Multi-parameter cuffless blood pressure monitoring system

Abstract

Techniques for cuffless blood pressure monitoring with multi-parameter correction are described herein and can be implemented to reduce measurement inaccuracies in wearable cuffless blood pressure devices. In an example, a wearable device includes a sensor arrangement to acquire physiological time-series data indicative of pulse transit time along a cardiovascular pathway and a correction sensor configured to measure a correction parameter that independently affects pulse transit time from corresponding blood pressure changes, such as body temperature, skin temperature, perfusion index, hydration level, or muscle activity. A processor of the wearable device is configured to process the physiological time-series data to determine pulse transit time and generate a blood pressure measurement based on the pulse transit time and the measured value of the correction parameter. Thus, the techniques described herein generate blood pressure measurements that account for physiological variables that have independent effects on pulse transit characteristics.

Description

[0001] Cross-references to related applications

[0002] This application claims priority to U.S. Application No. 63 / 703,665, filed October 4, 2024, entitled “Cuffless Blood Pressure Monitor with Additional Sensors for Calibration,” and U.S. Nonprovisional Application No. 19 / 341,842, filed September 26, 2025, entitled “Multi-parameter Cuffless Blood Pressure Monitoring System,” the entire contents of which are incorporated herein by reference. Background Technology

[0003] Blood pressure measurement can provide important insights into an individual's cardiovascular health. Traditionally, blood pressure is measured using an inflatable cuff that temporarily blocks blood flow in the arteries and detects pressure oscillations or sounds that can be correlated with blood pressure values ​​when the cuff deflates. While these devices provide accurate measurements, conventional cuff-based devices are often unsuitable for continuous monitoring in mobile settings due to the obstructive inflation / deflation cycle, their bulky form factor limiting wearability, and data gaps caused by intermittent measurements. Attached Figure Description

[0004] Figure 1 This is a block diagram of a non-limiting example of an environment operable with the multi-parameter cuffless blood pressure monitoring system and technology described herein.

[0005] Figure 2 A non-limiting example of a monitoring device is shown.

[0006] Figure 3 A non-limiting system is shown in an example implementation of a multi-parameter cuffless blood pressure monitoring system, which illustrates the operation of the blood pressure monitoring system in more detail.

[0007] Figure 4 A non-limiting example of multi-parameter cuffless blood pressure monitoring is shown, in which a wearable device generates blood pressure measurements based on time-series data and one or more measured calibration parameters.

[0008] Figure 5 A non-limiting example of multi-parameter cuffless blood pressure monitoring is shown, in which a temperature sensor is used to acquire skin temperature data as a calibration parameter.

[0009] Figure 6 Additional, non-limiting examples of multi-parameter cuffless blood pressure monitoring are shown, in which a temperature sensor is used to acquire skin temperature data as a calibration parameter.

[0010] Figure 7A non-limiting example of multi-parameter cuffless blood pressure monitoring is shown, in which a heat flux sensor is used to acquire body temperature data as a calibration parameter.

[0011] Figure 8 A non-limiting example of multi-parameter cuffless blood pressure monitoring is shown, in which a PPG sensor is used to acquire perfusion index data as a correction parameter.

[0012] Figure 9 A non-limiting example of multi-parameter cuffless blood pressure monitoring is shown, in which a bioimpedance sensor is used to acquire hydration level data as a correction parameter.

[0013] Figure 10 A non-limiting example of multi-parameter cuffless blood pressure monitoring is shown, in which an EMG sensor is used to acquire muscle activity data as a calibration parameter.

[0014] Figure 11 A non-limiting example of multi-parameter cuffless blood pressure monitoring is shown, in which multiple calibration parameters are used to generate blood pressure measurements.

[0015] Figure 12 A non-limiting example of multi-parameter cuffless blood pressure monitoring is shown, with a user interface for a blood pressure monitoring scenario illustrated.

[0016] Figure 13 A flowchart is shown, illustrating an algorithm for generating blood pressure using multi-parameter cuffless blood pressure monitoring technology as a step-by-step process in an example implementation. Detailed Implementation

[0017] While cuff-based blood pressure monitors are accurate and suitable for single measurements, they often face practical limitations that prevent their use in continuous health monitoring applications. Therefore, the development of cuffless blood pressure monitoring technology has emerged as an alternative for providing continuous, non-invasive measurements. Cuffless systems aim to estimate blood pressure using physiological signals rather than relying on mechanical blockage of blood vessels. For example, blood pressure can be estimated using pulse wave characteristics such as pulse conduction time (“PTT”) or pulse arrival time (“PAT”) based on the relationship between arterial stiffness and pulse wave velocity.

[0018] However, cuffless blood pressure monitoring systems can experience phenomena such as calibration "drift" and become inaccurate over time, requiring frequent recalibration against traditional cuff-based measurements to maintain accuracy. For example, conventional cuffless systems are susceptible to various physiological and environmental factors, as well as inter-individual variability, which affect pulse characteristics independently of blood pressure, thus reducing measurement reliability. These limitations can impair long-term accuracy and negate the advantages of conventional cuffless monitoring devices.

[0019] Therefore, this paper describes a multi-parameter cuffless blood pressure monitoring system that addresses these limitations by measuring and taking into account physiological and / or environmental factors that may cause calibration drift. For example, the system includes a wearable device with a sensor arrangement configured to acquire physiological temporal data indicating pulse wave characteristics (such as pulse propagation time, e.g., pulse conduction time (“PTT”) and / or pulse arrival time (“PAT”)) along the cardiovascular pathway between proximal and distal measurement points.

[0020] In this example, the sensor arrangement includes a proximal sensor (such as an electrocardiogram (“ECG”) sensor or a bioimpedance sensor) positioned to detect a first cardiovascular signal and a distal sensor (such as a photoplethysmography (“PPG”) sensor) positioned at a distal location to detect a corresponding second cardiovascular signal. The wearable device utilizes a processor to determine the pulse propagation time by detecting the time difference between the first and second cardiovascular signals; this pulse propagation time represents the duration of the pulse wave's journey along the cardiovascular pathway. The pulse propagation time can be correlated with blood pressure through an established relationship between arterial stiffness and pulse wave velocity. For example, relatively higher blood pressure may correspond to increased arterial stiffness and faster pulse wave propagation, resulting in a shorter propagation time.

[0021] However, environmental and / or physiological factors such as body temperature, skin temperature, hydration status, and / or muscle activity can affect pulse wave velocity, partially or entirely independently of corresponding blood pressure changes. This can lead to calibration drift and / or measurement inaccuracies when using conventional systems. To account for these factors, wearable devices also include one or more calibration sensors configured to measure calibration parameters that affect pulse propagation time independently of corresponding blood pressure changes. Therefore, wearable devices can generate accurate blood pressure measurements based on physiological time-series data and the measured values ​​of the calibration parameters.

[0022] In various embodiments, the calibration sensor may include one or more temperature sensors and / or heat flux sensors configured to measure body temperature and / or skin temperature to account for the effects of vasomotor (e.g., vasodilation or vasoconstriction) and / or systemic cardiovascular changes. Additionally or alternatively, the calibration sensor may include: a bioimpedance sensor configured to measure hydration levels to account for changes in arterial stiffness; a PPG sensor for calculating a perfusion index based on the ratio of pulsatile to non-pulsatile blood flow; and / or an electromyography (“EMG”) sensor for measuring muscle activity by detecting electrical signals from smooth muscle tissue.

[0023] The processor is configured to process physiological time-series data to determine pulse propagation time, such as PTT or PAT, and generate a blood pressure measurement based on both the pulse propagation time and measurements of one or more correction parameters. For example, the processor may generate an initial blood pressure measurement based on the pulse propagation time and apply correction factors based on measurements of one or more correction parameters to generate an adjusted blood pressure measurement.

[0024] In various examples, the system continuously integrates correction factors into blood pressure calculations, selectively activates corrections based on threshold conditions to optimize power consumption, and / or performs post-processing analysis on the acquired data based on the correction factors. The system is also operable to implement weighted schemes for multiple correction parameters, utilize various blood pressure measurement algorithms, and / or utilize machine learning models trained to calibrate blood pressure readings based on various correction parameters. This is merely an example and not a limitation, and the system can be operated and implemented in various ways to generate blood pressure measurements based on time-series data and measurements of correction parameters.

[0025] In this way, the technologies, devices, and components described herein overcome the limitations of conventional systems by preventing calibration drift in cuffless blood pressure monitoring. For example, a multi-parameter cuffless blood pressure monitoring system can take into account temperature-induced vasoconstriction affecting peripheral pulse wave velocity, changes in arterial stiffness related to hydration status, perfusion changes affecting peripheral circulation, smooth muscle movement affecting arterial wall properties, and so on. Therefore, the system can maintain accuracy over extended periods without manual recalibration, thus supporting continuous blood pressure monitoring and generating accurate cardiovascular health information during prolonged wear.

[0026] In some aspects, the technology described herein relates to a wearable device for cuffless blood pressure monitoring, the wearable device comprising: a sensor arrangement for acquiring physiological time-series data indicating pulse propagation time along the cardiovascular pathway; a calibration sensor configured to measure calibration parameters that affect pulse propagation time independently of corresponding blood pressure changes; and a processor configured to: process the physiological time-series data to determine pulse propagation time; and present blood pressure measurements generated based on the pulse propagation time and the calibration parameters.

[0027] In some respects, the technology described herein relates to a wearable device in which a sensor arrangement includes: a proximal sensor located at a proximal position along the cardiovascular pathway and configured to detect a first cardiovascular signal as part of physiological time-series data; and a distal sensor located at a distal position along the cardiovascular pathway and configured to detect a second cardiovascular signal as part of physiological time-series data and corresponding to the first cardiovascular signal.

[0028] In some aspects, the technology described herein relates to a wearable device in which the proximal sensor includes one or more of a proximal electrocardiogram sensor, a proximal bioimpedance sensor, or a proximal photoplethysmography (PPG) sensor, the distal sensor includes one or more of a distal PPG sensor, a distal pressure sensor, or a distal bioimpedance sensor, and the pulse propagation time includes one or more of pulse conduction time (PTT) or pulse arrival time (PAT).

[0029] In some respects, the technology described herein relates to a wearable device in which the calibration sensor includes one or more of a temperature sensor or a heat flux sensor, the temperature sensor or heat flux sensor being configured to measure skin temperature adjacent to the sensor arrangement as a calibration parameter, and the processor being further configured to generate a blood pressure measurement based on the skin temperature to take into account the effect of vasomotor activity on pulse propagation time.

[0030] In some respects, the technology described herein relates to a wearable device in which the calibration sensor includes one or more of a temperature sensor or a heat flux sensor, the temperature sensor or heat flux sensor being configured to measure body temperature as a calibration parameter, and the processor being further configured to generate a blood pressure measurement based on body temperature to account for the effect of systemic cardiovascular effects on pulse propagation time.

[0031] In some respects, the technology described herein relates to a wearable device in which the calibration sensor includes a bioimpedance sensor configured to measure hydration level as a calibration parameter, and the processor is further configured to generate blood pressure measurements based on the hydration level to account for the effect of arterial stiffness on pulse propagation time.

[0032] In some respects, the technology described herein relates to a wearable device in which the calibration sensor includes a photoplethysmography (PPG) sensor configured to acquire PPG data, and the processor is further configured to: determine the ratio of pulsatile to non-pulsatile blood flow based on the PPG data, calculate a perfusion index as a calibration parameter; and generate a blood pressure measurement based on the perfusion index to account for the influence of peripheral circulation on pulse propagation time.

[0033] In some respects, the technology described herein relates to a wearable device in which the calibration sensor includes an electromyography (EMG) sensor configured to measure muscle activity as a calibration parameter by detecting electrical signals associated with smooth muscle tissue, and the processor is further configured to generate blood pressure measurements based on muscle activity to account for the effect of arterial wall contractility on pulse propagation time.

[0034] In some respects, the technology described herein relates to a wearable device in which a processor is further configured to generate a blood pressure measurement by: calculating an initial blood pressure measurement based on pulse propagation time, and applying a correction factor to the initial blood pressure measurement based on the correction parameter in response to detecting that a correction parameter exceeds a threshold.

[0035] In some respects, the technology described herein relates to a wearable device in which the processor is further configured to determine systolic and diastolic blood pressure values ​​based on one or more characteristics of pulse propagation time and correction parameters, so as to include the systolic and diastolic blood pressure values ​​as part of the blood pressure measurement.

[0036] In some aspects, the technology described herein relates to a method implemented by a processing device, the method comprising: receiving physiological time-series data indicating pulse propagation time along a cardiovascular pathway and measurements of correction parameters affecting pulse propagation time independently of corresponding blood pressure changes; processing the physiological time-series data to determine pulse propagation time based on the time difference between corresponding points in the physiological time-series data; generating a blood pressure measurement based on the pulse propagation time and the measurements of the correction parameters; and outputting the blood pressure measurement.

[0037] In some respects, the techniques described herein relate to a method in which the correction parameters include at least one of body temperature, skin temperature, perfusion index, hydration level, or muscle activity.

[0038] In some respects, the techniques described herein relate to a method in which physiological time-series data includes a first physiological signal and a second physiological signal, and processing the physiological time-series data includes: determining one or more of pulse conduction time (PTT) or pulse arrival time (PAT) as pulse propagation time based on the first physiological signal and the second physiological signal.

[0039] In some respects, the techniques described herein relate to a method in which a first physiological signal includes an electrocardiogram (ECG) signal and a second physiological signal includes a photoplethysmography (PPG) signal.

[0040] In some respects, the technology described herein relates to a method in which generating a blood pressure measurement includes: generating an initial blood pressure measurement based on pulse propagation time, and adjusting the initial blood pressure measurement based on the measurement of a correction parameter.

[0041] In some respects, the techniques described herein relate to a method in which generating a blood pressure measurement includes: in response to detecting that a correction parameter exceeds a threshold, applying a correction factor to the blood pressure measurement based on the measurement of the correction parameter.

[0042] In some aspects, the technology described herein relates to a system for cuffless blood pressure monitoring, the system comprising: a sensor arrangement for acquiring physiological time-series data indicating pulse propagation time along a cardiovascular pathway; one or more calibration sensors configured to measure one or more calibration parameters that affect pulse propagation time independently of corresponding blood pressure changes; and a processor configured to: process the physiological time-series data to determine pulse propagation time; generate a blood pressure measurement based on the pulse propagation time and the measured values ​​of one or more calibration parameters; and output the blood pressure measurement.

[0043] In some aspects, the technology described herein relates to a system in which a sensor arrangement includes: a first sensor configured to measure a first physiological signal via contact with a user's skin surface; and a second sensor configured to measure a second physiological signal via contact with a user's skin surface, and a processor further configured to generate one or more of pulse conduction time (PTT) or pulse arrival time (PAT) as pulse propagation time based on the first and second physiological signals.

[0044] In some respects, the technology described herein relates to a system in which measurements of one or more correction parameters include at least two of body temperature, skin temperature, perfusion index, hydration level, or muscle activity, and a processor is further configured to implement a weighting scheme to apply weights to the measurements of one or more correction parameters to generate a blood pressure measurement.

[0045] In some respects, the technology described herein relates to a system in which a processor is configured to implement one or more of the following to generate blood pressure measurements: a blood pressure calibration algorithm, a machine learning model trained to calibrate blood pressure readings based on calibration parameters, or a calibration curve.

[0046] Figure 1 This is a block diagram of a non-limiting example 100 of an environment operable with the multi-parameter cuffless blood pressure monitoring system and technology described herein. Example 100 includes a person 102 shown wearing monitoring device 104. The environment also includes an analytics platform 106. Analytics platform 106 may be directly connected to monitoring device 104 via one or more wireless connections, or via one or more wired and / or wireless connections and one or more intermediate devices, such as computing devices, network routing devices and apparatuses, server equipment, and / or the Internet associated with person 102.

[0047] Monitoring device 104 can be used to monitor one or more aspects of person 102 to generate measurement values ​​108. For example, in some scenarios, monitoring device 104 can be provided to record the electrical activity of person 102's heart over an observation period, such as lasting several seconds or minutes, or several days, etc. For example, person 102 may have the amplitude of his or her heart's electrical potential that can be monitored over time to generate one or more electrocardiograms (ECGs) that can be used to predict any of a variety of events. In at least one example, monitoring device 104 can be configured to record physiological time-series data (e.g., data indicating one or more pulse wave characteristics) and / or measurements of correction parameters affecting pulse wave characteristics. Alternatively or additionally, monitoring device 104 can be used to output measurement values ​​108 (e.g., time series of measurement values, such as time series of potential measurements) that can indicate observations or be used to generate predictions for one or more events.

[0048] In conjunction with the monitoring device, instructions can be provided to person 102 on how to operate the monitoring device 104 and / or how to behave (e.g., sleep, engage in activities) while wearing the monitoring device 104. In one or more embodiments, these instructions may be provided as part of a kit, such as written instructions. Alternatively or additionally, the analysis platform 106 may enable these instructions to be transmitted and output (e.g., for display and / or audio output) via a computing device associated with person 102. In one or more embodiments, the analysis platform 106 may provide these instructions for output only after a predetermined amount of time (e.g., two days) has elapsed during the observation period of wearing the monitoring device 104 and / or based on patterns in various aspects of the person 102 being monitored.

[0049] Monitoring device 104 can be configured in various ways to monitor one or more aspects of personnel 102. Furthermore, Figure 1 and Figure 2 The external dimensions shown are merely an example, and the external dimensions of the monitoring device 104 may vary in different variations. It should be understood that the monitoring device 104 may be configured with one or more sensors, examples of which include one or more of the following: multiple electrodes (e.g., electrodes that can be placed on a person's skin), an accelerometer, and a pulse oximeter (e.g., for measuring and recording blood oxygen saturation (SpO2) and / or generating a photoplethysmogram of the person 102), etc. Of course, without departing from the described technology, the monitoring device 104 may be configured with any of a variety of sensor types.

[0050] While the monitoring device 104 can be configured in a manner similar to monitoring devices used for clinical monitoring of patients, in one or more embodiments, the monitoring device 104 may be configured differently from devices used for clinical monitoring and / or diagnosis of patients. For example, and not limitingly, the monitoring device 104 may be configured as a ring, watch, patch, and / or band, etc. Alternatively or additionally, the monitoring device 104 may have a similar form factor to a clinical setup but with different functions, such as a function to prevent the wearer from viewing the measurement value 108.

[0051] In one or more embodiments, the monitoring device 104 may be configured to offload the measurements 108 during the observation period. For example, the monitoring device 104 may offload the measurements 108 by transmitting them to an external computing device via a wired or wireless connection, for example, at predetermined time intervals and / or in response to establishing or re-establishing a connection with the computing device. In one or more embodiments, the measurements 108 and / or other data from the monitoring device 104 may be compressed by the monitoring device 104 for wireless transmission, for example, using one or more of a variety of data compression techniques. Compressing sensor data in this way can reduce the battery consumption of the monitoring device 104 during the observation period and is beneficial for wear during assessment of physiological conditions, such as blood pressure-related conditions.

[0052] In cases where monitoring device 104 can be configured to store measurement values ​​108 for the entire observation period, in one or more embodiments, monitoring device 104 can be configured without wireless transmission means, such as no antenna for wirelessly transmitting measurement values ​​108 and no hardware or firmware for generating data packets for such wireless transmission. Alternatively, monitoring device 104 can be configured with hardware for transmitting measurement values ​​108 via physical wired coupling. In this scenario, monitoring device 104 can be "plugged in" to retrieve measurement values ​​108 from the device's storage.

[0053] Therefore, the monitoring device 104 may be configured with one or more ports to enable wired transmission of the measurement value 108 to an external computing device. Examples of such physical coupling may include a micro universal serial bus (USB) connection, a mini USB connection, and a USB-C connection. Although the monitoring device 104 may be configured to extract the measurement value 108 via a wired connection as described above, in different scenarios, the monitoring device 104 may alternatively or additionally be configured to offload the measurement value 108 via one or more wireless connections.

[0054] Once the monitoring device 104 generates a measurement value 108, it provides the measurement value 108 to the analysis platform 106. As described above, the measurement value 108 can be transmitted to the analysis platform 106 via a wired and / or wireless connection.

[0055] For example, in a scenario where the analysis platform 106 is partially or entirely implemented on the monitoring device 104, the measured value 108 can be transferred from the device's local storage to the device's processing system via a bus. In a scenario where the monitoring device 104 is configured to generate one or more predicted values ​​110 by processing the measured value 108, the monitoring device 104 can also be configured to provide the generated one or more predicted values ​​110 as output, for example, by transmitting one or more predicted values ​​110 to an external computing device. In other scenarios, the measured value 108 can be processed by an external computing device configured to generate one or more predicted values ​​110. For example, the measured value 108 can be processed by a smartphone associated with a user, a smartphone or other dedicated device associated with the monitoring device 104, and / or one or more server computers in a data center or other location, which can be used by an entity associated with the monitoring device 104, to name a few. In other words, these other devices can implement at least a portion of the analysis platform 106 and / or the prediction system 114.

[0056] In one or more embodiments, the monitoring device 104 is configured to transmit measurement values ​​108 to an external device via a wired connection, such as via USB-C or some other physical communication coupling. Here, a connector can be inserted into the monitoring device 104, or the monitoring device 104 can be inserted into a device with a socket that engages with a corresponding contact of the device. The measurement values ​​108 can then be retrieved from the storage device of the monitoring device 104 via this wired connection, for example, by transmitting the measurement values ​​108 to the external device via the wired connection. This connection can be used in scenarios where, after an observation period, a person 102 mails the monitoring device 104 to, for example, a healthcare provider, a telemedicine service provider, the provider of the monitoring device 104, or a medical testing laboratory.

[0057] Alternatively or additionally, monitoring device 104 may transmit measurement values ​​108 to analysis platform 106 via one or more wireless connections. For example, monitoring device 104 may wirelessly transmit measurement values ​​108 to external computing devices such as mobile phones, tablets, laptops, smartwatches, other wearable health trackers, etc. Therefore, monitoring device 104 may be configured to communicate with external devices using one or more wireless communication protocols or technologies. For example, monitoring device 104 may communicate with external devices using one or more of Bluetooth (e.g., Bluetooth Low Energy link), Near Field Communication (NFC), or Long Term Evolution (LTE) standards such as 5G. In scenarios where measurement values ​​108 are transmitted to external devices for processing, monitoring device 104 may be configured with corresponding antennas and other wireless transmission devices. In these scenarios, measurement values ​​108 may be transmitted to analysis platform 106 in various ways, such as at predetermined time intervals (e.g., daily, hourly, or every five minutes), in response to the occurrence of some event (e.g., filling the storage buffer of monitoring device 104), or in response to the end of an observation period.

[0058] Therefore, regardless of where the analysis platform 106 is implemented (e.g., at monitoring device 104, at a smartphone associated with person 102, or at a server device), the analysis platform 106 will obtain the measurement value 108 generated by monitoring device 104. In one or more embodiments, the analysis platform 106 will also obtain additional measurement values ​​generated by monitoring device 104 and / or any other device used during the observation period (e.g., smartwatch, chest strap, etc.).

[0059] In one or more embodiments, the analysis platform 106 may be implemented wholly or partially at the monitoring device 104. Alternatively or additionally, the analysis platform 106 may be implemented wholly or partially using one or more computing devices external to the monitoring device 104, such as one or more computing devices associated with person 102 (e.g., mobile phones, tablets, laptops, desktop computers, or smartwatches) or one or more computing devices associated with a service provider (e.g., healthcare providers, telemedicine service providers, services corresponding to the provider of monitoring device 104, medical testing laboratory services, etc.). In the latter scenario, the analysis platform 106 may be implemented at least partially in one or more server devices.

[0060] In the illustrated example 100, the analytics platform includes a storage device 112. According to the technology, the storage device 112 is configured to store measurement values ​​108 and / or other measurement values ​​or information processed by the prediction system 114 to generate one or more predicted values ​​110. The storage device 112 may represent one or more databases and other types of storage devices capable of storing measurement values ​​108 and / or other types of measurement values. The storage device 112 may also store various other data, such as personal information, demographic information describing the person 102, information about healthcare providers, information about insurance providers, payment information, prescription information, determined health indicators, account information (e.g., username and password), etc. The storage device 112 may also store data of other users within the user group.

[0061] In the illustrated example 100, the analysis platform 106 also includes a prediction system 114. The prediction system 114 represents the functionality of processing measurements 108 to generate one or more predicted values ​​110. Alternatively or additionally, the prediction system 114 may output one or more time series indicating observations or predictions of one or more events over time. It should also be understood that, in different variations, the prediction system 114 may output different combinations of multiple predictions.

[0062] In at least one embodiment, prediction system 114 uses machine learning to generate one or more predicted values ​​110. For example, and not limitingly, prediction system 114 may include one or more neural networks trained based on historical measurements and historical outcome data of a user group. Prediction system 114 may include one or more machine learning models (e.g., an ensemble of models). Alternatively or additionally, prediction system 114 may include logic (machine learning models and / or other types of logic) for preprocessing the acquired measurements to extract various cardiovascular features and / or other features from the sequence of measurements. Example 100 also includes predicted values ​​110 corresponding to the output of prediction system 114.

[0063] In various examples, prediction system 114 represents and / or includes blood pressure monitoring system 116, and predicted value 110 includes and / or represents a measurement of blood pressure 118. For example, as described in further detail below, blood pressure monitoring system 116 is operable to implement a multi-parameter cuffless blood pressure monitoring technique that takes into account physiological and environmental factors affecting the characteristics of the pulse wave to maintain measurement accuracy over long periods without recalibration. In various examples, operation of one or more of the analysis platform 106, prediction system 114, and / or blood pressure monitoring system 116 may be performed by monitoring device 104, devices and systems of a healthcare provider, and / or one or more additional devices not shown.

[0064] Figure 2 A non-limiting example 200 of a monitoring device is shown. Example 200 shows a monitoring device 104.

[0065] According to the technology, the monitoring device 104 includes one or more sensors 202, examples of which include, but are not limited to, a pair or more pairs of electrodes, an accelerometer, a pulse oximeter, and a sweat sensor. The monitoring device 104 may also include a transmitter 204. In this example 200, the monitoring device 104 also includes one or more adhesive portions 206. In operation, the monitoring device 104 is configured to be applied to the skin via one or more adhesive portions 206, such that, for example, one or more sensors 202 are set to detect and record the electrical activity of the heart of the person 102, for example, to generate an electrocardiogram (ECG and / or EKG). In at least one embodiment, the monitoring device 104 can be removed by peeling one or more adhesive portions 206 off the skin.

[0066] It should be understood that the monitoring device 104 and its various components are only one form factor, and the monitoring device 104 and its components may have different form factors without departing from the spirit or scope of the technology described.

[0067] In one or more embodiments, monitoring device 104 may include a processor and / or memory (not shown). Monitoring device 104 may utilize the processor to generate measurement values ​​108 based on communication with one or more sensors 202 that indicate aspects of person 102, such as person 102's blood pressure 118. In one or more embodiments, the processor also generates one or more communicable data packets that include measurement value 108 and / or one or more other measurement values. Alternatively or additionally, the processor generates and / or prompts the storage of other data that may be used to predict the classification of physiological conditions, such as conditions related to blood pressure 118.

[0068] In embodiments where monitoring device 104 is configured for wireless transmission, transmitter 204 can wirelessly transmit measurement value 108 as a data stream to a computing device. For example, in one or more embodiments, monitoring device 104 is configured to transmit (e.g., send and / or receive) information (e.g., potential measurement value) via a Bluetooth Low Energy (BLE) connection. Alternatively or additionally, monitoring device 104 can buffer measurement value 108 (e.g., buffer it in memory) and have transmitter 204 subsequently transmit the buffered measurement value at various intervals (e.g., time intervals (per second, every thirty seconds, per minute, every five minutes, per hour, etc.), storage intervals (when the buffered measurement value reaches a threshold data amount), etc.).

[0069] Multi-parameter cuffless blood pressure monitoring system

[0070] The following discussion describes techniques that can be implemented using the systems, components, and devices described above and below. Aspects of the processes can be implemented in hardware, firmware, software, or a combination thereof. These processes are shown as a set of boxes specifying operations that can be performed by one or more devices, and are not necessarily limited to the order in which the operations are performed as shown by the respective boxes. For example, one or more boxes of a process specify operations that can be programmed into instructions by hardware (e.g., a processor, microprocessor, controller, firmware) to create a dedicated machine for executing the algorithm shown in the flowchart. Therefore, these instructions can be stored on a computer-readable storage medium that enables the hardware to execute the algorithm. References will be made to the following discussion. Figures 1 to 13 .

[0071] Figure 3 A non-limiting system in an example embodiment 300 of multi-parameter cuffless blood pressure monitoring is shown, wherein the operation of the blood pressure monitoring system 116 is shown in more detail.

[0072] In various examples, blood pressure monitoring system 116 refers to a wearable device (such as monitoring device 104) that can be attached to a user's skin surface, supports the functions of the wearable device, can be implemented by the wearable device, and / or (partially or wholly) includes the wearable device. In additional or alternative examples, blood pressure monitoring system 116 includes one or more components and / or operations that can be performed outside the wearable device (e.g., by analytics platform 106).

[0073] The blood pressure monitoring system 116 is operable to generate accurate blood pressure measurements that take into account physiological factors that may affect pulse propagation characteristics, either partially or entirely, independently of corresponding blood pressure changes. For example, various physiological parameters may have a disproportionate effect on pulse propagation time, leading to inaccurate blood pressure derivations. The blood pressure monitoring system 116 is shown as including a sensor arrangement 302, one or more calibration sensors 304, and one or more processors 306 operable to acquire and process physiological data to generate blood pressure 118, such as a blood pressure measurement derived using the techniques described herein. The blood pressure monitoring system 116 is also shown as including a presentation module 308, which is configured to output blood pressure 118 and / or perform various functions using blood pressure 118.

[0074] In this example, firstly, sensor arrangement 302 and calibration sensor 304 may be configured to interact with cardiovascular, physiological, and / or environmental systems to acquire various measurements used to determine blood pressure. For example, sensor arrangement 302 is configured to acquire physiological time-series data 314, which indicates cardiovascular signal characteristics such as electrical activity from the heart, optical changes in blood volume, and / or changes in bioimpedance that occur as blood pulsates through blood vessels. The time-series data 314 may be processed to derive temporal relationships between proximal and distal measurement locations, as described further in detail below, to generate pulse arrival time (“PAT”) and / or pulse conduction time (“PTT”) along the cardiovascular pathway.

[0075] For example, sensor arrangement 302 may include a proximal sensor 310 located at a proximal position along the cardiovascular pathway and a distal sensor 312 located at a distal position along the cardiovascular pathway. Proximal sensor 310 may be configured to detect a first cardiovascular signal as part of physiological timing data 314, while distal sensor 312 may be configured to detect a second cardiovascular signal corresponding to the first cardiovascular signal. In some cases, proximal sensor 310 may include one or more of a proximal electrocardiogram (“ECG”) sensor, a proximal bioimpedance sensor, or a proximal photoplethysmography (“PPG”) sensor. Distal sensor 312 may include one or more of a distal PPG sensor, a distal pressure sensor, or a distal bioimpedance sensor.

[0076] The calibration sensor 304 is configured to measure calibration parameters 316 that affect pulse propagation characteristics (e.g., timing, amplitude, rate, etc.) independently of corresponding blood pressure changes. Calibration parameters 316 can include various physiological measurements that affect pulse characteristics through mechanisms other than blood pressure changes. For example, a particular calibration parameter 316 may affect blood pressure but cause a non-proportional change in the timing data 314 that does not reflect the corresponding blood pressure change; if ignored, this could lead to measurement errors. Therefore, due to the presence of one or more calibration parameters 316, calibration drift frequently occurs in conventional cuffless blood pressure techniques.

[0077] Various correction parameters 316 may be considered. In at least one example, correction parameter 316 includes body temperature 318 measured using a temperature sensor and / or a heat flux sensor to account for the influence of systemic cardiovascular effects on pulse propagation. Additionally or alternatively, correction parameter 316 may include skin temperature 320 measured adjacent to sensor arrangement 302 to account for the influence of vasomotor activity (e.g., vasoconstriction and / or vasodilation) on pulse propagation timing. In some examples, correction parameter 316 includes a perfusion index 322, such as a perfusion index 322 calculated based on PPG data acquired by correction sensor 304, to account for the influence of peripheral circulation on pulse wave characteristics. In various embodiments, correction parameter 316 may include hydration level 324 measured using bioimpedance analysis to account for changes in arterial stiffness, and / or muscle activity 326 detected by electromyography to account for the effects of arterial wall contraction. This is merely illustrative and not limiting, and various additional or alternative correction parameters 316 may be considered.

[0078] In various examples, processor 306 can adjust one or more properties of sensor arrangement 302 and / or calibrate sensor 304 based on measurements of timing data 314 and / or calibration parameters 316. For example, processor 306 can adjust the sensor sampling rate to optimize signal quality during periods of physiological change, modify sensor gain settings to accommodate changes in signal amplitude, change filtering parameters to reduce noise artifacts, selectively activate or deactivate specific sensors based on detected physiological conditions, and so on. In some examples, processor 306 can dynamically reconfigure sensor properties (e.g., measurement window or timing interval) to improve measurement accuracy when calibration parameters 316 indicate a higher probability of inaccuracy than a threshold.

[0079] One or more processors 306 are coupled to sensor arrangement 302 and / or calibration sensor 304 and can generate blood pressure 118 based on measurements of timing data 314 and / or calibration parameters 316. For example, processor 306 is configured to process physiological timing data 314 to determine pulse propagation time 328 based on the time difference between corresponding points in the cardiovascular signal. In various examples, processor 306 calculates pulse propagation time 328 as one or more of pulse conduction time (“PTT”) or pulse arrival time (“PAT”) based on signal analysis from proximal sensor 310 and distal sensor 312.

[0080] For example, and not as a limitation, processor 306 can determine PTT by measuring the time difference between a characteristic of the ECG signal detected by proximal sensor 310 (e.g., the peak value of the R wave) and the corresponding peak value of the PPG signal detected by distal sensor 312. In this example, PTT represents the time it takes for the pulse wave to travel from a proximal location (e.g., the heart) to a distal location (e.g., a peripheral measurement site). In alternative or additional examples, processor 306 can determine PAT by measuring the time difference between an ECG characteristic (e.g., the Q wave onset point) and the corresponding PPG characteristic at a distal location (e.g., the “foot point” of the corresponding PPG waveform).

[0081] Based on the pulse propagation time 328, the processor 306 can generate an initial blood pressure 330, for example, by utilizing a known relationship between pulse wave velocity / timing and vascular pressure. For example, the processor 306 can apply empirical models, mathematical algorithms, and / or calibration curves that correlate pulse propagation characteristics with systolic and / or diastolic blood pressure values. In some embodiments, the processor 306 can utilize a machine learning model trained on population data to establish a relationship between pulse propagation time 328 and blood pressure ranges for different population groups or physiological conditions.

[0082] Subsequently, processor 306 is operable to generate blood pressure 118 based on initial blood pressure 330 and correction parameter 316. For example, processor 306 can apply correction factor 332 based on correction parameter 316 to adjust initial blood pressure 330 and generate blood pressure 118. Processor 306 can generate correction factor 332 based on various considerations, such as, but not limited to, threshold of correction parameter 316, deviation magnitude of correction parameter 316 relative to baseline measurement, rate of change of correction parameter 316, historical patterns in correction parameter 316, user-specific calibration data, relationships between one or more correction parameters 316, demographic information, time of day, activity level, posture, environmental conditions, etc. Processor 306 can also employ various techniques to generate correction factor 332, such as, but not limited to, mathematical algorithms, machine learning models, statistical regression techniques, lookup tables, calibration curves, fuzzy logic systems, neural networks, decision trees, ensemble methods, Bayesian inference, hybrid methods combining multiple processing techniques, etc.

[0083] In at least one example, processor 306 implements threshold-based correction, wherein a correction factor 332 is applied in response to detecting that correction parameter 316 exceeds a predetermined range. For example, correction parameter 316 includes skin temperature 320. When skin temperature 320 is above a first threshold (e.g., above 35°C), processor 306 may apply a vasodilator correction algorithm including the first correction factor 332 to compensate for the increase in pulse propagation time caused by peripheral vasodilation. When skin temperature 320 drops below a second threshold (e.g., below 30°C), processor 306 may activate and apply a vasoconstrictor correction algorithm including the second correction factor 332 to adjust initial blood pressure 330, thereby taking into account the shortened pulse propagation time. When the skin temperature 320 is between the first threshold and the second threshold (e.g., between 30°C and 35°C), the processor 306 may apply a reduced correction factor 332 and / or not apply a correction factor to generate blood pressure 118 primarily based on pulse propagation time 328, thereby saving computational resources and achieving low-power operation by selectively activating the correction mechanism only when necessary for the detected physiological condition.

[0084] In an additional or alternative example, processor 306 can implement an "always-on" correction method that continuously applies correction factor 332. In this implementation, processor 306 can continuously monitor correction parameter 316 and dynamically adjust correction factor 332 in real time based on physiological conditions. For example, correction factor 332 can be calculated as a continuous function of correction parameter 316, where the amplitude of correction factor 332 varies proportionally to the measured parameter value. This method can provide fine-grained blood pressure adjustment that takes into account subtle physiological changes.

[0085] In one or more examples, processor 306 may generate one or more correction factors 332 based on multiple correction parameters 316. For example, processor 306 may implement a weighting scheme to assign weights to various correction parameters 316 based on their relative impact on pulse propagation time 328, measurement reliability, and / or physiological significance. In the example of measuring skin temperature 320 and hydration level 324, processor 306 may apply a relatively high weight to skin temperature 320 during periods of significant temperature variation, while assigning a lower weight to hydration level 324. The weighting scheme may be dynamically adjusted based on various factors such as signal quality, measurement confidence level, physiological state over time, user-specific calibration data, relationships between correction parameters 316, etc.

[0086] In various examples, processor 306 generates a first correction factor 332 for systolic blood pressure and a second correction factor 332 for diastolic blood pressure. For example, the amplitude and / or direction of the first and second correction factors 332 may differ, for example, based on how the measured correction parameter 316 has different effects on the systolic and diastolic blood pressure measurements. For example, certain physiological conditions (e.g., changes in vascular compliance) may have a relatively large effect on systolic blood pressure, while other physiological conditions (e.g., changes in peripheral resistance) may have a relatively large effect on diastolic blood pressure.

[0087] Therefore, the blood pressure monitoring system 116 can implement various techniques to generate blood pressure 118 based on various considerations. The processor 306 can implement these techniques in various time scenarios to optimize system performance and accuracy. For example, the processor 306 can apply a correction factor 332 in real time during data acquisition to provide immediate blood pressure measurements, or it can perform post-processing analysis on the acquired time-series data 314 and correction parameters 316 to generate retrospective blood pressure calculations with higher accuracy. In some embodiments, the processor 306 can utilize a hybrid approach combining real-time preliminary correction with subsequent post-processing refinement to balance immediate feedback requirements with computational efficiency.

[0088] Once generated, the blood pressure monitoring system 116 can perform various functions based on blood pressure 118. For example, the presentation module 308 is configured to output and / or present blood pressure 118 through various display and communication mechanisms. The presentation module 308 can generate indications 334 that provide the user with visual, auditory, and / or tactile feedback regarding the blood pressure status. In various examples, the presentation module 308 prompts the display of systolic and diastolic blood pressure values ​​determined by the processor 306 (e.g., based on characteristics of pulse propagation time 328 and correction parameter 316). The presentation module 308 can also provide information about the currently activated correction factor, measurement confidence level, and trends in blood pressure readings over time. Additionally or alternatively, the presentation module 308 may include wireless transmission capabilities to transmit the blood pressure 118 measurement to external devices, cloud systems, or healthcare provider networks for further analysis and monitoring. In various embodiments, presenting blood pressure 118 includes modifying one or more attributes of the wearable device to adjust the sensor properties of sensor arrangement 302 and / or calibrate sensor properties of sensor 304 (e.g., sensitivity, acquisition rate, etc.).

[0089] In this way, the technologies, devices, and components described herein overcome the limitations of conventional systems by mitigating inaccuracies caused by calibration drift in cuffless blood pressure monitoring and taking into account the influence of various physiological factors on blood pressure measurements. Therefore, the blood pressure monitoring system 116 can maintain accuracy for extended periods without manual recalibration, supporting continuous blood pressure monitoring and generating accurate cardiovascular health analysis results over prolonged wear periods. Furthermore, selectively applying correction factors based on threshold conditions saves computational resources by deploying processing-intensive correction algorithms only when detected physiological parameters indicate their necessity. This improves the operation of the computing devices implementing the system by reducing unnecessary or redundant computations during stable physiological conditions, thereby extending battery life and enabling efficient power management in wearable implementations.

[0090] Figure 4 A non-limiting example 400 of multi-parameter cuffless blood pressure monitoring is shown, wherein the wearable device generates blood pressure measurements based on time-series data and one or more measured calibration parameters.

[0091] In this example 400, wearable device 402 is configured to perform cuffless blood pressure monitoring via multi-parameter correction, for example, through one or more components including blood pressure monitoring system 116. Wearable device 402 may be positioned adjacent to the skin surface 404 of a user (e.g., person 102) to monitor physiological parameters along cardiovascular pathway 406 (e.g., arteries, arterioles, capillaries, veins, etc.) along which blood flow 408 flows toward / out of the heart. Wearable device 402 includes a sensor arrangement 302 comprising multiple sensors located at different locations along cardiovascular pathway 406 to acquire physiological time-series data (e.g., time-series data 314) and correction parameters 316 for accurate pulse propagation time measurement.

[0092] In the illustrated example 400, sensor arrangement 302 includes a proximal sensor 310 located at a proximal position along cardiovascular pathway 406 and a distal sensor 312 located at a distal position along cardiovascular pathway 406. Proximal sensor 310 can be configured to detect a first signal 410 as part of physiological timing data, while distal sensor 312 can be configured to detect a second signal 412 as part of physiological timing data, the second signal 412 corresponding to the first signal 410. As described further in detail above, proximal sensor 310 may include one or more of a proximal ECG sensor, a proximal bioimpedance sensor, or a proximal PPG sensor. Distal sensor 312 may include one or more of a distal PPG sensor, a distal pressure sensor, or a distal bioimpedance sensor. The time difference between the characteristics of the first signal 410 and the second signal 412 enables determination of the pulse propagation time 328 along cardiovascular pathway 406.

[0093] Wearable device 402 also includes a calibration sensor 304 configured to measure a calibration parameter 316 that affects pulse travel time 328 independently of corresponding changes in blood pressure 118. The calibration sensor 304 can be integrated into the same form factor as sensor arrangement 302 to allow simultaneous acquisition of timing data and calibration parameter 316. Processor 306 processes physiological timing data from first signal 410 and second signal 412 to determine pulse travel time 328, which may include one or more of PTT or PAT.

[0094] Subsequently, processor 306 generates a blood pressure measurement of 118 based on both the pulse propagation time 328 and the measured value of correction parameter 316, to compensate for the influence of physiological factors affecting pulse wave velocity independently of blood pressure changes. For example, processor 306 generates a correction factor 332 based on the characteristics of correction parameter 316 (e.g., amplitude, frequency, oscillation, duration, rate of change, pattern variability, threshold, etc.). Then, processor 306 applies correction factor 332 to the initial blood pressure 330 to generate blood pressure 118.

[0095] Wearable device 402 can output blood pressure 118 in various ways, such as by displaying it on an integrated screen, transmitting it to a paired mobile device, storing it in local memory for later retrieval, wirelessly transmitting it to a cloud-based health platform, using an audible alarm or notification, a haptic feedback mode, a visual indicator such as an LED light, integrating it with an electronic health record system, streaming it in real time to a healthcare monitoring service, or a combination thereof. Presentation module 308 can select an appropriate output method based on user preferences, device capabilities, connectivity status, the urgency of the reading, or clinical requirements.

[0096] Although in this example 400, the proximal sensor 310, distal sensor 312, and calibration sensor 304 are shown as being included within the housing of the wearable device 402, this is merely an example and not a limitation. In one or more examples, one or more of the processor 306, proximal sensor 310, distal sensor 312, and / or calibration sensor 304 may be located remotely from the wearable device 402, for example, included in a separate monitoring device. For example, one or more sensors and / or components of the blood pressure monitoring system 116 may be distributed across multiple wearable devices or form factors to provide comprehensive physiological monitoring. For example, the blood pressure monitoring system 116 may be implemented across a combination of devices, such as a chest patch for proximal measurements, a wrist-worn device for distal measurements, and a ring-type sensor for additional calibration parameters. Thus, the techniques described herein can combine data acquired from various devices at different physiological locations to provide a variety of analytical results.

[0097] Figure 5 A non-limiting example 500 of multi-parameter cuffless blood pressure monitoring is shown, wherein a temperature sensor is used to acquire skin temperature data as a correction parameter.

[0098] In this example 500, vasoconstriction occurs in cardiovascular pathway 406, for example, caused by cold exposure. For instance, low temperatures cause arterial walls to constrict and reduce the inner diameter of the vessel, as indicated by the narrowing lumen 502 of cardiovascular pathway 406. The narrowing lumen 502 affects the characteristics of blood flow 408 and the resulting pulse propagation measurements, thereby reducing the accuracy of blood pressure readings. For example, the reduced diameter of the narrowing lumen 502 may cause pulse waves to pass relatively faster through the constricted cardiovascular pathway 406 without a proportional change in blood pressure.

[0099] Therefore, the pulse propagation time 328, based on the time difference between the characteristics of the first signal 410 and the second signal 412, is relatively short, for example, 230 ms relative to a baseline time of 250 ms. Based on the pulse propagation time 328, the processor 306 generates an initial blood pressure 330 of 142 / 99 mmHg. Therefore, the initial blood pressure 330, based solely on the pulse propagation time 328 and without correction for vasoconstriction, is artificially inflated relative to the actual blood pressure of the person 102.

[0100] Therefore, the calibration sensor 304 in this example includes a temperature sensor 504 operable to measure the skin temperature 320 adjacent to the sensor arrangement 302 as a calibration parameter 316 to account for the effect of vasomotor activity on pulse propagation time 328. In example 500, the temperature sensor 504 detects a skin temperature 320 of 28 degrees Celsius, for example, four degrees lower than a particular user's baseline. The processor 306 generates a calibration factor 332 based on the measured skin temperature 320.

[0101] Subsequently, processor 306 applies a correction factor 332 based on the measured skin temperature 320 to generate an adjusted blood pressure 118 that takes into account the effects of vasomotor activity. In Example 500, correction factor 332 adjusts the initial blood pressure 330 from 142 / 99 mmHg to a corrected blood pressure 118 of 130 / 88 mmHg. This correction compensates for the faster pulse propagation caused by the narrowed lumen 502 and provides an accurate representation of blood pressure independent of changes in pulse propagation time 328 in peripheral vessels.

[0102] Figure 6 Additional non-limiting example 600 of multi-parameter cuffless blood pressure monitoring is shown, wherein a temperature sensor is used to acquire skin temperature data as a correction parameter.

[0103] In this example 600, vasodilation occurs in the cardiovascular pathway 406, for example, due to heat exposure. The increased temperature relaxes the arterial walls and increases the inner diameter of the blood vessels, as shown by the widening lumen 602 of the cardiovascular pathway 406. The widening lumen 602 affects the characteristics of blood flow 408 and the resulting pulse propagation measurements, for example, causing the pulse wave to pass relatively slowly through the dilated cardiovascular pathway 406.

[0104] Therefore, the pulse propagation time 328, based on the time difference between the first signal 410 and the second signal 412, is relatively long, for example, 275 ms relative to a baseline time of 250 ms. Based on the pulse propagation time 328, the processor 306 generates an initial blood pressure measurement 330 of 108 / 72 mmHg. Therefore, the initial blood pressure 330, based solely on the pulse propagation time 328 and without correction for vasodilation, is inaccurately lower than the user's actual blood pressure.

[0105] To compensate for this, in the illustrated example, temperature sensor 504 detects a skin temperature 320 of 36 degrees Celsius, for example, four degrees higher than the user's baseline temperature. Processor 306 generates a correction factor 332 based on the skin temperature 320 and applies the correction factor 332 to the initial blood pressure 330 to account for the effects of vasomotor activity. In this example, correction factor 332 adjusts the initial blood pressure 330 from 108 / 72 mmHg to a corrected blood pressure 118 of 116 / 77 mmHg. This correction compensates for the relatively slow pulse propagation time 328 caused by the widened lumen 602, thus providing an accurate representation of blood pressure 118.

[0106] Figure 7 A non-limiting example 700 of multi-parameter cuffless blood pressure monitoring is shown, wherein a heat flux sensor is used to collect body temperature data as a correction parameter.

[0107] In this example, wearable device 402 includes a heat flux sensor 702, serving as a calibration sensor 304, to measure body temperature 318 as a calibration parameter 316 to account for the influence of systemic cardiovascular effects on pulse propagation time 328. The heat flux sensor 702 is operable to detect heat transfer between the user's body and the surrounding environment, thus providing measurements of core body temperature changes that may affect cardiovascular function (e.g., caused by fever, dehydration, activity, physical exertion, metabolic changes, etc.). In some cases, the heat flux sensor 702 may be positioned adjacent to the skin surface 404 to optimize thermal contact and measurement accuracy. In this example, body temperature 318 is elevated, for example, reaching 38.5 degrees Celsius due to fever.

[0108] When body temperature (318) rises above the normal physiological range, systemic changes occur in cardiovascular function, which may alter the relationship between pulse propagation time (328) and actual blood pressure. For example, if the pulse propagation time (328) is relatively fast, say 240 ms, then the initial blood pressure (330) generated solely based on the pulse propagation time (328) will be inaccurately high, for example, 135 / 90 mmHg.

[0109] To account for the elevated body temperature 318, processor 306 generates a correction factor 332 based on measurements from heat flux sensor 702 and applies the correction factor 332 to the initial blood pressure 330. In this example, processor 306 applies the correction factor 332 to adjust the initial blood pressure 330 from 135 / 90 mmHg to a corrected blood pressure 118 of 127 / 85 mmHg, thereby providing an accurate representation of blood pressure 118.

[0110] Figure 8 A non-limiting example 800 of multi-parameter cuffless blood pressure monitoring is shown, wherein a PPG sensor is used to acquire perfusion index data as a correction parameter.

[0111] In this example 800, the wearable device 402 includes a PPG sensor 802 to measure the perfusion index 322 as a correction parameter 316, thereby taking into account the influence of peripheral circulation characteristics on the pulse travel time 328. The PPG sensor 802 is operable to acquire various physiological data for the processor 306 to calculate the ratio of pulsatile to non-pulsatile blood flow, which provides an indication of peripheral circulation. In various examples, changes in peripheral circulation cause changes in cardiovascular function, which may alter the relationship between the pulse travel time 328 and actual blood pressure.

[0112] For example, in this example, the perfusion index 322 is relatively low, for example, 0.6%, which results in a relatively fast pulse propagation time 328, for example, 225 ms. Therefore, the initial blood pressure 330 generated solely based on the pulse propagation time 328 is inaccurately high, for example, 145 / 98 mmHg. To account for the relatively low perfusion index 322, the processor 306 generates a correction factor 332 based on measurements from the PPG sensor 802 and applies the correction factor 332 to the initial blood pressure 330. In the example shown, the correction factor 332 adjusts the initial blood pressure 330 from 145 / 98 mmHg to a corrected blood pressure 118 of 130 / 88 mmHg, which is physiologically accurate.

[0113] Figure 9 A non-limiting example 900 of multi-parameter cuffless blood pressure monitoring is shown, wherein a bioimpedance sensor is used to acquire hydration level data as a correction parameter.

[0114] In this example, wearable device 402 includes a bioimpedance sensor 902 that measures hydration level 324 as a correction parameter 316 to account for the effect of arterial stiffness caused by tissue dehydration on pulse propagation time 328. For example, bioimpedance sensor 902 excites electrodes to acquire impedance data for hydration estimation, thereby providing a quantitative assessment of the fluid state of various tissues within the body. In this example, bioimpedance sensor 902 applies an electric current through electrodes located on the skin surface 404 and measures the resulting impedance characteristics of the body tissue. For example, body tissues with relatively high water content exhibit relatively low impedance, while dehydrated tissues exhibit relatively high impedance values.

[0115] Bioimpedance sensor 902 analyzes these impedance changes to determine hydration level 324, which in various examples is correlated with the fluid content of blood vessels and / or surrounding tissues. In this example, hydration level 324 was determined to be 45%, a decrease from a baseline hydration level of 55%. Decreased hydration may stiffen the arterial walls and reduce compliance, which could affect pulse wave velocity and alter pulse propagation time 328 independently of actual blood pressure changes.

[0116] For example, in this example, the pulse propagation time 328 is measured to be relatively fast, such as 235 ms, and the processor 306 calculates an initial blood pressure 330 of 138 / 92 mmHg based on this measurement of pulse propagation time 328. However, dehydration can make the arterial walls more rigid, which can affect the relationship between pulse propagation time 328 and actual blood pressure 118, and the initial blood pressure 330 may be inaccurately elevated relative to the actual blood pressure. Therefore, the processor 306 generates a correction factor 332 based on the degree of hydration 324 and applies the correction factor 332 to the initial blood pressure 330 to generate a blood pressure 118 measurement of, for example, 130 / 88 mmHg, which takes into account the effect of dehydration on pulse wave velocity and provides an accurate representation of the actual cardiovascular stress condition.

[0117] Figure 10 A non-limiting example 1000 of multi-parameter cuffless blood pressure monitoring is shown, wherein an electromyography (“EMG”) sensor is used to acquire muscle activity data as a correction parameter.

[0118] In this example, wearable device 402 includes an EMG sensor 1002 as a calibration sensor 304 to measure muscle activity 326 as a calibration parameter 316, thereby taking into account the effect of arterial wall contractility on pulse propagation time 328. For example, muscle activity 326 can affect pulse propagation time 328 through changes in arterial wall tension and compliance that occur independently of changes in blood pressure 118. The EMG sensor 1002 can detect electrical activity generated by smooth muscle contraction within the arterial wall to provide a measurement of the effect of muscle on vascular properties that affect the measurement of pulse propagation time 328.

[0119] For example, as shown in the figure, muscle activity 326 is measured as a relative increase, for example, 85 µV. Therefore, the pulse propagation time 328 is relatively fast, for example, 220 ms, which causes the initial blood pressure 330 generated by the processor 306 based on the pulse propagation time 328 to be inaccurately high, for example, 150 / 99 mmHg. Therefore, the processor 306 generates a correction factor 332 based on the data acquired by the EMG sensor 1002 and applies the correction factor 332 to the initial blood pressure 330 to generate a blood pressure 118 measurement of, for example, 132 / 88 mmHg, which takes into account the effect of muscle activity 326.

[0120] Figure 11 A non-limiting example 1100 of multi-parameter cuffless blood pressure monitoring is shown, in which multiple calibration parameters are used to generate blood pressure measurements.

[0121] In this example 1100, the wearable device 402 includes various calibration sensors 304, such as a temperature sensor 504, a heat flux sensor 702, a PPG sensor 802, a bioimpedance sensor 902 (“BIA”), and an EMG sensor 1002. The processor 306 can coordinate measurements from the various calibration sensors to generate a calibration factor 332 applied to an initial blood pressure 330 based on various calibration parameters 316, such as skin temperature 320, body temperature 318, perfusion index 322, hydration level 324, and / or muscle activity 326.

[0122] In various examples, processor 306 may implement a weighting scheme 1102 to weight the contributions of one or more correction parameters 316 to the generation of correction factor 332. For example, weighting scheme 1102 applies weights to each correction parameter 316 based on factors such as measurement reliability, physiological relevance, amplitude, rate of change, and time stability. In some implementations, weighting scheme 1102 assigns relatively high weights to correction parameters that exhibit a stronger correlation with changes in pulse propagation time 238, independent of changes in blood pressure.

[0123] In some examples, processor 306 adjusts the weights based on the relationship between two or more correction parameters 316. For example, when both skin temperature 320 and perfusion index 322 indicate peripheral vasoconstriction, processor 306 may increase the weights assigned to these parameters compared to examples where these parameters indicate contradictory physiological states. Similarly, if hydration level 324 decreases while body temperature 318 increases, processor 306 may increase the weight of hydration measurements, for example, because dehydration amplifies the effect of temperature on vascular compliance. Processor 306 is operable to dynamically adjust the weights within weighting scheme 1102 based on real-time evaluation of characteristics of the correction parameters 316, such as amplitude, rate of change, signal quality, measurement confidence levels from each correction sensor 304, etc.

[0124] In some examples, the calibration parameter measurements can be incorporated into the blood pressure calculation algorithm executed by processor 306 through continuous integration. In these examples, processor 306 continuously monitors the calibration parameter 316 and adjusts the calculated blood pressure values ​​in real time. This continuous integration allows processor 306 to account for progressive physiological changes that may affect pulse propagation time 328 over a long monitoring period.

[0125] Alternatively or additionally, processor 306 may apply a correction factor 332 to initial blood pressure 330 in response to detecting that one or more correction parameters 316 exceed a threshold. For example, processor 306 monitors each correction parameter 316 against a predetermined threshold indicating when a physiological condition may significantly affect pulse propagation time. When processor 306 detects that one or more correction parameters 316 exceed the set threshold, processor 306 activates one or more correction algorithms and / or increases the weight of the affected parameters within weighting scheme 1102. The threshold-based approach allows processor 306 to conserve computational resources during periods of stable physiological condition while providing enhanced correction when appropriate.

[0126] Processor 306 can implement various computational methods to generate blood pressure 118. For example, processor 306 can utilize one or more mathematical algorithms that apply predetermined correction equations (e.g., a weighted combination based on correction parameters 316) to adjust the initial blood pressure 330. As described in further detail below, in some examples, processor 306 employs one or more machine learning models trained to calibrate blood pressure readings based on correction parameters 316. For example, a particular machine learning model is trained to understand the complex relationships between multiple correction parameters 316 and their corresponding effects on pulse propagation time 328.

[0127] The processor 306 can also implement statistical regression techniques that establish a mathematical relationship between the correction parameter 316 and blood pressure adjustment by analyzing training data. Additionally or alternatively, the processor 306 can access a lookup table that provides a predetermined correction factor 332 for various combinations of the correction parameter range, or utilize a calibration curve that maps the correction parameter values ​​to the corresponding blood pressure adjustment.

[0128] Processor 306 may determine systolic and / or diastolic blood pressure values, for example, based on characteristics of pulse propagation time 328 and / or measurements of correction parameter 316, to include them as part of blood pressure 118. In one or more examples, correction parameter 316 may have different effects on systolic and diastolic blood pressure measurements; therefore, processor 306 is operable to apply individual correction factors 332 and / or weighting schemes 1102 for each blood pressure component.

[0129] In the illustrated example 1100, processor 306 generates a pulse propagation time 328 based on a first signal 410 and a second signal 412. The pulse propagation time 328 is relatively fast, for example, 218 ms, therefore the initial blood pressure 330 is inaccurately elevated, for example, to 155 / 105 mmHg. Therefore, processor 306 applies a correction factor 332 based on various correction parameters 316 to generate a physiologically accurate blood pressure measurement of 139 / 95 mmHg. Thus, the technique described herein can utilize multiple physiological measurements to maintain measurement accuracy under different physiological conditions and individual differences.

[0130] Figure 12 A non-limiting example 1200 of multi-parameter cuffless blood pressure monitoring is shown, wherein a user interface 1202 for a blood pressure monitoring scenario is shown.

[0131] User interface 1202 enables users to monitor blood pressure trends and view the status of various calibration parameters that may affect measurement accuracy. For example, user interface 1202 includes a first panel 1204 that includes blood pressure readings and status information. In this example, the first panel 1204 displays a blood pressure measurement of "128 / 82 mmHg" and a status indicator showing "Within Range" to convey that the measured value is within the normal parameter range. Additionally, the first panel 1204 includes a trend indicator showing "Stable" to inform the user of the recent direction of blood pressure changes. In some embodiments, the first panel 1204 may include color coding or other visual indicators to convey whether the blood pressure reading is within a target range or requires attention.

[0132] Below the first panel 1204, the second panel 1206 includes a graphical representation of the trend of blood pressure changes over time. The second panel 1206 presents systolic and diastolic blood pressure measurements as separate trend lines, providing a visual representation of blood pressure patterns and changes over long periods. The graphs included in the second panel 1206 can show trends, fluctuations, and / or periods of stability in blood pressure measurements. In various embodiments, the second panel 1206 may include interactive features that allow users to zoom in on specific time periods and / or view detailed measurement data at specific points along the trend line.

[0133] The third panel 1208 is located below the second panel 1206 and displays calibration status information related to various calibration parameters 316 that may affect pulse propagation time measurements. The third panel 1208 provides status indicators for multiple calibration parameters 316, including measurements of temperature, perfusion, hydration, muscle activity, and body temperature. Each calibration parameter in the third panel 1208 may display an active or inactive state to indicate whether a calibration is currently applied based on the measured value of the corresponding calibration parameter 316. For example, the third panel 1208 may show that calibrations based on skin temperature and perfusion are currently active, while calibrations based on hydration, muscle activity, and body temperature are still inactive. This is merely an example and not a limitation, and the user interface 1202 can be configured in various ways and include various features.

[0134] Figure 13 A flowchart is shown illustrating an algorithm as a step-by-step process 1300 in an example implementation, one or more steps of which can be executed by a processing device to generate blood pressure 118 using multi-parameter cuffless blood pressure monitoring technology.

[0135] In this example, firstly, physiological time-series data indicating the pulse propagation time are acquired (box 1302). For example, time-series data 314 can be acquired by one or more sensors of sensor arrangement 302 of wearable device 402 during the monitoring period. For example, time-series data 314 includes a first physiological signal acquired by a proximal sensor 310 located at a proximal position along cardiovascular pathway 406, and a second physiological signal acquired by a distal sensor 312 located at a distal position along cardiovascular pathway 406. The first signal 410 and / or the second signal 412 can include various types of data, such as, but not limited to, various combinations of ECG data, PPG data, bioimpedance data, etc. Therefore, time-series data 314 can indicate the time difference between corresponding points in the first physiological signal and the second physiological signal.

[0136] Measurements of correction parameters that affect pulse propagation time independently of corresponding blood pressure changes are acquired (box 1304). For example, wearable device 402 also includes one or more correction sensors 304 to acquire measurements of one or more correction parameters 316. Correction parameters may include at least one of body temperature 318, skin temperature 320, perfusion index 322, hydration level 324, or muscle activity 326. Correction sensors 304 measure these parameters to account for physiological factors that affect pulse propagation measurements (e.g., pulse propagation time 328) partially or completely independently of actual blood pressure changes.

[0137] Physiological time-series data is processed to generate pulse propagation time (box 1306). For example, processor 306 is operable to analyze a first physiological signal and a second physiological signal to determine pulse propagation time 328, such as pulse conduction time (“PTT”) and / or pulse arrival time (“PAT”), based on the time difference between corresponding points in physiological time-series data 314. For example, processor 306 may identify characteristic features in the first physiological signal (such as the R-wave peak in an ECG signal) and corresponding features in the second physiological signal (such as pulse arrival at a distal location indicated by PPG data). The processor may determine the time difference between these corresponding features to generate pulse propagation time 328.

[0138] Subsequently, a blood pressure measurement is generated based on the pulse propagation time and the measured value of the correction parameter (box 1308). For example, processor 306 may generate an initial blood pressure 330 based on the pulse propagation time 328 and adjust the initial blood pressure 330 based on the measured value of the correction parameter 316. For example, processor 306 may apply a correction factor 332 to the initial blood pressure 330 based on the measured value of the correction parameter 316.

[0139] In some examples, in response to detecting that one or more correction parameters 316 are higher or lower than a threshold, the initial blood pressure 330 is adjusted using a correction factor 332. Therefore, the processor 306 can implement different processing modes above and below the threshold. The processor 306 can also perform real-time optimization of device operation based on threshold conditions by switching between different processing modes, such as activating temperature-based correction when body temperature 318 deviates from the baseline range, or implementing hydration-based adjustment when hydration level 324 drops below a predetermined value.

[0140] Subsequently, the blood pressure measurement is presented (step 1310). For example, the blood pressure monitoring system 116 may perform various functions based on the generated blood pressure measurement 118. In one or more examples, the blood pressure monitoring system 116 may output an indication 334 of the blood pressure 118 via various display methods or communication interfaces. The indication 334 may include systolic and diastolic blood pressure values ​​taking into account the measured correction parameter 316. The indication 334 may include visual indicators, numerical displays, or trend information conveying the blood pressure status to the user or healthcare provider.

[0141] In some examples, blood pressure 118 can be transmitted (e.g., wirelessly) to an external device and / or stored locally for subsequent analysis. In at least one example, processor 306 is operable to alter one or more conditions of wearable device 402 based on blood pressure 118 to increase the sampling rate, activate / deactivate one or more sensors, etc. Therefore, the techniques described herein achieve improved accuracy in continuous and / or periodic blood pressure monitoring by systematically incorporating correction parameters that address common sources of measurement drift in cuffless blood pressure devices.

[0142] Machine learning in multi-parameter cuffless blood pressure monitoring systems

[0143] The foregoing examples, such as those for blood pressure monitoring system 116 and / or prediction system 114, describe various instances of machine learning models. In one or more examples, a machine learning model (e.g., an AI model) refers to a computer representation that can be tuned based on inputs (e.g., through training and retraining) without being actively programmed by the user to automatically and without user intervention approximate an unknown function. For example, the term "machine learning model" includes models that utilize algorithms to learn from known data by analyzing training data and making predictions based on known data, thereby generating outputs that reflect the patterns and properties of the training data through learning and relearning.

[0144] In the context of multi-parameter cuffless blood pressure monitoring, machine learning models can be implemented (e.g., by one or more processors 306 of the blood pressure monitoring system 116) to analyze physiological data patterns and correction parameters 316 to generate an accurate measurement of blood pressure 118 that takes into account calibration drift. For example, processor 306 can utilize one or more machine learning models to process physiological time-series data 314 (such as ECG readings, PPG waveforms, bioimpedance measurements, etc.) acquired by sensor arrangement 302 and / or correction parameters 316 acquired by calibration sensor 304, including body temperature 318, skin temperature 320, perfusion index 322, hydration level 324, and muscle activity 326.

[0145] Examples of machine learning models suitable for multi-parameter blood pressure monitoring include: neural networks, convolutional neural networks (“CNNs”) for analyzing pulse waveform characteristics, long short-term memory (“LSTM”) neural networks for analyzing temporal patterns in pulse propagation time 328 and correction parameters 316, generative adversarial networks (“GANs”), decision trees (e.g., for threshold-based correction activation), support vector machines (“SVMs”), linear regression models for establishing the relationship between correction parameters 316 and blood pressure adjustment, logistic regression for binary correction factor activation, Bayesian networks for probabilistic correction factor determination, random forest learning for feature importance in various weighting schemes 1102, dimensionality reduction algorithms, boosting algorithms, deep learning neural networks, and so on.

[0146] For example, a machine learning model can be configured using multiple layers, each with multiple nodes. These multiple layers can be configured to include an input layer, an output layer, and one or more hidden layers. In the context of multi-parameter cuffless blood pressure monitoring, the input layer can receive various physiological parameters from the measurement 108, such as pulse propagation time 328 features, correction parameters 316 including temperature measurements, perfusion index 322 values, hydration level 324 data, muscle activity 326 signals, and so on. For example, the hidden layer processes these inputs through weighted connections to identify complex relationships between pulse propagation characteristics affecting blood pressure accuracy and correction parameters 316, such as relationships that cannot be detected using conventional calibration patterns. The output layer can produce a blood pressure measurement 118 incorporating a correction factor 332 based on the measured correction parameters 316, or generate a weighted scheme 1102 for combining multiple correction parameters 316. Various blood pressure calibration and correction tasks are achieved through computation performed by nodes within each layer via hidden states using a weighted connection system “learned” during the training of the machine learning model.

[0147] To train a machine learning model for multi-parameter blood pressure monitoring, training data is received, which provides examples of what the machine learning model "will learn," i.e., serves as the basis for learning patterns from the data. For multi-parameter blood pressure monitoring applications, the training data may include a labeled dataset of physiological measurements from individuals with known blood pressure values ​​(measured using a cuff-based reference device), along with corresponding pulse propagation times 328 and correction parameters 316 (such as temperature changes, hydration levels, perfusion index, and muscle activity status). For example, a machine learning system including the machine learning model acquires and preprocesses the training data, which includes input features (e.g., PTT measurements, PAT measurements, skin temperature readings 320, body temperature values ​​318, perfusion index calculated values ​​322, hydration level assessed values ​​324, muscle activity signals 326) and corresponding target labels, such as accurate systolic and diastolic blood pressure values, appropriate correction factors 332, optimized weighting schemes 1102 for multiple correction parameters 316, etc.

[0148] The machine learning system can also be operated to initialize various parameters of the machine learning model, which can be used by the machine learning model as internal variables to represent and process information during training. These parameters can also be used to represent inferences obtained through training regarding the relationship between pulse propagation characteristics and physiological factors that cause calibration drift. In one or more embodiments, training data is divided into multiple batches to improve the efficiency of parameter processing and optimization of the machine learning model during training, which is particularly beneficial for model accuracy when processing large amounts of physiological time-series data, including pulse time-series measurements and changes in correction parameters over long monitoring periods.

[0149] Subsequently, the machine learning model receives training data as input and uses it as the basis for generating predicted values ​​based on the current state of the parameters of each layer and corresponding node of the model. The results are output as output data, such as corrected blood pressure 118, correction factor 332 values, weighting scheme 1102 parameters, etc. For example, the blood pressure monitoring system 116 includes a machine learning model trained to identify variations in pulse propagation time 328 and patterns in correction parameters 316 associated with accurate blood pressure measurements. This enables the processor 306 to generate accurate blood pressure 118 values ​​that take into account physiological factors affecting pulse wave characteristics independently of actual blood pressure variations.

[0150] Training a machine learning model can include computing a loss function to quantify the loss associated with the operations performed by the nodes of the machine learning model. Loss functions can be configured in various ways to control the operation and / or functionality of the machine learning model. For example, a loss function can be designed to prioritize the accuracy of blood pressure estimation while minimizing calibration drift over long monitoring periods and taking into account individual physiological differences in the response of the correction parameters. For instance, the computation of the loss function may involve comparing the difference between a predicted value specified in the output data (e.g., a predicted blood pressure value with a correction factor of 332 applied) and a target label specified by the training data (e.g., a cuff-based reference blood pressure measurement). Loss functions can be configured in various ways, examples of which include regret values, quadratic loss functions as part of a least-squares technique for continuous blood pressure parameter estimation, cross-entropy loss for threshold-based correction activation decisions, custom loss functions incorporating physiological constraints for pulse propagation and vascular mechanics, mean absolute error for blood pressure accuracy optimization, and so on.

[0151] The configuration of training data can be used to support various use cases in multi-parameter cuffless blood pressure monitoring. For example, machine learning models can be trained to detect specific patterns in changes in pulse propagation time 328 related to temperature-induced vasoconstriction or vasodilation, identify hydration-related changes in arterial stiffness affecting pulse wave velocity, identify perfusion changes affecting peripheral circulation and pulse timing, detect muscle activity patterns affecting arterial wall properties, or establish complex relationships among multiple calibration parameters 316 that collectively affect the accuracy of blood pressure measurement. The model can be configured to operate in a threshold-based calibration activation mode to provide accurate calibration while conserving computational resources when calibration parameters 316 exceed predetermined ranges. The model can also be reconfigured to achieve continuous calibration integration when enhanced accuracy is required under various physiological conditions. This adaptive approach efficiently utilizes computational resources dedicated to the machine learning process while ensuring comprehensive calibration capabilities using physiological time-series data 314 and calibration parameters 316 acquired during continuous monitoring periods.

[0152] It should be understood that many variations are possible based on the disclosed content of this document. Although the features and elements have been described above in specific combinations, each feature or element may be used alone without other features and elements, or in various combinations with or without other features and elements.

[0153] Clause 1. A wearable device for cuffless blood pressure monitoring, comprising: a sensor arrangement for acquiring physiological time-series data indicating pulse propagation time along a cardiovascular pathway; one or more calibration sensors configured to measure calibration parameters that affect pulse propagation time independently of corresponding blood pressure changes; and a processor configured to: process the physiological time-series data to determine pulse propagation time; and present blood pressure measurements generated based on the pulse propagation time and the calibration parameters.

[0154] Clause 2. The wearable device according to Clause 1, wherein the sensor arrangement includes: a proximal sensor located at a proximal position along the cardiovascular pathway and configured to detect a first cardiovascular signal as part of physiological time-series data; and a distal sensor located at a distal position along the cardiovascular pathway and configured to detect a second cardiovascular signal as part of physiological time-series data and corresponding to the first cardiovascular signal.

[0155] Clause 3. The wearable device according to Clause 1 or 2, wherein one or more calibration sensors include one or more of a temperature sensor or a heat flux sensor configured to measure skin temperature adjacent to the sensor arrangement as a calibration parameter, and the processor is further configured to generate a blood pressure measurement based on the skin temperature to take into account the effect of vasomotor activity on pulse propagation time.

[0156] Clause 4. The wearable device according to any of the preceding clauses, wherein one or more calibration sensors include one or more of a temperature sensor or a heat flux sensor configured to measure body temperature as a calibration parameter, and the processor is further configured to generate a blood pressure measurement based on body temperature to take into account the effect of systemic cardiovascular effects on pulse propagation time.

[0157] Clause 5. The wearable device according to any of the preceding clauses, wherein one or more calibration sensors include a bioimpedance sensor configured to measure hydration level as a calibration parameter, and the processor is further configured to generate a blood pressure measurement based on the hydration level to take into account the effect of arterial stiffness on pulse propagation time.

[0158] Clause 6. The wearable device according to any of the preceding clauses, wherein one or more calibration sensors include a photoplethysmography (PPG) sensor configured to acquire PPG data, and the processor is further configured to: determine the ratio of pulsatile blood flow to non-pulsatile blood flow based on the PPG data, calculate a perfusion index as a calibration parameter; and generate a blood pressure measurement based on the perfusion index to take into account the effect of peripheral circulation on pulse propagation time.

[0159] Clause 7. A wearable device according to any of the preceding clauses, wherein one or more calibration sensors include an electromyography (EMG) sensor configured to measure muscle activity as a calibration parameter by detecting electrical signals associated with smooth muscle tissue, and the processor is further configured to generate a blood pressure measurement based on muscle activity to take into account the effect of arterial wall contractility on pulse propagation time.

[0160] Clause 8. The wearable device according to any of the preceding clauses, wherein the processor is further configured to generate the blood pressure measurement by: calculating an initial blood pressure measurement based on pulse propagation time, and applying a correction factor to the initial blood pressure measurement based on the correction parameter in response to detecting that a correction parameter exceeds a threshold.

[0161] Clause 9. The wearable device according to any of the preceding clauses, wherein the processor is further configured to: determine systolic and diastolic blood pressure values ​​based on one or more characteristics and correction parameters of pulse propagation time, so as to include the systolic and diastolic blood pressure values ​​as part of the blood pressure measurement.

[0162] Clause 10. A method implemented by a processing device, the method comprising: receiving physiological time-series data indicating pulse propagation time along a cardiovascular pathway and measurements of a correction parameter affecting the pulse propagation time independently of corresponding blood pressure changes; processing the physiological time-series data to determine the pulse propagation time based on the time difference between corresponding points in the physiological time-series data; generating a blood pressure measurement based on the pulse propagation time and the measurements of the correction parameter; and outputting the blood pressure measurement.

[0163] Clause 11. The method according to Clause 10, wherein the correction parameter includes at least one of body temperature, skin temperature, perfusion index, hydration level, or muscle activity.

[0164] Clause 12. The method according to Clause 10 or 11, wherein the physiological time series data includes a first physiological signal and a second physiological signal, and processing the physiological time series data includes: determining one or more of the pulse conduction time (PTT) or pulse arrival time (PAT) as the pulse propagation time based on the first physiological signal and the second physiological signal.

[0165] Clause 13. The method according to any one of Clauses 10 to 12, wherein the physiological timing data includes electrocardiogram (ECG) signals and photoplethysmography (PPG) signals.

[0166] Clause 14. The method according to any one of Clauses 10 to 13, wherein generating a blood pressure measurement includes: generating an initial blood pressure measurement based on pulse propagation time, and adjusting the initial blood pressure measurement based on the measurement of a correction parameter.

[0167] Clause 15. The method according to any one of Clauses 10 to 14, wherein generating a blood pressure measurement includes: in response to detecting that a correction parameter exceeds a threshold, applying a correction factor to the blood pressure measurement based on the measured value of the correction parameter.

Claims

1. A wearable device for cuffless blood pressure monitoring, comprising: Sensors are arranged to collect physiological time-series data indicating the time it takes for the pulse to travel along the cardiovascular pathway; One or more calibration sensors are configured to measure calibration parameters that affect the pulse propagation time independently of corresponding blood pressure changes; as well as The processor is configured as follows: Processing the physiological time-series data to determine the pulse propagation time; and The blood pressure measurement value is generated based on the pulse propagation time and the measurement value of the correction parameter.

2. The wearable device according to claim 1, wherein, The sensor arrangement includes: a proximal sensor located at a proximal position along the cardiovascular pathway and configured to detect a first cardiovascular signal as part of the physiological time-series data; and a distal sensor located at a distal position along the cardiovascular pathway and configured to detect a second cardiovascular signal as part of the physiological time-series data and corresponding to the first cardiovascular signal.

3. The wearable device according to claim 1 or 2, wherein, The one or more calibration sensors include one or more of a temperature sensor or a heat flux sensor, the temperature sensor or the heat flux sensor being configured to measure skin temperature adjacent to the sensor arrangement as the calibration parameter, and the processor being further configured to generate the blood pressure measurement based on the skin temperature to take into account the effect of vasomotor activity on the pulse propagation time.

4. The wearable device according to any of the preceding claims, wherein, The one or more calibration sensors include one or more of a temperature sensor or a heat flux sensor, the temperature sensor or the heat flux sensor being configured to measure body temperature as the calibration parameter, and the processor being further configured to generate the blood pressure measurement based on the body temperature to take into account the influence of systemic cardiovascular effects on the pulse propagation time.

5. The wearable device according to any of the preceding claims, wherein, The one or more calibration sensors include a bioimpedance sensor configured to measure hydration level as the calibration parameter, and the processor is further configured to generate the blood pressure measurement based on the hydration level to take into account the effect of arterial stiffness on the pulse propagation time.

6. The wearable device according to any of the preceding claims, wherein, The one or more calibration sensors include a photoplethysmography (PPG) sensor, the PPG sensor being configured to acquire PPG data, and the processor being further configured to: The perfusion index is calculated as the correction parameter by determining the ratio of pulsatile to non-pulsatile blood flow based on the PPG data; and The blood pressure measurement is generated based on the perfusion index to account for the influence of peripheral circulation on the pulse propagation time.

7. The wearable device according to any of the preceding claims, wherein, The one or more calibration sensors include an electromyography (EMG) sensor configured to measure muscle activity as the calibration parameter by detecting electrical signals associated with smooth muscle tissue, and the processor is further configured to generate the blood pressure measurement based on the muscle activity to account for the effect of arterial wall contractility on the pulse propagation time.

8. The wearable device according to any of the preceding claims, wherein, The processor is also configured to generate the blood pressure measurement by: calculating an initial blood pressure measurement based on the pulse propagation time, and applying a correction factor to the initial blood pressure measurement based on the correction parameter in response to detecting that the correction parameter exceeds a threshold.

9. The wearable device according to any of the preceding claims, wherein, The processor is further configured to determine systolic and diastolic blood pressure values ​​based on one or more characteristics of the pulse propagation time and the correction parameters, so as to include the systolic and diastolic blood pressure values ​​as part of the blood pressure measurement.

10. A method implemented by a processing device, the method comprising: Receive physiological time-series data indicating the pulse propagation time along the cardiovascular pathway, as well as measurements of correction parameters that affect the pulse propagation time independently of corresponding blood pressure changes; The physiological time series data is processed to determine the pulse propagation time based on the time difference between corresponding points in the physiological time series data; Blood pressure measurements are generated based on the pulse propagation time and the measured values ​​of the correction parameters; and Output the blood pressure measurement value.

11. The method according to claim 10, wherein, The correction parameters include at least one of body temperature, skin temperature, perfusion index, hydration level, or muscle activity.

12. The method according to claim 10 or 11, wherein, The physiological time series data includes a first physiological signal and a second physiological signal, and processing the physiological time series data includes: determining one or more of the pulse conduction time (PTT) or pulse arrival time (PAT) as the pulse propagation time based on the first physiological signal and the second physiological signal.

13. The method according to any one of claims 10 to 12, wherein, The physiological time-series data includes electrocardiogram (ECG) signals and photoplethysmography (PPG) signals.

14. The method according to any one of claims 10 to 13, wherein, Generating the blood pressure measurement includes: generating an initial blood pressure measurement based on the pulse propagation time, and adjusting the initial blood pressure measurement based on the measurement value of the correction parameter.

15. The method according to any one of claims 10 to 14, wherein, Generating the blood pressure measurement includes: in response to detecting that the correction parameter exceeds a threshold, applying a correction factor to the blood pressure measurement based on the measured value of the correction parameter.

Images