Wearable blood pressure sensor

Abstract

A smartwatch-integrable, low-power wearable antenna biosensor measuring RF bioimpedance (BioZ) changes from radial artery blood flow using radio frequencies (RF) for continuous blood pressure estimation is provided (100). This non-contact RF BioZ-based approach (300) offers a more equitable solution across diverse demographics than current solutions. The wearable device has shown performance equivalent to clinical pulse oximeter PPG sensor, with mean absolute errors (MAE) of 2.41 ± 3.2 mmHg for RF BioZ and 2.52 ± 2.54 mmHg for PPG for systolic blood pressure (SBP). It meets British Hypertension Society (BHS) and Association for the Advancement of Medical Instrumentation (AAMI) error threshold requirements for grade A / passing score for non-cuff-based BP measuring technology, respectively.

Description

WEARABLE BLOOD PRESSURE SENSORCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims benefit of U.S. Provisional Application Serial No. 63 / 675,895, filed on Inly 26, 2024, and titled “Towards Clinically Accurate Continuous Blood Pressure Smart Watch Biosensor Using Equitable Bioimpedance Modality and Al.” The disclosure of which is hereby incorporated by reference in its entirety.BACKGROUND

[0002] Photoplethysmography (PPG), a technology used in wrist watches and rings to monitor heart rate and other vital signs, has gained popularity due to its non-invasive nature and convenience. However, its effectiveness can be compromised for individuals with darker skin tones, raising concerns about the accuracy of health monitoring and the potential for biased Al algorithms.

[0003] PPG relies on detecting changes in light absorption by red blood cells, which is influenced by skin color. The technology typically uses green or infrared light to measure these changes and assess heart rate, oxygen saturation, and other parameters. In individuals with darker skin tones, the increased melanin concentration can affect the amount of light absorbed and scattered, leading to less accurate readings. Darker skin absorbs more light, which can interfere with the PPG sensors’ ability to detect subtle changes in blood volume, resulting in less reliable data.

[0004] This issue has broader implications for the accuracy of Al algorithms that analyze PPG data. These algorithms are often trained on datasets that may not be sufficiently diverse, leading to biases in their performance. If the training data mostly includes individuals with lighter skin tones, the algorithms may not be calibrated to account for the variations in lightabsorption and scattering in darker skin tones. Consequently, this can lead to systematic inaccuracies in health monitoring for individuals with darker skin, exacerbating health disparities and inequities.

[0005] Another problem associated with PPG is that the PPG sensors always need to be in contact with the wearer. This makes data integrity highly user dependent on the fit of the associated device. Furthermore, the contact requirement also makes these sensors more susceptible to motion artifacts making their usage in high intensity ambulatory conditions difficult.SUMMARY

[0006] A wearable low-power wearable antenna biosensor for continuous blood pressure (BP) estimation using RF bioimpedance (BioZ) changes from radial artery blood flow is provided. The biosensor can be integrated into wearable devices such as smartwatches or rings. The proposed biosensor addresses critical limitations in current wearable BP monitoring technologies, particularly those using PPG.

[0007] The proposed biosensor extracts arterial RF BioZ using an antenna, relying on the reflection of an incident RF signal at the tissue-artery interface. By leveraging existing smartwatch antennas for RF BioZ measurements, the proposed biosensor reduces integration complexity and provides a more stable platform. Preliminary results show that the RF BioZ sensor performs comparably to pulse oximeter PPG sensor while offering advantages such as enhanced wearability through noncontact format, lower power consumption, and the ability to be used without skin color-influenced data corruption. The proposed biosensor works with many types of antennas in general, not just Wi-Fi antennas, and in both skin contact and noncontact applications .

[0008] This RF-based approach not only improves technical aspects of cuffless BP monitoring but also addresses the potential for Al bias in wearable health technology. By providing more consistent raw data across diverse populations, it creates a foundation for developing non-biased Al algorithms, ultimately leading to more equitable health monitoring solutions. Furthermore, because contact is not required, there are fewer motion artifacts on the collected data. In addition, the proposed biosensor can access deeper seated artery locations than previous techniques, which makes the biosensor more resistant to perfusion issues at cold temperatures than current devices.

[0009] In some aspects, the techniques described herein relate to a biosensor including: an antenna adapted to transmit a signal towards an artery of a wearer and to receive a reflected signal in return; and a processing device adapted to: measure a bioimpedance based on the transmitted signal and the reflected signal; and based on the measured bioimpedance, estimate a blood pressure of the wearer.

[0010] In some aspects, the techniques described herein relate to a biosensor, wherein the antenna is a Wi-Fi antenna or a Bluetooth antenna. The antenna may be a commercially available antenna typically used in smartwatches or other wearable devices. Other types of antennas may be supported.

[0011] In some aspects, the techniques described herein relate to a biosensor, wherein the biosensor is part of smartwatch.

[0012] In some aspects, the techniques described herein relate to a biosensor, wherein the processing device is further adapted to: measure a change in RF bioimpedance based on the transmitted signal and the reflected signal; and based on the measured change in RF bioimpedance, estimate a change in blood pressure of the wearer. Other parameters may be measured including HRV, BP and other such cardiovascular parameters.

[0013] In some aspects, the techniques described herein relate to a biosensor, wherein the processing device is further adapted to: receive an Al model; measure RF bioimpedance based on the transmitted signal and the reflected signal; and based on the measured RF bioimpedance and the Al model, estimate a blood pressure of the wearer. Other measured parameters may include heart rate, heart rate variability, respiration rate, etc.

[0014] In some aspects, the techniques described herein relate to a biosensor, wherein the Al model is a machine learning model. The model may also be a physics-informed neural network model or a mix of both.

[0015] In some aspects, the techniques described herein relate to a biosensor, wherein the antenna is an inverted F type Wi-Fi antenna, or other type of antenna.

[0016] In some aspects, the techniques described herein relate to a biosensor, wherein the antenna is not in skin contact with the wearer. Alternatively, the sensor may be in skin contact with the wearer.

[0017] In some aspects, the techniques described herein relate to a biosensor, wherein the measurement is not affected by a skin color of the wearer.

[0018] In some aspects, the techniques described herein relate to a biosensor, wherein the biosensor is part of a smart watch. The biosensor may be part of any wearable device such as patches, rings, bracelets, head bands, etc.

[0019] In some aspects, the techniques described herein relate to a biosensor, wherein the processing device is further adapted to: based on the measured RF bioimpedance, estimate one or more of a mood, a hydration, heart rate, heart rate variability, arterial stiffness, stroke volume, or a stress level of the wearer.

[0020] In some aspects, the techniques described herein relate to a wearable smart device including: an antenna adapted to transmit a signal towards an artery of a wearer and toreceive a reflected signal in return; and a processing device adapted to: measure an RF bioimpedance based on the transmitted signal and the reflected signal; and based on the measured bioimpedance, estimate a blood pressure of the wearer.

[0021] In some aspects, the techniques described herein relate to a wearable smart device, wherein the antenna is a Wi-Fi antenna or a Bluetooth antenna. The antenna may be a commercially available antenna typically used in smartwatches or other wearable devices, or other antenna designs which can be adapted for wearability .

[0022] In some aspects, the techniques described herein relate to a wearable smart device, wherein the wearable smart device is a smartwatch. Other wearable devices may be supported.

[0023] In some aspects, the techniques described herein relate to a wearable smart device, wherein the processing device is further adapted to: measure a change in RF bioimpedance based on the transmitted signal and the reflected signal; and based on the measured change in RF bioimpedance, estimate a change in blood pressure of the wearer. Other cardiac parameters may be estimated.

[0024] In some aspects, the techniques described herein relate to a wearable smart device, wherein the processing device is further adapted to: receive an Al model; measure RF bioimpedance based on the transmitted signal and the reflected signal; and based on the measured RF bioimpedance and the Al model, estimate a blood pressure of the wearer.

[0025] In some aspects, the techniques described herein relate to a wearable smart device, wherein the Al model is a machine learning model. The model may also be a PINN model, or a mixture of both types of models.

[0026] In some aspects, the techniques described herein relate to a wearable smart device, wherein the antenna is an inverted F type Wi-Fi antenna. Other types of antennas, including Bluetooth antennas, may be supported.

[0027] In some aspects, the techniques described herein relate to a wearable smart device, wherein the antenna is not in skin contact with the wearer.

[0028] In some aspects, the techniques described herein relate to a wearable smart device, wherein the measurement is not affected by the skin color of the wearer.

[0029] In some aspects, the techniques described herein relate to a wearable smart device, wherein the wearable smart device is a wearable ring.

[0030] In some aspects, the techniques described herein relate to a wearable smart device, wherein the processing device is further adapted to: based on the measured RF bioimpedance, estimate one or more of a mood, a hydration, arterial stiffness, stroke volume or a stress level of the wearer.

[0031] Additional features will be set forth in part in the description which follows or may be learned by practice. The features will be realized and attained by means of the elements and combinations particularly pointed out in the appended claims. It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive, as claimed.BRIEF DESCRIPTION OF THE DRAWINGS

[0032] FIG. 1 is a diagram of an example wearable device including a sensor for estimating the blood pressure of a wearer;

[0033] FIG. 2 is a block diagram of a system for training a model to estimate the blood pressure of a wearer using bioimpedance;

[0034] FIG. 3 is a block diagram of an example method for estimating the blood pressure of a wearer using a bioimpedance chip and a model;

[0035] FIG. 4 is an operational flow of an implementation of a method for training a model to estimate the blood pressure of a wearer;

[0036] FIG. 5 is an operational flow of an implementation of a method for selecting frequency to use to estimate the blood pressure of a wearer;

[0037] FIG. 6 shows an exemplary computing environment in which example embodiments and aspects may be implemented.

[0038] Various objects, aspects, and features of the disclosure will become more apparent and better understood by referring to the detailed description taken in conjunction with the accompanying drawings, in which like reference characters identify corresponding elements throughout. In the drawings, like reference numbers generally indicate identical, functionally similar, and / or structurally similar elements.DETAILED DESCRIPTION

[0039] FIG. 1 is a diagram of an example wearable device including a sensor for estimating the blood pressure of a wearer. The wearable device 100 may be a smartphone, ring, or a set of glasses or goggles. Other types of wearable devices may be supported.

[0040] As shown, the wearable device 100 includes a bioimpedance chip 120. The bioimpedance chip 120 is also referred to herein as a biosensor and may be configured to estimate the blood pressure of a wearer of the wearable device 100. The estimated blood pressure may then be displayed to the wearer of the wearable device, may be recorded for later use and / or analysis, or may be transmitted to one or more health providers. Other uses may be supported. In some embodiments, the processing may be performed remotely by a server or cloud-based processing component.

[0041] The bioimpedance chip 120 may be configured to estimate the blood pressure of the wearer based on the RF bioimpedance (“BioZ) of blood flow of the wearer. RF bioimpedance as used herein in a measure of the opposition to the flow of electrical current through the tissue of the wearer. In particular, the bioimpedance chip 120 may use a determined change in RF bioimpedance to estimate the blood pressure of the wearer. A change in RF bioimpedance may be caused by a change in blood volume, and therefore a measured change in RF bioimpedance may be indicate a corresponding change in blood pressure of the wearer.

[0042] As shown, the bioimpedance chip 120 includes an RF antenna 110, a processing component 130, and a model 140. The processing component 130 may include one or more processors. The RF antenna may be a Wi-Fi antenna and may be adapted to transmit and receive Wi-Fi signals. An example antenna 110 may be an F-type Wi-Fi antenna. Other types of Wi-Fi antennas may be used. Note that the RF antenna 110 is not limited to Wi-Fi antennas, and other types of electromagnetic waves and corresponding antenna types may be used such as Bluetooth and other appropriate antenna geometries.

[0043] In some embodiments, the bioimpedance chip 120 may be a stand-alone chip and may be integrated into a wearable device 100 by a manufacturer of the wearable device 100. Alternatively, the bioimpedance chip 120 may utilize existing components of the wearable device 100. For example, the bioimpedance chip 120 may leverage an existing RF antenna 110 that is already part of the wearable device 100 and may utilize processing resources of the wearable device 100 for the processing component 130.

[0044] The processing component 130 may use the RF antenna 110 to determine a change of RF bioimpedance of the wearer of the wearable device 100. In some embodiments, the processing component 130 may measure the change in RF bioimpedance by transmitting a signal towards the wearer and then measuring a strength of a corresponding receivedreflection signal. The strength of the transmitted signal and the strength of the received corresponding reflection signal may be used to measure the RF bioimpedance of the wearer. The measured RF bioimpedance may be compared to a previously received RF bioimpedance to determine the change in RF bioimpedance.

[0045] In some embodiment, when measuring the RF bioimpedance, the RF antenna may transmit the signals towards a “tissue-artery” interface of the wearer. An example tissueartery interface may be above a radial artery of a wrist of the wearer. Other locations may be supported and may include locations where blood vessels other than arteries are located.

[0046] The bioimpedance chip 120 may use the change in RF bioimpedance and a model 140 to estimate the blood pressure of the wearer. The model 140 may be a machine learning model that is trained to estimate the blood pressure of the wearer based on the change in measured bioimpedance. The model may be a machine learning regression model, a physics- informed neural network model, or a combination of both. However, other types of models may be used. The model may be individual specific and trained using data collected from the wearer of the bioimpedance chip 120. Alternatively or additionally the model may be a global model that may apply to multiple individuals.

[0047] In some embodiments, before measuring the RF bioimpedance for a wearer the bioimpedance chip 120 may select a frequency to use for the transmitted signal in an initial calibration step. As may be appreciated, different individuals may respond better to different RF frequencies. Accordingly, the bioimpedance chip 120 may do a frequency sweep from approximately 1.8 - 2.4 ghz and may select the frequency that provides the strongest reflection signal. Other frequencies may be considered. The selected frequency may be stored for the wearer and may be used for subsequent measurements. Depending on the embodiment, the selected frequency may depend on the antenna geometry and may beconstricted to a 100MHz range. In other embodiments, a set frequency may be used along with a matching impedance network to automatically tune for possible frequency drift.

[0048] In some embodiments, the frequency may be reselected every few time intervals. The re-selection may be performed more or less frequently depending on the activity level of the wearer and the overall signal quality. The frequency may be reselected when a corresponding RF bioimpedance fails a quality check. For example, if an RF bioimpedance is below a threshold or is less than some percentage of a previous measurement, the bioimpedance chip 120 may determine that a new frequency should be selected. Other methods for determining when to select a new frequency may be used. Alternatively, instead of frequency selection a matching impedance network may be tuned at regular intervals to allow for constant frequency operation.

[0049] As will be described further with respect to FIG. 2, the model 140 may have been trained using training data that includes measured bioimpedances and corresponding blood pressure measurements taken with a blood pressure measuring device. In some embodiments, the training data may be taken from a variety of different wearers in a variety of scenarios. Alternatively, the training data may be taken only from the wearer who will use model 140. Once trained, the model 140 may be provided and stored in memory of the bioimpedance chip 120 and / or the wearable device 100. The model 140 may be periodically updated as new training data is received and the model 140 is improved.

[0050] FIG. 2 is a block diagram of a system 200 for training a model to estimate the blood pressure of a wearer using bioimpedance. As shown, the system 200 includes one or more components including a bioimpedance chip 120, a blood pressure monitor 220, and a computing device 210. The computing device 210 may be a general -purpose computing device such as the computing device 600 illustrated with respect to FIG. 6.

[0051] As shown, the bioimpedance chip 120 may include an analog front end 122 and a digital front end 123. The analog front end 122 may handle initial analog signal conditioning (filtering, amplification, impedance matching). The analog signal may then be and then be converted to digital by an Analog-to-Digital Converter (ADC). The digital output of the ADC is then fed into the digital front end 123 which performs further digital signal processing.

[0052] The system 200 may be configured to generate training data 230 that may be used to generate the model 140. The training data 230 may include RF data 211 generated by the bioimpedance chip 120 and BP data 221 generated by the blood pressure monitor 220. The RF data 211 may be impedance data and may include the RF bioimpedance measurements (generated by the bioimpedance chip 120 as described previously with respect to FIG. 1).The BP data 221 may be blood pressure measurements generated by the blood pressure monitor 220. The blood pressure monitor 220 may be a continuous blood pressure monitor such as the Finapress NOVA. Other blood pressure monitors may be used.

[0053] To generate the training data 230, the bioimpedance chip 120 and the blood pressure monitor 220 may be simultaneously connected to a subject. The RF data 211 and BP data 221 may then be collected for the subject by the computing device 210. Each piece of RF data 21 1 and BP data 221 may be time stamped such that the computing device 210 can determine which RF bioimpedance measurements correspond to which blood pressure measurements. Depending on the embodiment, the subject may be connected for a standardized period of time such as 30 minutes or one hour. Other time periods may be selected. The RF data 211 and BP data 221, and associated time stamps, may be collected for the subject over the period of time.

[0054] Depending on the embodiment, while connected to the bioimpedance chip 120 and blood pressure monitor 220, the subject may be asked to perform certain activities or exercises to vary the blood pressure of the subject. These exercises may include resting, squeezing, or breathing exercises. For example, in one embodiment, the training data 230 collection was divided into four trials. In each trial the subject was asked to rest for one minute, perform hand squeezing for four minutes, dip their hand in an ice bath for 30 seconds, and then perform four successive Valsalva maneuvers. Other activities or exercises may be used.

[0055] Further, in embodiments where one model 140 is used for multiple wearers, the training data 230 generation procedure described above may be repeated for a variety of different subjects. The subjects may be selected such that many different demographics of subjects are represented including age, race, sex, weight, height, etc. Selecting a diverse group of subjects may help ensure that the model 140 trained using the training data 230 is applicable to a large number of individuals.

[0056] After collecting the training data 230, the computing device 210 may use the training data 230 to train the model 140. The model 140 may be trained to receive RF data 211 from the bioimpedance chip 120 (i.e., impedance data), and to output a predicted blood pressure. For individual specific models 140, once trained on the individual, the model 140 may be uploaded or provided to the bioimpedance chip 120 used by the individual. For more broadly applicable models 140, the models 140 may be distributed to the various bioimpedance chips 120 for use.

[0057] FIG. 3 is an illustration of a method 300 for predicting the blood pressure of a wearer of a bioimpedance chip 120. The method 300 may be implemented by the bioimpedance chip 120.

[0058] At 301, a signal is transmitted. The signal may be RF power being sent to the tissue which gets modified due to impedance changes due to artery pulsation and gets reflected back. The signal may be transmitted by an RF antenna 110 of a bioimpedance chip 120. Other types of RF signals may be used. The bioimpedance chip 120 may be part of a wearable device 100 such as a smartwatch or ring. The signal may be transmitted towards the skin of a wearer of the wearable device 100.

[0059] At 303, a reflected signal corresponding to the transmitted signal is received. The reflected signal may be received by the RF antenna 110 of the bioimpedance chip 120. The reflected signal may be a reflection of the signal transmitted at 301. The reflected signal may have been reflected off of the skin of the wearer.

[0060] At 305, an RF bioimpedance is measured based on the transmitted signal and the received signal. The RF bioimpedance may be measured by the processing component 130 of the bioimpedance chip 120. The RF bioimpedance may be measured based on a comparison of the strength of the transmitted signal with the strength of the received reflected signal.

[0061] At 307, the measured RF bioimpedance is provided. The measured bioimpedance may be provided by the processing component 130 to the model 140. Depending on the embodiment, instead of the measured RF bioimpedance, the processing component 130 may provide a change in RF bioimpedance. The change in bioimpedance may be a difference between the currently measured RF bioimpedance and a previously measured RF bioimpedance.

[0062] At 309, an estimated blood pressure is received. The estimated blood pressure may be received from the model 140 by the processing component 130. The estimated blood pressure may be the output of the model 140 and may be based on the measuredbioimpedance (or change in bioimpedance. The processing component 130 may then provide the estimated blood pressure for storage by a health monitoring application associated with the wearer of the wearable device 100, may cause the estimated blood pressure to be displayed to the wearer of the wearable device 100, or may provide the estimated blood pressure to a health care provider selected by the wearer of the wearable device 100.

[0063] FIG. 4 is an operational flow of an implementation of a method 400 for training a model to estimate the blood pressure of a wearer. The method 400 may be implemented by the computing device 210, or remoting by a server or cloud-based computing device.

[0064] At 401, impedance data and blood pressure data are collected. The impedance data may be the RF data 211 and may be collected along with the blood pressure data 221 by the computing device 210. The impedance data may be the RF bioimpedance measurements received from the bioimpedance chip 120 and the blood pressure data 221 may be collected by a blood pressure monitor 220. In some embodiments, each individual wearer may have their own trained model 140 trained only using data collected from them. Alternatively, a universal model may be used.

[0065] At 403, a model is trained using the collected impedance data and blood pressure data. The model 140 may be trained by the computing device 210. In some embodiments, a portion of the training data 230 may be set aside for verification and adjustment of the trained model 140.

[0066] At 405, the model is adjusted using additional impedance and blood pressure data. The model 140 may be adjusted by the computing device 210 using the portion of the training data 230 not used to train the model 140.

[0067] At 407, the trained model is provided. The trained model 140 may be provided for use by the bioimpedance chip 120 of the wearable device 100 by the individual.

[0068] FIG. 5 is an operational flow of an implementation of a method for maintaining the operational frequency of the antenna over long periods of time or during motion artifacts experienced by the wearer. The method 500 may be implemented by the bioimpedance chip 120 and / or a cloud-based computing system.

[0069] At 501 , a signal is acquired using the selected frequency. The signal may be acquired by the bioimpedance chip 120. The signal may be used to estimate the blood pressure or other cardiac parameters of a wearer of a device that includes the bioimpedance chip 120.

[0069] At 503, a quality of the signal is checked. The quality of the signal may be checked by the bioimpedance chip 120. The quality of the signal may be checked by looking for variance and drift in signal quality over 30 beats and may include measuring some or all of the following parameters: Amplitude Range; Standard Deviation; Skewness; Kurtosis; Signal-to-Noise Ratio (SNR); Power Band Ratio (e.g., 0.5-4 Hz vs total power); Spectral Entropy; Peak Symmetry; Template Correlation; Dynamic Time Warping (DTW) Distance; Envelope Standard Deviation; High Derivative Ratio; Clipping Detection; Baseline Wander; Variance; and Outlier Ratio (e.g., % of samples beyond ±3 SD). Other methods for checking the quality of a signal may be used. If the signal passes the quality check, then the method 500 may return to 501 where a next signal may be acquired by the bioimpedance chip 120. If the signal fails the signal quality check, then the bioimpedance chip 120 may continue at 505.

[0070] At 505, the matching network is tuned. The matching network may be tuned by the bioimpedance chip 120 to recenter to the operational frequency of the antenna, leading to the best impedance matching between the antenna and the tissue for acceptable signal quality. After the tunning the method 500 may return to 501.

[0071] FIG. 6 shows an exemplary computing environment in which example embodiments and aspects may be implemented. The computing device environment is only one example of a suitable computing environment and is not intended to suggest any limitation as to the scope of use or functionality.

[0072] Numerous other general purpose or special purpose computing devices environments or configurations may be used. Examples of well-known computing devices, environments, and / or configurations that may be suitable for use include, but are not limited to, personal computers, server computers, handheld or laptop devices, multiprocessor systems, microprocessor-based systems, network personal computers (PCs), minicomputers, mainframe computers, embedded systems, distributed computing environments that include any of the above systems or devices, and the like.

[0073] Computer-executable instructions, such as program modules, being executed by a computer may be used. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. Distributed computing environments may be used where tasks are performed by remote processing devices that are linked through a communications network or other data transmission medium. In a distributed computing environment, program modules and other data may be located in both local and remote computer storage media including memory storage devices.

[0074] With reference to FIG. 6, an exemplary system for implementing aspects described herein includes a computing device, such as computing device 600. In its most basic configuration, computing device 600 typically includes at least one processing unit 602 and memory 604. Depending on the exact configuration and type of computing device, memory 604 may be volatile (such as random access memory (RAM)), non-volatile (such as read-onlymemory (ROM), flash memory, etc.), or some combination of the two. This most basic configuration is illustrated in FIG. 6 by dashed line 606.

[0075] Computing device 600 may further include the analog front end 122. The analog front end 122 may interface with the processing unit 602 and one or more RF antennas 613.

[0076] Computing device 600 may have additional features / functionality. For example, computing device 600 may include additional storage (removable and / or non-removable) including, but not limited to, magnetic or optical disks or tape. Such additional storage is illustrated in FIG. 6 by removable storage 608 and non-removable storage 610.

[0077] Computing device 600 typically includes a variety of computer readable media. Computer readable media can be any available media that can be accessed by the device 600 and includes both volatile and non-volatile media, removable and non-removable media.

[0078] Computer storage media include volatile and non-volatile, and removable and nonremovable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Memory 604, removable storage 608, and non-removable storage 610 are all examples of computer storage media. Computer storage media include, but are not limited to, RAM, ROM, electrically erasable program read-only memory (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by computing device 600. Any such computer storage media may be part of computing device 600.

[0079] Computing device 600 may contain communication connection(s) 612 that allow the device to communicate with other devices. Computing device 600 may also have inputdevice(s) 614 such as a keyboard, mouse, pen, voice input device, touch input device, etc. The communication connection(s) 612 may include one or more antenna including the RF antenna 613. Output device(s) 616 such as a display, speakers, printer, etc. may also be included. All these devices are well known in the art and need not be discussed at length here.

[0080] It should be understood that the various techniques described herein may be implemented in connection with hardware components or software components or, where appropriate, with a combination of both. Illustrative types of hardware components that can be used include Field-programmable Gate Arrays (FPGAs), Application-specific Integrated Circuits (ASICs), Application-specific Standard Products (ASSPs), System-on-a-chip systems (SOCs), Complex Programmable Logic Devices (CPLDs), etc. The methods and apparatus of the presently disclosed subject matter, or certain aspects or portions thereof, may take the form of program code (i.e., instructions) embodied in tangible media, such as floppy diskettes, CD-ROMs, hard drives, or any other machine-readable storage medium where, when the program code is loaded into and executed by a machine, such as a computer, the machine becomes an apparatus for practicing the presently disclosed subject matter.

[0081] Although exemplary implementations may refer to utilizing aspects of the presently disclosed subject matter in the context of one or more stand-alone computer systems, the subject matter is not so limited, but rather may be implemented in connection with any computing environment, such as a network or distributed computing environment. Still further, aspects of the presently disclosed subject matter may be implemented in or across a plurality of processing chips or devices, and storage may similarly be effected across a plurality of devices. Such devices might include personal computers, network servers, and handheld devices, for example.

[0082] Although the subject matter has been described in language specific to structural features and / or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.

Claims

WHAT IS CLAIMED IS:

1. A biosensor comprising : an antenna adapted to transmit a signal towards an artery of a wearer and to receive a reflected signal in return; and a processing device adapted to: measure an RF bioimpedance based on the transmitted signal and the reflected signal; and based on the measured RF bioimpedance, estimate a blood pressure of the wearer.

2. The biosensor of claim 1 , wherein the antenna is a Wi-Fi antenna.

3. The biosensor of claim 1 , wherein the biosensor is part of smartwatch.

4. The biosensor of claim 1 , wherein the processing device is further adapted to: measure a change in RF bioimpedance based on the transmitted signal and the reflected signal; and based on the measured change in RF bioimpedance, estimate a change in blood pressure of the wearer.

5. The biosensor of claim 1 , wherein the processing device is further adapted to: receive an Al model; measure an RF bioimpedance based on the transmitted signal and the reflected signal; and based on the measured RF bioimpedance and the Al model, estimate a blood pressure of the wearer.

6. The biosensor of claim 5, wherein the Al model is a machine learning model.

7. The biosensor of claim 1 , wherein the antenna is an inverted F type Wi-Fi antenna.

8. The biosensor of claim 1 , wherein the antenna is not in skin contact with the wearer.

9. The biosensor of claim 1 , wherein the measurement is not affected by a skin color of the wearer.

10. The biosensor of claim 1 , wherein the biosensor is part of a wearable ring.

11. The biosensor of claim 1 , wherein the processing device is further adapted to: based on the measured RF bioimpedance, estimate one or more of a mood, a pulse, a stroke volume, a hydration, or a stress level of the wearer.

12. A wearable smart device comprising: an antenna adapted to transmit a signal towards an artery of a wearer and to receive a reflected signal in return; and a processing device adapted to: measure an RF bioimpedance based on the transmitted signal and the reflected signal; and based on the measured RF bioimpedance, estimate a blood pressure of the wearer.

13. The wearable smart device of claim 12, wherein the antenna is a Wi-Fi antenna.

14. The wearable smart device of claim 12, wherein the wearable smart device is a smartwatch.

15. The wearable smart device of claim 12, wherein the processing device is further adapted to: measure a change in RF bioimpedance based on the transmitted signal and the reflected signal; and based on the measured change in RF bioimpedance, estimate a change in blood pressure of the wearer.

16. The wearable smart device of claim 12, wherein the processing device is further adapted to: receive an Al model; measure an RF bioimpedance based on the transmitted signal and the reflected signal; and based on the measured RF bioimpedance and the Al model, estimate a blood pressure of the wearer.

17. The wearable smart device of claim 16, wherein the Al model is a machine learning model.

18. The wearable smart device of claim 12, wherein the antenna is an inverted F type Wi-Fi antenna.

19. The wearable smart device of claim 12, wherein the antenna is not in skin contact with the wearer.

20. The wearable smart device of claim 12, wherein the processing device is further adapted to: based on the measured RF bioimpedance, estimate one or more of a mood, a hydration, a stroke volume, a pulse, stroke volume, arterial stiffness or a stress level of the wearer.

Images