daily activities. The devices, systems, and methods may be easy for a patient to use and require little training, and provide a comfortable and portable way to determine and track health status. Moreover, the sensing device may remain comfortably and continuously wearable for several 1days to several weeks without interfering in a patient’s activities and without significant upkeep.

[0072] Generally, cardiac data and accelerometer data may be processed and analyzed to determine a biomarker or physiological parameter of a user. For example, a sensing device such as sensing device 110 can be configured to measure accelerometer data and PPG data of a patient and provide that data to a processor (e.g., processor 112, processor 122) for further processing and / or analysis. In some embodiments, the sensing device can be configured to measure data associated with a plurality of biomarkers. In some embodiments, the biomarker or physiological parameter of interest is heart rate or pulse rate. In some embodiments, the methods for determining heart rate can include receiving cardiac data (e.g., PPG data) and accelerometer data (e.g., acceleration data) of the patient using a non-invasive patient measurement device, determining if movement is present and / or classifying a type of movement, optionally applying motion artifact removal, optionally generating a probability distribution indicative of likely values of the user’s heart rate, and then determining the user’s heart rate using a machine learning algorithm. Any of the system and devices described herein may be used in the methods described herein.

[0073] FIG. 4 depicts a flow 400 for determining a heart rate or pulse rate of a user, according to embodiments. The flow 400 can be executed by one or more compute devices (e.g., structurally and / or functionally similar to sensing device 110, compute device 120, etc.). The flow 400 can operate continuously, periodically, or sporadically on the compute device to determine the user’s heart rate information. The flow 400 may enable more robust determination of heart rate when a user is moving or engaging in high intensity activity, as the flow 400 takes into consideration various movement data in determining heart rate.

[0074] At step 402, the sensing device captures input data. In the example flow depicted, the input data can include acceleration data 404 and PPG data 406. In other embodiments, additional types of sensor data or other patient information may also be captured and / or received. In some embodiments, the acceleration data 404 can include information related to the motion and / or position of the patient such as acceleration, movement, posture, orientation, and / or the like. The PPG data 406 can include information related to cardiac information of the user. In some embodiments, the PPG data 406 can include green wavelength PPG data and / or red wavelength PPG data. In some embodiments, the sensing device can continuously and / or repeatedly at various times capture acceleration data 404 and / or PPG data 406. In some embodiments, the sensing device may capture acceleration data 404 and / or PPG data 406 basedon an input from a user (e.g., instructing the device to capture data for determining heart rate). In greater detail, the acceleration data 404 and / or the PPG data 406 may be measured using one or more sensing device(s), e.g., sensing device(s) 110, 210, 310, 310', 310", 800, 900. The acceleration data 404 and the PPG data 406 may be measured in the same time period(s). In some embodiments, the set time period may be between about 5 seconds and about one week, including all sub-values in-between. For example, the set time period may be up to about 1 minute, up to about 5 minutes, up to about 10 minutes, up to about 30 minutes, up to about 60 minutes, up to about 2 hours, up to about 6 hours, up to about 12 hours, up to about 24 hours, up to about 36 hours, up to about 48 hours, up to about 72 hours, up to about 96 hours, up to about 120 hours, and up to about 1 week.

[0075] In some embodiments, at step 408, the compute device can determine whether the PPG data is saturated. For example, the compute device can determine that the PPG data is saturated low or high, e.g., based on whether the data hits a low saturation value or a high saturation value. In some embodiments, the compute device may include processing circuitry and / or implement signal processing algorithms to remove saturation from sensor signals. For example, processing circuitry for removing saturation from biometric signals such as electrodermal activity are described in U.S. Patent No. 10,506,944, filed March 17, 2014, titled “Apparatus for electrodermal activity measurement with current compensation,” the disclosure of which is incorporated herein by reference.

[0076] At step 410, the acceleration data 404 is divided into fixed time windows. Similarly, at 412, the PPG data 406 is divided into fixed time windows. In some embodiments, the window is between about 0.1 second and about 30 seconds, including all ranges and / or values therebetween, including, for example, between about 5 seconds and about 15 seconds.

[0077] After the acceleration data 404 and the PPG data 406 are divided into windows, the acceleration data 404 and the PPG data 406 is preprocessed at steps 414 and 416, respectively. As described above with reference to FIG. IB, pre-processing can include filtering, smoothing, amplification, data rejection, and / or removing portions of data that are not useful. In some embodiments, pre-processing may include applying one or more band-pass filters to the acceleration data 404 and / or the PPG data 406. For example, certain predetermined ranges of values may be associated with the acceleration data 404 and / or the PPG data 406, and therefore values outside of those ranges may be filtered out or rejected.

[0078] In some embodiments, the acceleration data 404 and / or the PPG data 406 can be processed using motion artifact removal, at step 418. As described above, motion artifact removal can include applying adaptive filtering algorithms, e.g., such as those described in P.C.T. Patent Application No. PCT / US2024 / 039838, incorporated above by reference. The motion artifact removal can remove artifacts or noise in the data due to motion of the user, e.g., based on activity, gesturing, etc. In some embodiments, the motion artifact removal is applied when it is determined that the user is engaging in movement (e.g., when the user is exercising and / or moving) and not when the user is not engaging in movement (e.g., when the user is at rest). For example, in some embodiments, the compute device may first determine whether there is motion or movement, e.g., based on the acceleration data, and then when there is motion, apply the motion artifact removal to the acceleration data 404 and / or the PPG data 406. In some embodiments, depending on whether movement is present, the acceleration data and PPG data with or without motion artifact removal is used to determine heart rate, as further described below.

[0079] At step 420, the movement of the patient is classified, e.g., based on acceleration data. In some embodiments, the user’s movement can be classified based on raw acceleration data. In some embodiments, the user’s movement can be classified based on pre-processed acceleration data. In some embodiments, both raw and pre-processed acceleration data may be used to classify the user’s movement. Including both raw and pre-processed acceleration data 404 may enable more robust determination of movement. For example, in some embodiments, pre-processing at step 414 may have removed certain information in the acceleration data such as data indicative of posture of the patient. Including the raw acceleration data may therefore provide an additional dimension of information that can be useful for classifying movement. As described above with reference to FIG. IB, in some embodiments, movement classification can include implementing a cascading decision tree. In some embodiments, the decision tree can include determining, at a first decision node, if motion is present. If motion is present, the decision tree can continue to determining if the motion is locomotion (e.g., patient moving around) or gesturing (e.g., patient communicating with hands, etc.). If locomotion is determined, then the decision tree can continue to determining what type of locomotion is present. In some embodiments, movement classification uses a machine learning model to determine a type of movement or activity based on the acceleration data 404.

[0080] In some embodiments, if movement is present (422: YES), then the acceleration data 404 and the PPG data 406 after pre-processing and motion artifact removal is furtherprocessed and analyzed to determine the user’s heart rate. If movement is not present (422: NO), then the acceleration data 404 and the PPG data 406 without motion artifact removal is further processed and analyzed to determine the user’s heart rate. As described above, this can avoid over-filtering of the data, e.g., due to the motion artifact removal, that may introduce inaccuracy in the heart rate detection.

[0081] Additionally, the determination of whether there is movement and / or the type of movement or activity can be used for transition probability estimation, at step 426. At step 426, the preprocessed PPG data 406, the preprocessed acceleration data 404, and the classified movement can be used to generate a probability distribution, as described above with reference to FIGS. IB and 5A-5D. The probability distribution can set the likelihood that a particular value is determined to be the user’s heart rate. The probability distribution can be input into a model that is used to determine the heart rate of the user, and impact the output of the model. For example, values associated with lower probabilities in the probability distribution may be assigned a lower weight than values associated with higher probabilities in the probability distribution. In some embodiments, the compute device can be configured to set one or more parameters of the probability distribution, e.g., skewness, standard deviation, center or peak value, etc., based on one or more factors including: (1) the type of movement, (2) the intensity of movement, (3) a noise level in the acceleration data, (4) a noise level in the PPG data, (5) clinical history and / or demographic information associated with the user, etc. In some embodiments, the compute device can be configured to generate the probability distribution based on one or more predetermined rules associated with these factors. In other embodiments, the compute device can be configured to implement a trained model that is configured to output a probability distribution based on receiving these factors as inputs.

[0082] Optionally, at step 428, the compute device can apply one or more heart rate or pulse rate change thresholds to the probability distribution, at step 428. For example, the compute device may apply a lower bound or a higher bound threshold to the probability distribution to prevent determination of heart rate values outside of a predetermined range set by the thresholds. This can be used to prevent spikes and / or unexpected values from being determined. Such spikes and / or unexpected values may result from improper sensor placement or wear, a broken and / or faulty sensor, water interference, etc.

[0083] In some embodiments, if no movement is detected, the probability distribution may also be reset to a predetermined default distribution, at step 430. In particular, the transition probability estimation may be reset in the case where the probability estimation includesaspects (e.g., skewness) that may falsely represent whether there should be a higher or lower heart rate. Therefore, at step 430, the compute device can reset the probability distribution to prevent a misidentified motion from affecting the pulse rate estimation. If movement is detected, then no reset may be necessary at step 430.[00841 At step 424, the acceleration data 404 and the PPG data 406 can be transformed into the spectral domain, e.g., using a Fourier transform. At step 434, power spectrum analysis may be performed on the data, while also taking into the account the probability distribution. As described with reference to FIG. IB, the acceleration data, the PPG data, and the probability distribution can be input into a machine learning model. The machine learning model can then process the data to determine a heart rate value for the user. In some embodiments, the machine learning model can be configured to perform power spectrum analysis based on a summation of power spectrum analyses of the most recent window of data as well as previous windows of data. The power spectrum analysis then output a frequency plot that shows the summed power at each frequency. The frequency with the highest power can then be used to determine the heart rate of the user for the current window of time.

[0085] At step 436, the heart rate determined for multiple windows falling within a larger time period can be aggregated to generate the heart rate or PR biomarker 438. In some embodiments, the larger period of time can be on the order of a minute or minutes (e.g., between about 1 minute and about several minutes). In some embodiments, the aggregation at 436 can include finding a mean, median, or mode over the larger period of time. In some embodiments, the aggregation at 436 can use a statistical model to generate the PR biomarker 438.

[0086] Additionally or alternatively, a notification may be provided to one or more users (e.g., patient, health care professional) based on a determined metric, physiological parameter, or biomarker. In some embodiments, one or more notifications (e.g., alerts) may be provided to a predetermined set of contacts (e.g., family, partner, caregiver, health care professional) based on predetermined criteria. The set of contacts may be selected, for example, by the patient or a caregiver, and may receive notifications or other communications via one or more of a telephone call, email, text message, push notification on a mobile device, web portal, and the like. In some embodiments, the notification may include one or more of the calculated metric and confidence level of the calculated metric. In some embodiments, one or more of frequency, timing, and content of patient notification may be modified based on patient data such as demographic data. For example, patients who are in a high-risk demographic group (e.g.,elderly, pre-existing condition, socioeconomic status) may receive a patient infection notification more frequently than a patient in a low-risk group (e.g., young, healthy).

[0087] Additionally or alternatively, in some embodiments, a metric, physiological parameter, or biomarker of a patient may be utilized to remotely monitor and / or manage a patient. For example, health care professionals (e.g., primary care physicians) may use the information to adjust a medication regimen, clinical assessments, and / or to inform therapy decision-making. In some embodiments, a health care professional (e.g., care provider) may be provided access to the patient data via a graphical user interface on one or more compute devices such as through a browser-based web access portal. The health care professional may, for instance, log in to the browser-based web access portal via a personal computer. The health care professional could review all of the monitored data from one or more patients. The health care professional could additionally input lab test results, notes from patient appointments, patient therapy changes, patient infections and complications, and any other findings into the patient’s health record.

[0088] The systems, devices, and / or methods described herein may be performed by software (executed on hardware), hardware, or a combination thereof. Hardware modules may include, for example, a general-purpose processor (or microprocessor or microcontroller), a field programmable gate array (FPGA), and / or an application specific integrated circuit (ASIC). Software modules (executed on hardware) may be expressed in a variety of software languages (e.g., computer code), including C, C++, Java®, Python, Ruby, Visual Basic®, and / or other object-oriented, procedural, or other programming language and development tools. Examples of computer code include, but are not limited to, micro-code or microinstructions, machine instructions, such as produced by a compiler, code used to produce a web service, and files containing higher-level instructions that are executed by a computer using an interpreter. Additional examples of computer code include, but are not limited to, control signals, encrypted code, and compressed code.

[0089] It should be understood that the disclosed embodiments are not intended to be exhaustive, and functional, logical, operational, organizational, structural and / or topological modifications can be made without departing from the scope of the disclosure. As such, all examples and / or embodiments are deemed to be non-limiting throughout this disclosure.

[0090] All definitions, as defined and used herein, should be understood to control over dictionary definitions, definitions in documents incorporated by reference, and / or ordinary meanings of the defined terms.[00911 Examples of computer code include, but are not limited to, micro-code or microinstructions, machine instructions, such as produced by a compiler, code used to produce a web service, and files containing higher-level instructions that are executed by a computer using an interpreter. For example, embodiments can be implemented using Python, Java, JavaScript, C++, and / or other programming languages and development tools. Additional examples of computer code include, but are not limited to, control signals, encrypted code, and compressed code.

[0092] The drawings primarily are for illustrative purposes and are not intended to limit the scope of the subject matter described herein. The drawings are not necessarily to scale; in some instances, various aspects of the subject matter disclosed herein can be shown exaggerated or enlarged in the drawings to facilitate an understanding of different features. In the drawings, like reference characters generally refer to like features (e.g., functionally similar and / or structurally similar elements).

[0093] The acts performed as part of a disclosed method(s) can be ordered in any suitable way. Accordingly, embodiments can be constructed in which processes or steps are executed in an order different than illustrated, which can include performing some steps or processes simultaneously, even though shown as sequential acts in illustrative embodiments. Put differently, it is to be understood that such features can not necessarily be limited to a particular order of execution, but rather, any number of threads, processes, services, servers, and / or the like that can execute serially, asynchronously, concurrently, in parallel, simultaneously, synchronously, and / or the like in a manner consistent with the disclosure. As such, some of these features can be mutually contradictory, in that they cannot be simultaneously present in a single embodiment. Similarly, some features are applicable to one aspect of the innovations, and inapplicable to others.|0094| Where a range of values is provided, it is understood that each intervening value, to the tenth of the unit of the lower limit unless the context clearly dictates otherwise, between the upper and lower limit of that range and any other stated or intervening value in that stated range is encompassed within the disclosure. That the upper and lower limits of these smaller ranges can independently be included in the smaller ranges is also encompassed within thedisclosure, subject to any specifically excluded limit in the stated range. Where the stated range includes one or both of the limits, ranges excluding either or both of those included limits are also included in the disclosure.

[0095] The phrase “and / or,” as used herein in the specification and in the embodiments, should be understood to mean “either or both” of the elements so conjoined, i.e., elements that are conjunctively present in some cases and disjunctively present in other cases. Multiple elements listed with “and / or” should be construed in the same fashion, i.e., “one or more” of the elements so conjoined. Other elements can optionally be present other than the elements specifically identified by the “and / or” clause, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, a reference to “A and / or B”, when used in conjunction with open-ended language such as “comprising” can refer, in one embodiment, to A only (optionally including elements other than B); in another embodiment, to B only (optionally including elements other than A); in yet another embodiment, to both A and B (optionally including other elements); etc.

[0096] As used herein in the specification and in the embodiments, “or” should be understood to have the same meaning as “and / or” as defined above. For example, when separating items in a list, “or” or “and / or” shall be interpreted as being inclusive, i.e., the inclusion of at least one, but also including more than one of a number or list of elements, and, optionally, additional unlisted items. Only terms clearly indicated to the contrary, such as “only one of’ or “exactly one of,” or, when used in the embodiments, “consisting of,” will refer to the inclusion of exactly one element of a number or list of elements. In general, the term “or” as used herein shall only be interpreted as indicating exclusive alternatives (i.e., “one or the other but not both”) when preceded by terms of exclusivity, such as “either,” “one of,” “only one of,” or “exactly one of.” “Consisting essentially of,” when used in the embodiments, shall have its ordinary meaning as used in the field of patent law.

[0097] As used herein in the specification and in the embodiments, the phrase “at least one,” in reference to a list of one or more elements, should be understood to mean at least one element selected from any one or more of the elements in the list of elements, but not necessarily including at least one of each and every element specifically listed within the list of elements and not excluding any combinations of elements in the list of elements. This definition also allows that elements can optionally be present other than the elements specifically identified within the list of elements to which the phrase “at least one” refers, whether related or unrelated to those elements specifically identified. Thus, as a non-limitingexample, “at least one of A and B” (or, equivalently, “at least one of A or B,” or, equivalently “at least one of A and / or B”) can refer, in one embodiment, to at least one, optionally including more than one, A, with no B present (and optionally including elements other than B); in another embodiment, to at least one, optionally including more than one, B, with no A present (and optionally including elements other than A); in yet another embodiment, to at least one, optionally including more than one, A, and at least one, optionally including more than one, B (and optionally including other elements); etc.

[0098] In the embodiments, as well as in the specification above, all transitional phrases such as “comprising,” “including,” “carrying,” “having,” “containing,” “involving,” “holding,” “composed of,” and the like are to be understood to be open-ended, i.e., to mean including but not limited to. Only the transitional phrases “consisting of’ and “consisting essentially of’ shall be closed or semi-closed transitional phrases, respectively, as set forth in the United States Patent Office Manual of Patent Examining Procedures, Section 2111.03.

[0099] The specific examples and descriptions herein are exemplary in nature and embodiments may be developed by those skilled in the art based on the material taught herein without departing from the scope of the present invention, which is limited only by the attached claims.

Claims

CLAIMSWe claim:

1. An apparatus, comprising: a photoplethysmography (PPG) sensor configured to measure a PPG signal associated with a user; an accelerometer configured to measure an acceleration signal associated with the user; a processor operatively coupled to the PPG sensor and the accelerometer, the processor configured to: determine whether the user is engaging in movement based on the acceleration signal; in response to determining that the user is engaging in movement, determine a type of movement of the user based on the acceleration signal; generate a probability distribution associated with a pulse rate of the user based on the type of movement; and determine the pulse rate of the user based on the PPG signal, the acceleration signal, and the probability distribution.

2. The apparatus of claim 1, wherein the processor is configured to determine the pulse rate of the user by inputting the PPG signal, the acceleration signal, and the probability distribution into a model to obtain the pulse rate of the user3. The apparatus of claim 2, wherein the PPG signal and the acceleration signal are measured over a current time period, and the pulse rate determined by the processor is for the current time period, the model configured to output a frequency plot corresponding to a summation of frequency components of the PPG signal measured over the current time period and of one or more PPG signals measured over time periods prior to the current time period.

4. The apparatus of claim 1, wherein the processor is configured to apply an adaptive filter algorithm to the acceleration signal and the PPG signal prior to using the PPG signal and the acceleration signal to determine the pulse rate of the user.

5. The apparatus of claim 4, wherein the processor is configured to apply the adaptive filter algorithm to the acceleration signal and the PPG signal in response to determining that the user is engaging in movement.

6. The apparatus of claim 1, wherein the type of movement includes at least one of locomotion, walking, running, or gesturing.

7. The apparatus of claim 1, wherein the PPG signal and the acceleration signal are measured over a current time period, and the pulse rate determined by the processor is for the current time period, wherein the probability distribution includes a peak that is centered at the pulse rate determined for the user for a time period prior to the current time period.

8. A method, comprising: receiving a photoplethysmography (PPG) signal associated with a user from a PPG sensor; receiving an acceleration signal associated with the user from an accelerometer; determining whether the user is engaging in movement based on the acceleration signal; in response to determining that the user is engaging in movement, determining a type of movement of the user based on the acceleration signal; and determining, using a machine learning model, a pulse rate of the user based on a power spectrum analysis of the PPG signal and the acceleration signal.

9. The method of claim 8, further comprising: generating a probability distribution associated with the pulse rate of the user based on the type of movement, wherein determining the power spectrum analysis is based on the probability distribution.

10. The method of claim 8, further comprising: applying an adaptive filter algorithm to the acceleration signal and the PPG signal prior to using the PPG signal and the acceleration signal to determine the pulse rate of the user.

11. The method of claim 10, wherein the adaptive filter algorithm is applied to the acceleration signal and the PPG signal in response to determining that the user is engaging in movement.

12. The method of claim 10, wherein the adaptive filter algorithm is configured to remove motion artifacts from the PPG signal.

13. The method of claim 8, wherein the type of movement includes at least one of locomotion, walking, running, or gesturing.

14. The method of claim 8, wherein the PPG signal and the acceleration signal are measured over a current time period, and the pulse rate is for the current time period.

15. A method, comprising: receiving a photoplethysmography (PPG) signal associated with a user from a PPG sensor, the PPG signal including data measured over a plurality of time windows; receiving an acceleration signal associated with the user from an accelerometer, the PPG signal including data measured over the plurality of time windows; determining whether the user is engaging in movement based on the acceleration signal; in response to determining that the user is engaging in movement, determining a type of movement of the user based on the acceleration signal; determining, for each time window of the plurality of time windows, a pulse rate of the user based on the PPG signal and the acceleration signal; and determining, based on the pulse rate for each time window of the plurality of time windows, a pulse rate biomarker.

16. The method of claim 15, wherein determining the pulse rate biomarker can include determining at least one of a mean, a median, or a mode of the pulse rate for each time window of the plurality of time windows.

17. The method of claim 15, further comprising: determining if the pulse rate biomarker is outside of a predetermined range.

18. The method of claim 17, further comprising: based on determining that the pulse rate biomarker is outside of a predetermined range, sending, to a user device, an alert.

19. The method of claim 15, further comprising: determining a power spectrum analysis for the PPG signal and the acceleration signal, wherein determining the pulse rate is based on the power spectrum analysis.

20. The method of claim 19, wherein determining the pulse rate is based on a highest power frequency associated with the power spectrum analysis.

Images