Mobile device-based oscillometric blood pressure measurement

Abstract

Systems, devices, and methods for mobile device-based oscillometric blood pressure measurement are disclosed. In some aspects, a mobile device includes a vibration motor, an inertial motion unit (IMU), a camera, a processor, and a non-transitory computer readable memory, where the processor is configured to perform operations including: vibrating the vibration motor; acquiring, using the inertial measurement unit of the mobile device, IMU data indicative of vibration damping caused by a user applying a force to the camera using a body part; determining an applied force based on a force model and the IMU data; capturing photoplethysmography (PPG) data of a set of blood vessels in the body part while applying the force using the camera of the mobile device; and determining the blood pressure of the user using oscillometric techniques based on the PPG data and the applied force.

Description

[0001] Cross Reference to Related Applications

[0002] This patent document claims priority to and the benefit of U.S. Provisional Patent Application No. 63 / 385,383, filed November 29, 2022, entitled “SMARTPHONE-BASED BLOOD PRESSURE MEASUREMENT.” The entire contents of the aforementioned patent application are incorporated herein by reference as part of the disclosure of this patent document. TECHNICAL FIELD

[0003] This patent document relates to mobile device-based oscillometric blood pressure measurement. BACKGROUND

[0004] The American Heart Association (AHA) and the American Medical Association (AMA) issued a joint statement emphasizing the importance of self-measurement of blood pressure at home and explicitly stating that oscillometry is the preferred method for calculating blood pressure. Oscillometry is the standard calculation method performed by most FDA-approved blood pressure cuffs. SUMMARY

[0005] This document discloses methods, devices, and systems for blood pressure measurement using oscillometric techniques using sensor data acquired by built-in sensors of a mobile device that does not have additional accessories.

[0006] One aspect of this document relates to a mobile device configured to measure blood pressure using oscillometric techniques using sensor data acquired by built-in sensors of a mobile device that does not have accessories. The mobile device can include a vibration motor, an inertial motion unit (IMU), a camera, a processor, and a non-transitory computer-readable memory storing instructions that, when executed by the processor, cause the processor to perform at least the following: vibrate the vibration motor; acquire, by an inertial measurement unit of the mobile device, IMU data indicative of vibration dampening caused by a force applied by a body part of a user to the camera; determine the applied force based on a force model and the IMU data; capture, by the camera of the mobile device, photoplethysmography (PPG) data of a set of blood vessels in the body part while the force is applied; and determine, using oscillometric techniques, a blood pressure of the user based on the PPG data and the applied force.

[0007] One aspect of the present document relates to a method for measuring blood pressure based on oscillometric technique using sensor data acquired by in-built sensors of a mobile device that does not have an accessory. The method can include vibrating a vibration motor of the mobile device; acquiring inertial measurement unit (IMU) data using an IMU of the mobile device, the IMU data indicative of vibration damping caused by a body part of a user exerting a force on a camera of the mobile device; determining the exerted force based on a force model and the IMU data; capturing photoplethysmography (PPG) data of a set of blood vessels in the body part using a camera of the mobile device while the force is being exerted; and determining blood pressure of the user using oscillometric technique based on the PPG data and the exerted force.

[0008] Another aspect of the present document relates to one or more non-transitory computer readable media storing instructions that, when executed by one or more processors of a mobile device, cause the mobile device to perform any one or more of the solutions described herein.

[0009] The above and other aspects of the present document and embodiments thereof are more fully described in the following detailed description with reference to the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0010] Figure 1 A diagram showing an example system of the disclosed technology is shown.

[0011] Figure 2 An exemplary block diagram showing various components of a mobile device according to some embodiments of the present document is shown.

[0012] Figure 3 An example application for blood pressure measurement based on a mobile device according to some embodiments of the present disclosure is shown.

[0013] Figure 4 A flowchart showing a process for blood pressure measurement based on a mobile device according to some embodiments of the present document is shown.

[0014] Figure 5 Regression and Bland Altman plots for force measurement of three different smartphones according to some embodiments of the present document are shown.

[0015] Figure 6 (a) exemplary PPG and exerted force data measured by a mobile device and corresponding generated oscillograms according to some embodiments of the present document are shown.

[0016] Figure 6 (b) average oscillograms corresponding to various blood pressure ranges according to some embodiments of the present document are shown.

[0017] Figure 7 A diagram showing a force model according to some embodiments of the present document is shown.Figure 6 (a) and Figure 6 Confusion matrix and ROC analysis of the results in (b). DETAILED DESCRIPTION

[0018] This document describes systems, devices, and methods for measuring blood pressure using an oscillographic method that functions essentially the same as a blood pressure cuff. This method is applicable to any mobile device with a camera, a vibration motor, and an inertial measurement unit (IMU). To determine blood pressure using the oscillographic method, the device needs to: 1) apply a known force to a blood vessel, and at the same time 2) measure the local blood volume. For a blood pressure cuff, an inflatable cuff applies pressure to a set of blood vessels, while integrated sensors measure the change in blood volume within the set of blood vessels. For the mobile phone application disclosed herein, the user applies a force by pressing a body part (e.g., the user's finger) against the camera of the mobile device, while the camera can measure the blood volume within the body part using photoplethysmography (PPG) at the same time the force is applied. Vibrations can be induced by a vibration motor inside the mobile device, and the damping of these vibrations caused by the applied force is monitored using the IMU of the mobile device to quantify the applied force. The mobile device can determine the user's blood pressure based on the applied force and PPG data acquired using the built-in sensors already present in the mobile device, without hardware modifications, without physical accessories, without any prior blood pressure measurement for calibration, and can be enabled with only software by downloading an application on any mobile device, as discussed in further detail below. Thus, the technology can provide a large-scale screening opportunity for hypertension. As a software-enabled measurement technology for any mobile device with a camera, an IMU, and a vibration motor, this technology can rapidly proliferate in communities that do not have adequate blood pressure monitoring.

[0019] Widespread blood pressure screening is a key step to improving global healthcare. Hypertension, defined herein as systolic and / or diastolic blood pressure exceeding 130 / 80 mmHg, has long been a major risk factor for heart disease, stroke, and chronic kidney disease. In recent years, hypertension has been shown to accelerate cognitive decline in the elderly, and there is growing evidence that hypertension is associated with increased risk of all-cause mild cognitive impairment (MCI) and non-amnestic MCI. According to the World Health Organization, nearly half (46%) of hypertensive adults are unaware of their condition, and thus, BP screening is critical to achieving the United Nations' goal of reducing hypertension prevalence by 30% between 2010 and 2030. In addition, large-scale screening can greatly improve health literacy and enable high-risk groups to utilize non-traditional healthcare settings to address hypertension. Due to the widespread popularity of mobile devices, mobile device-based blood pressure measurement can provide a highly scalable and cost-effective screening solution.

[0020] Most previous studies of smartphone and smartwatch blood pressure measurement have used optical methods for relative blood pressure measurement, which only require PPG measurements, including, for example, pulse transit time (PTT) or pulse wave analysis (PWA). However, these techniques are fundamentally different from the mobile device-based oscillometric blood pressure measurement techniques disclosed herein. PTT and PWA only provide relative changes in blood pressure, but cannot provide the valuable systolic and diastolic pressure measurements that physicians commonly use. Instead, blood pressure cuff measurements need to be taken daily or weekly to calibrate the measurements. These relative change measurements assume that the individual has access to a blood pressure cuff (or other absolute blood pressure measurement); otherwise, the relative measurements cannot be meaningfully interpreted. Existing smartphone-based absolute blood pressure measurements typically employ custom accessories (e.g., added force sensors) to measure force.

[0021] The mobile device-based oscillometric blood pressure measurements disclosed herein do not require accessories and operate on any model of mobile device that has a vibration motor, camera, and IMU. The oscillometric blood pressure measurements are not "optical" blood pressure measurements that rely on pulse transit time (PTT), pulse arrival time (PAT), pulse wave analysis (PWA), or any other technique that estimates relative changes in blood pressure. The present application can make absolute blood pressure measurements based on the oscillometric method used in clinical blood pressure cuffs and recommended by the American Medical Association (AMA) and the American Heart Association (AHA). As discussed elsewhere in this document, blood pressure measurements based on oscillometric techniques require measurement of two key indicators: applied force and local blood volume. Local blood volume of an artery can be accurately estimated via photoplethysmography (PPG) using the camera of a smartphone. Measuring applied force without an accessory is key to enabling the blood pressure measurements disclosed herein. By relying on components common to different types of mobile devices, the oscillometric blood pressure measurements disclosed herein work across different types of mobile devices (e.g., types can correspond to manufacturer's models).

[0022] Some embodiments of the present document disclose a force damping technique that utilizes only the vibration motor and IMU of a mobile device to measure applied force. Most mobile devices already contain an IMU with an accelerometer and gyroscope for other purposes, including, for example, measuring linear acceleration and angular velocity so that these mobile devices can perform position inference, which are widely used for rotating screens, playing games, counting steps, gesture recognition, etc. Similarly, most mobile devices already contain a vibration motor to provide haptic feedback for messages, phone calls, and screen interactions, etc. To measure force, the mobile device can be set to vibrate (e.g., maximum vibration) while the IMU measures the motion of the device as it oscillates (caused by the vibration). When force is applied during the vibration, the IMU signal changes as the oscillation is altered or damped. According to embodiments of the present document, the vibration damping is measured and relied upon to model the applied force as a damping force on the vibration.

[0023] According to some embodiments of the present document, IMU data containing multi-axis accelerometer data and multi-axis gyroscope data (collectively representing vibrational damping) is used to determine the corresponding force causing the vibrational damping. This approach recognizes the coupling between different axes of gyroscope and accelerometer data that respond differently to an applied force. Due to conservation of energy, an applied force does not simply result in a decrease in vibrational damping or amplitude along all IMU axes. Instead, some axes can even increase, as the applied force to dampen vibrational motion on one axis can result in transfer of vibrational energy to another axis. The disclosed approach takes multiple inputs from both accelerometers and gyroscopes as input, studying the frequency, amplitude, and relationships between different axes.

[0024] An applied force causing vibrational damping of a mobile device can be quantified based on a force model trained to correlate (1) IMU data measured by the mobile device indicative of vibrational damping of the mobile device with (2) forces applied to the mobile device and causing corresponding vibrational damping. The force model can include a machine learning algorithm, such as a multivariate linear regression algorithm. The correlation can depend on a configuration of the mobile device. Damping of vibrations can depend on one or more factors affecting how the mobile device absorbs and dissipates vibrational energy. Examples of these factors include vibrational intensity, load distribution within the mobile device including, for example, weight and distribution of components within the mobile device, materials used in the mobile device including, for example, density, elasticity, and internal friction, overall design including, for example, shape and structural integrity, features such as ribs, gussets, etc., interactions between components within the device, etc., or combinations thereof. Further, damping measured by the IMU of the mobile device can depend on one or more factors including, for example, a position of the IMU relative to one or more vibrational motors, sensor parameters including, for example, sensitivity, resolution, noise level, etc., or combinations thereof. Cross-device reliability studies disclosed herein show that the techniques disclosed herein provide similar force estimation performance across all tested smartphone models, indicating that the techniques are widely applicable and transferable across mobile devices. A force model can be determined for a type of mobile device, e.g., a manufacturer’s model, and used to calibrate mobile devices of the same type.

[0025] The calibration step can allow for standardized measurements across different types of mobile devices. The calibration can ensure that different mobile devices of the same type or different types are configured to consistently measure force. Thus, different mobile devices of the same type or different types can use the same blood pressure estimation model, even though the force model (also referred to as a force estimation algorithm) can be different. This is beneficial because developing a force model is an automated or at least partially automated process that can be completed in a short amount of time (e.g., one day) and does not require a large number of participant recruitment or clinical measurements. On the other hand, developing a blood pressure model can require many participants with different demographics to have clinical blood pressure measurements, which can take months or even longer.

[0026] Figure 1 An exemplary system demonstrating the disclosed technology implementing mobile device-based oscillometric blood pressure measurements is shown. The system 100 includes a mobile device 102, which can include at least one of a camera 104, a vibration motor 105, an IMU 106, a processor 108, a wireless transmitter 110, and a display 112. The processor 108 can control the operation of the mobile device 102 (e.g., cause the vibration motor 105 to vibrate, cause the camera 104 and / or the IMU 106 to acquire data), receive and process sensor data (e.g., image data acquired by the camera 104, IMU data acquired by the IMU 106), and run algorithms (e.g., a force model) on the sensor data and generate results (e.g., applied force causing vibration damping, a force plot, a PPG plot, blood pressure, a diagnosis based on determined blood pressure, etc.). As Figure 1 As shown in FIG. 1, the mobile device 102 can communicate with a user 114 or an external device or system (e.g., a cloud server 116). For example, the mobile device 102 can send a report of the system status to the cloud server 116 using, for example, the wireless transmitter 110. As another example, the mobile device 102 can communicate with the user 114 via the display 112. The display 112 can be a touchscreen configured as a graphical user interface, such that the mobile device 102 can present data or results to the user 114 via the display 112 and receive user input via the display 112. In some embodiments, the mobile device 102 can be a smartphone, a tablet computer, etc.

[0027] Figure 2 An exemplary block diagram showing various components of a mobile device according to some embodiments of the present document is shown. The mobile device 200 is Figure 1The example mobile device 102 shown in FIG. 1. In some embodiments, the mobile device 200 can be a smartphone, tablet, etc. In some embodiments, the mobile device 200 can be vibrated by the driving of a built-in vibration motor 203, measure the vibration damping caused by a user applying a force to the mobile device 200, capture PPG data that records changes in blood volume in a set of blood vessels caused by the applied force, and estimate blood pressure using a force model based on the PPG data and the applied force. The mobile device 200 includes one or more sensors 202 that can collect data, a processing unit 204 connected to the one or more sensors 202 and capable of executing a force model on the collected data, a wireless transceiver 206 connected to the processing unit 204, and a display 208 connected to the processing unit 204. The one or more sensors 202 can include one or more of a camera or an IMU. The IMU of the mobile device 200 can include an accelerometer and a gyroscope.

[0028] Figure 3 An example application for mobile device-based blood pressure measurement according to some embodiments of the disclosure is shown. For illustrative purposes only and not intending to be limiting, a smartphone application is shown. It should be understood that the mobile device-based blood pressure measurement disclosed herein can be implemented on other mobile devices including, for example, tablets.

[0029] Figure 3 (a) shows a simplified diagram of smartphone force damping. The smartphone includes a vibration motor, a camera, an IMU, and a display. The vibration motor drives oscillatory motion. A user applies a damping force by pressing a finger on the camera. The IMU of the smartphone records the motion (vibration), and the smartphone camera simultaneously records PPG. The base or other side of the fingernail is pressed on the camera of the smartphone so that the smartphone camera can measure real-time blood volume within the finger when the finger applies a force by pressing on the camera.

[0030] Figure 3 (b) shows the movement of the vibration motor along the Z-axis (perpendicular to the display of the smartphone) in (I), the applied finger force along the Z-axis in (II), and the linear Z-axis acceleration of the smartphone IMU plotted vertically in (III). As shown, as the applied force increases, the oscillatory acceleration or vibration amplitude of the Z-axis decreases. The raw PPG signal in (IV) is also plotted vertically to show how the increase in applied force affects the blood volume in (V).

[0031] The IMU of the smartphone can include an accelerometer and a gyroscope. Although only linear acceleration along the Z-axis is shown (in (III)), the IMU of the smartphone measures acceleration in multiple directions (e.g., three perpendicular directions, including the direction along the Z-axis and two directions in the plane perpendicular to the Z-axis), as well as gyroscope data in multiple directions (e.g., the same three directions as the accelerometer). The vibration of the smartphone and its damping can be monitored based on IMU data including multi-axis accelerometer data in conjunction with multi-axis gyroscope data. Due to the manner in which vibration energy dissipates, the linear acceleration of the IMU in various axes (indicative of the motion of the smartphone) can change in different patterns. For example, as shown in (II) and (III), the linear acceleration along the Z-axis decreases as the applied force increases; however, the linear acceleration along different directions can decrease at different rates, or even increase for certain periods of time while the damping force is applied. Similarly, the angular velocity of the IMU along different directions (indicative of the motion of the smartphone) can change in different patterns in response to the user-applied damping force. Thus, IMU data including multi-axis accelerometer data in conjunction with multi-axis gyroscope data can be used to determine the applied force. See additional description elsewhere in this document regarding this aspect.

[0032] Figure 3 (c) shows a graphical user interface (GUI) of the smartphone application with visual signifiers to guide the user in performing a blood pressure measurement using the application. The first visual signifier includes an image of a fingertip near the smartphone camera, which can provide the user with intuitive guidance on how to position the finger. The second visual signifier includes a plot of the applied force during the force-damping measurement overlaid in real-time with a force guide line. The force guide can assist the user in applying the correct amount of force during the measurement.

[0033] The GUI can also include a plot of the PPG signal acquired during data collection. The acquired PPG signal can provide useful information to the user or others (e.g., researchers) during data collection. The PPG signal can be plotted in real-time before and during the measurement, so the user and / or researchers can ensure data quality. For example, if the PPG signal is not within the desired range, the user or researchers can press a button before the measurement to calibrate the PPG signal to ensure that the PPG signal is not over- or under-saturated. At the end of the measurement, the complete force and PPG signals can be plotted within the application to provide an overview of the data. The GUI can also present the predicted results, including, for example, systolic blood pressure (sysBP), diastolic blood pressure (diaBP), heart rate (HR), etc. The GUI can further present the user’s identity (e.g., user ID), the number of attempts (or count) by the user. As shown, the smartphone is inverted so that the camera of the smartphone is close to the user to facilitate the user placing the finger on the camera to apply the force and acquire the PPG signal.

[0034] Additionally or alternatively, the application can provide guidance to the user on how to use the application and the mobile device to perform a blood pressure measurement. Such guidance information can be in the form of text, audio, video, images, etc., or combinations thereof. For example, the GUI can remind the user that a software update or calibration is needed, that the mobile device should be placed on a specific type of surface (e.g., a flat wooden surface, such as an office desk or table), that a surface calibration is needed (as described elsewhere in this document), that the user needs to follow a protocol. The protocol can be substantially the same as a standard blood pressure procedure, including that the user needs to sit straight, that the measurement device (mobile device) is at heart level, that the user's feet need to be flat on the floor, that the user should remain calm and breathe normally during the measurement, that for a mobile device-based measurement, the user needs to place the mobile device flat on the surface, screen facing up, front-facing camera closest to the user (so that the user can easily press the camera while simultaneously viewing the GUI of the application). The application can provide instructions and / or guidance on how to properly perform the measurement. For example, the instructions / guidance can include that the user should place the hand flat on the surface, with the index finger over the front-facing camera of the mobile device.

[0035] Figure 3 (d) showing a plot showing the correlation between the amplitude of the plotted blood volume oscillations and the applied finger pressure. For ease of explanation, the approximate locations of diastolic blood pressure (DBP), mean arterial pressure (MAP), and systolic blood pressure (SBP) are indicated.

[0036] Figure 4 A flowchart showing a process for mobile device-based blood pressure measurement according to some embodiments of the present document is shown. Process 400 can be implemented on a mobile device (e.g., mobile device 102, mobile device 200).

[0037] At block 410, process 400 includes causing a vibration motor of the mobile device to vibrate. The vibration motor can be a vibration motor that is already built into the mobile device. For example, the vibration can be driven by a vibration motor that is used to provide haptic feedback for messages, phone calls, and screen interactions, etc. The vibration motor can be caused to vibrate at a particular level of the vibration motor (e.g., a maximum level).

[0038] In some embodiments, the mobile device can be placed on a surface of a particular type to perform the oscillometric blood pressure measurement, taking into account that the vibration damping involves characteristics of the particular type of surface. For example, process 400 can include providing information to guide or remind that the mobile device is to be placed on such a type of surface (e.g., a flat wooden surface, such as a desk, office desk, etc.) as part of the preparation for the blood pressure measurement. Process 400 can include providing additional information, including, for example, how the user should sit, how to apply the force, etc., or combinations thereof, as part of the preparation.

[0039] In some embodiments, process 400 can include performing surface calibration by measuring a characteristic of a surface on which the mobile device is placed. Example characteristics of a surface include flatness, levelness, material, etc., or combinations thereof. In some embodiments, the mobile device can measure whether a surface is level using built-in sensors including an IMU (including an accelerometer and / or a gyroscope) by detecting the direction of gravity relative to the orientation of the mobile device and / or the tilt angle and tilt direction of the surface. In some embodiments, the mobile device can identify the material of the surface. By way of example only, process 400 can include causing a mobile phone placed on a surface to vibrate in a vibration pattern of a known vibration profile under the drive of a built-in vibration motor, measuring the vibration of the mobile device, and determining the material of the surface by comparing the measured vibration profile to the known vibration profile. As another example, process 400 can include causing a mobile phone placed on a surface to vibrate in different vibration patterns of a known vibration profile under the drive of a built-in vibration motor, measuring the vibration profile of the mobile device, and determining the material of the surface by comparing the changes in the measured vibration profile to the changes in the known vibration profile. Additionally or alternatively, process 400 can include identifying a characteristic of a surface based on one or more images of the surface. Process 400 can include performing surface calibration based on the characteristic of the surface when determining the applied force, as described below.

[0040] At block 420, process 400 includes acquiring inertial measurement unit (IMU) data using an IMU of the mobile device, the IMU data being indicative of vibration dampening caused by a body part of a user applying a force to a camera of the mobile device. The body part can be a tip portion of a finger of the user (e.g., the index finger of the user). In some embodiments, process 400 can include providing visual guidance to be presented on a display of the mobile device. The visual guidance (also referred to as visual cues) can include markings to guide the user to place a surface of the body part onto the camera.

[0041] The IMU of the mobile device can include a multi-axis accelerometer and a multi-axis gyroscope. The vibration motor can be a vibration motor that has been built-in in the mobile device for other purposes, including, for example, measuring linear acceleration and angular velocity, enabling these mobile devices to do position inference, which are widely used for rotating screens, playing games, counting steps, gesture recognition, etc.

[0042] The IMU data can include multi-axis accelerometer data and multi-axis gyroscope data. For example, the IMU data can include three-axis accelerometer data and three-axis gyroscope data. The three-axis accelerometer data can include linear acceleration of the IMU in each axis (indicative of the smart phone motion). Depending on the degree of energy dissipation during the vibration damping, the linear acceleration of the IMU can change in different patterns. For example, the linear acceleration along different directions can decrease at different rates, or even follow opposite trends (e.g., linear acceleration in one direction increases while linear acceleration in one direction decreases) for certain time periods of the applied damping force. Similarly, the angular velocity of the IMU along different directions (indicative of the smart phone motion) can change in different patterns in response to the user-applied damping force. Thus, the IMU data including multi-axis accelerometer data and multi-axis gyroscope data can be used in combination to provide a comprehensive representation of the vibration damping, and thus improve the accuracy of the determined applied force. As used herein, "damping" or "vibration damping" refers to a recorded alteration in the device motion due to the applied force.

[0043] Each axis of the multi-axis accelerometer data and multi-axis gyroscope data can include a time series signal. The signals of the axes of the IMU data can contain high frequency components from the vibration motor and low frequency components corresponding to noise. In some embodiments, the raw IMU data can be sampled from the mobile device at the maximum rate without device filtering or post-processing. In some embodiments, the raw IMU data can undergo a series of processing. For example, the raw IMU data can be processed using one or more of the following techniques: low pass filtering, high pass filtering, band pass filtering, savgol filtering, standard deviation, or empirical mode decomposition. The filtering can also involve multiple steps, such as the case of a filter bank or multiple stages with different filter types. From the filtered or raw IMU data, additional features can be obtained, such as the signal power of the axis combinations. The one or more features can be assessed using various metrics, including, for example, principal component analysis, recursive feature elimination, LASSO regression, and correlation.

[0044] At block 430, the process 400 includes determining the applied force based on the force model and the IMU data.

[0045] In some embodiments, the process 400 can include decomposing the IMU data based on an empirical mode decomposition (EMD) technique. According to the EMD technique, the process 400 can include, for each axis of the multi-axis accelerometer data and multi-axis gyroscope data, decomposing the signal of the axis into a plurality of intrinsic mode functions (IMFs) representing different frequency components of the signal, such that the applied force can be determined based on at least a portion of the IMFs corresponding to the respective axes of the multi-axis accelerometer data and multi-axis gyroscope data.

[0046] Since the high frequency components of the signals of the individual axes of the vibrating motor are the strongest components of the corresponding signals, the first few IMFs of the individual axes of the IMU data primarily correspond to the vibrating motor signals. The upper envelope of the first IMF of each axis of the multi-axis accelerometer data and the multi-axis gyroscope data can be used as a feature for identifying the applied force. In some embodiments, the upper envelope of the first IMF on each IMU axis acts as a representative feature, and the first IMF can collectively be most significantly correlated with the force data and thus used as an input to a force model to determine the applied force, while other features can be excluded. Thus, for example only, the process 400 can include, for each axis of the multi-axis accelerometer data and the multi-axis gyroscope data, identifying, from a plurality of IMFs of the signals of the axis, a first IMF corresponding to the highest frequency; and identifying an upper envelope of the first IMF of the axis; and inputting the upper envelopes of the first IMF corresponding to all axes of the multi-axis accelerometer data and the multi-axis gyroscope data, respectively, into a force model.

[0047] The EMD techniques described herein for processing the IMU data are for illustrative purposes only and are not intended to be limiting. As described elsewhere in this document, the IMU data can be processed by at least one of filtering, filter banks, averaging, standard deviation, wavelet decomposition, spectral analysis, or EMD, etc. before being input into a force model. In some embodiments, different techniques can be used to process the signals of at least two different axes of the IMU data, including the multi-axis accelerometer data and the multi-axis gyroscope data. In some embodiments, the raw IMU data can be directly input into a force model.

[0048] The force model can include a machine learning model trained to relate (1) IMU data measured by a mobile device indicative of a vibration damping of the mobile device to (2) a force applied to the mobile device and causing the corresponding vibration damping. The force model can be a data driven model, including parametric modeling, linear regression models, ensemble learning models (random forest, adaboost, etc.), support vector machine models, neural networks (e.g., transformer models, convolutional neural networks (CNNs), etc.), etc., or variations thereof, or combinations thereof. For example only, the force model can include a multivariate linear regression model providing values of the applied force on a continuous scale.

[0049] The force model can be trained using training data that includes applied forces measured using a force sensor (e.g., force sensitive resistor (FSR)) and measured dampening resulting from the applied forces. Example test cases were performed across various smartphone manufacturers, physical shapes, costs, and embedded components, which involved five participants and three different types of smartphones, including Google Pixel 4, Samsung Galaxy A53, and Motorola Moto G Power. In the example test cases, each of the five participants in a group applied pressure with an index finger on a 0.3mm thick force sensor positioned on top of the front-facing camera of the smartphone. The participants applied the force with the index finger in the same vertical position as a blood pressure measurement to cause a hydrostatic force. The force sensor (a calibrated linear force sensor (SingleTact 15mm 4.5N)) was placed above the front-facing camera of the smartphone. The participants pressed their fingers on the force sensor of the smartphone (and the camera of the smartphone) to measure the force applied by the finger. In the example case, the smartphone was placed on a flat wooden desktop during the measurement. Real-time readings of the force sensor were displayed on the screen in front of the user through force guidance during the measurement. The user was instructed to use the real-time feedback to follow the force guidance to apply a series of pressures. Similar to a cuff-based blood pressure measurement, the force guidance instructed the user to continuously increase the applied force. The study included repeating a phase of applying the series of forces forty-two times three times. Each participant performed the three repetitions on each of the three different types of smartphones to obtain ten minutes of raw applied force data for each phone (two minutes per participant). The acquired IMU data was processed based on EMD in substantially the same manner as described above to provide vibration dampening data. The force measured using the force sensor and the corresponding vibration dampening data were used to train and validate the force model. Table 1 and Figure 5 FIG. 1 show the mean absolute error, correlation coefficient, and bias for each smartphone. In Figure 5 FIG. 1, solid dots represent data corresponding to the Motorola Moto G Power, crosses represent data corresponding to the Samsung Galaxy A53, and stars represent data corresponding to the Google Pixel 4. The mean correlation coefficient for all phones was 0.86. The Google Pixel 4 smartphone used in the blood pressure validation study had the lowest mean absolute error (MAE).

[0050] Table 1.

[0051] Smartphone Model Dimensions (mm) Cost ($) MAE (N) STD Bias R Google Pixel 4 147.1 x 68.8 x 8.2 799 0.96 1.34 -0.37 0.84 Samsung Galaxy A53 159.6 x 74.8 x 8.1 399 0.85 1.69 -0.12 0.84 Moto G Power 2022 159.9 x 75.8 x 9.6 199 0.70 1.46 -0.11 0.90

[0052] As discussed elsewhere in this document, the vibration damping behavior of a mobile device can depend on its configuration. Results of example test cases indicate that relevant differences in force measurements for different types of mobile devices (e.g., types can correspond to manufacturer models) can be calibrated using different force models. When a new type of mobile device (e.g., a new model by a manufacturer) is released, a force model can be trained, and the obtained force model can be used to calibrate mobile devices of the same type. A mobile device type can correspond to a model and / or a manufacturer of a mobile device, and be associated with multiple factors, including, for example, a configuration of a vibration motor of the mobile device, a configuration of an IMU, a configuration of a camera, etc., or combinations thereof.

[0053] A model library can be constructed by aggregating force models corresponding to different types of mobile devices. In some embodiments, process 400 can include obtaining a force model from such a model library based on a type of a mobile device, as preparation for a blood pressure measurement procedure.

[0054] Process 400 can determine the applied force substantially in real-time. In some embodiments, during a measurement, process 400 can include providing visual guidance to be presented on a display of the mobile device. The visual guidance can include a force plot showing the applied force substantially in real-time. In some embodiments, the visual guidance can further include a force guide line superimposed on the applied force. The user can be instructed to follow the force guide to apply a series of forces using the real-time feedback. By way of example only, similar to the way a cuff-based blood pressure measurement is performed, the force guide can instruct the user to continuously increase the applied force within a range.

[0055] In some embodiments, process 400 can include measuring a surface area of a body part, where the force is applied by pressing a surface of the body part on a camera. By way of example only, when the body part is pressed on the camera to apply a damping force, process 400 can cause the camera to take a picture of the surface of the body part, and determine the area based on the image. As another example, the mobile device includes a display including a touch screen configured to sense user contact, and process 400 can determine the area by identifying an area of contact of the body part with the display. Process 400 can include determining a pressure applied based on the applied force and the area.

[0056] At block 440, the process 400 includes capturing photoplethysmography (PPG) data of a set of blood vessels in the body part while the force is applied using a camera of the mobile device. The body part of the user can include a tip portion of a finger, and the set of blood vessels can include at least one blood vessel in a palmar arch branch of a digital artery in the tip portion of the finger. The PPG data can be related to blood flow within the set of blood vessels that varies with the magnitude of the applied force. The PPG data can record intensity in a red channel of the camera. The blood volume flowing through the artery near the nail bed varies with the magnitude of the applied force, and the reflectance properties of the tissue in the region change. Thus, the camera of the mobile device can record the change in blood volume by measuring the change in reflectance properties.

[0057] By way of example only, the display (or screen) of the mobile device can be set to a pure white background with maximum brightness. The user can be instructed to place a body part (e.g., index finger) over the camera (e.g., with the aid of a visual symbol). The bright white screen illuminates the finger, and the camera records the change in pixel intensity in the red channel, which is representative of the change in blood volume. To account for skin color and light variations, the mobile device can adjust the ISO (sensitivity of the camera sensor to light) by calibrating with the user's finger over the camera. By way of example, the red channel intensity is between 70 and 200. This calibration can prevent PPG measurements from being undersaturated or oversaturated.

[0058] At block 450, the process 400 includes determining the user's blood pressure using an oscillogram technique based on the PPG data and the applied force. The PPG measurements can be aligned with the applied force measurements so that an applied force value can be assigned to each PPG peak. The prominence of each PPG peak can be determined by finding the peak to trough difference. During a blood pressure measurement, the camera of the mobile device records the PPG signal while the user applies force with a body part (e.g., index finger) to the camera. The applied force causes the PPG signal to change, resulting in an oscillogram shape. When the applied force is greater than the diastolic pressure, the blood volume during the systolic pulse period increases, and the blood volume during the diastolic period decreases. The total volume of blood flowing through the artery is roughly the same, but the pressure on the artery causes more blood to flow during the systolic pulse period and less blood to flow during the diastolic pulse period. This increases the prominence of the PPG signal. When the applied force exceeds the mean arterial pressure, the blood volume starts to be restricted even during the systolic period of the pulse. Thus, the systolic PPG prominence starts to decrease, and the blood volume during the diastolic period approaches zero to remain roughly constant. The PPG prominence continues to decrease even after the applied force exceeds the systolic pressure. As the pressure increases, the signal will decrease to zero (or noise) as the blood flowing through the artery approaches zero during both the systolic and diastolic periods. The pulse prominence plot drawn by the range of applied force (from less than diastolic pressure to higher than systolic pressure) for each PPG peak results in a Gaussian or Skew-Gaussian like shape, which can be referred to as an oscillogram. The blood pressure can be determined based on the oscillogram and a blood pressure model.

[0059] As described elsewhere in this disclosure, the blood pressure model can be substantially independent of the type of mobile device. Thus, no calibration of the blood pressure model based on the type of mobile device is needed.

[0060] The blood pressure model can be a data driven model, including parametric modeling, linear regression models, ensemble learning models (random forest, adaboost, etc.), support vector machine models, neural networks (e.g., transformer models, CNNs), etc., or variations thereof, or combinations thereof.

[0061] Most FDA-approved blood pressure cuffs utilize the oscillometric method, which utilizes the shape of the oscillometric graph to assess blood pressure. Thus, a blood pressure model can be trained to assess blood pressure based on the oscillometer. To train, the oscillometer can be processed to extract characteristic features including, for example, the maximum value of the oscillometric graph, the applied force at the maximum value (a variable parameter of the maximum value of the oscillometric graph), the extreme gradient of the oscillometric graph (e.g., the maximum gradient, the minimum gradient), a variable parameter of the extreme gradient (a variable parameter of the maximum gradient, a variable parameter of the minimum gradient), the prominence at the extreme gradient (e.g., the prominence at the maximum gradient, the prominence at the minimum gradient), and the like, or combinations thereof. In Figure 6 (a) and Figure 6 These exemplary characteristic features are observed in

[0062] Figure 6 (a) shows exemplary PPG and applied force data measured by a mobile device and the corresponding generated oscillometric graph according to some embodiments of the present document. Specifically, Figure 6 (a) shows filtered PPG data and applied force data captured using a smartphone, where the amplitude of the filtered PPG signal is computed and the force is converted to pressure via a fixed estimate of the fingertip surface area. The filtered amplitude of the PPG signal is plotted against the applied pressure, called the oscillometric graph. The shape of the oscillometric graph resembles a skewed Gaussian, where the peak indicates the mean arterial pressure.

[0063] Figure 6 (b) shows average oscillometric graphs corresponding to various blood pressure ranges according to some embodiments of the present document. The graph of the average oscillometric graph plotted by blood pressure range shows that the oscillometric graph shape indicates blood pressure. The average oscillometric graph shape is plotted as a solid line. The average skewed Gaussian fit is plotted as a dashed line. For higher blood pressures, the oscillometric graph peak shifts to the right because, for example, the mean arterial pressure is larger.

[0064] In the example case of training a blood pressure model, the characteristic features of the sphygmography that include the maximum of the sphygmogram, the applied force at the maximum (a variable parameter of the maximum of the sphygmogram), the extreme gradient of the sphygmogram (e.g., maximum gradient, minimum gradient), a variable parameter of the extreme gradient (a variable parameter of the maximum gradient, a variable parameter of the minimum gradient), the prominence at the extreme gradient of each of the training sphygmograms (e.g., prominence at the maximum gradient, prominence at the minimum gradient) are the only inputs to the blood pressure model. In the example case, motion-related attributes, including heart rate and pulse shape, are not included in the features provided to the model; indirect measurement of heart rate or pulse is avoided by interpolating all sphygmograms to have the same number and same force values of points. A Least Absolute Shrinkage and Selection Operator (LASSO) regression model is trained to predict systolic and diastolic measurements of blood pressure. The model is trained and tested using hold one participant out validation. In this training regimen, the model is trained using all data except for one or more blood pressure measurements of a single participant. The model is then used to predict blood pressure of the holdout participant. This process is repeated for each participant.

[0065] Based on the trained blood pressure model, an N = 30 validation study was performed. Table 2 shows participant demographics.

[0066] Table 2

[0067]

[0068]

[0069] In Table 2 of participant demographics, “Skin” refers to Fitzpatrick Skin Type, “Phone” refers to mean smartphone estimates (rounded to the nearest integer), and “Cuff” refers to cuff measurements for baseline ground truth. The phone and cuff mean measurements shown are only from natural (non-exercise induced) measurements. Each of the thirty participants participated in multiple measurements.

[0070] In the blood pressure validation study, the smartphone device was compared to an FDA-approved cuff-based device (Omron BP7350). Participants first used the smartphone to record a trial measurement to learn how to use the device and practice applying force. The trial measurement was not used in the analysis. The cuff-based device was used to measure the participant's blood pressure following a standard protocol. Immediately after the cuff-based measurement, the participant used the smartphone to perform three secondary blood pressure measurements. After the three blood pressure measurements were completed on the smartphone, the blood pressure was again measured using the cuff-based device. The complete measurement sequence was as follows: cuff, smartphone, smartphone, smartphone, cuff. If the systolic BP readings from the two cuff measurements differed by less than 3 mmHg, the first cuff measurement was used as the label for all three smartphone measurements in the analysis. If the two blood pressure readings differed by more than 3 mmHg, the average of the two readings was taken.

[0071] Participants were also asked to optionally complete a second phase of data collection involving exercise. The purpose of this exercise phase was to obtain high blood pressure data. After consent, participants were asked to perform a one-minute wallsit. During the wallsit, participants measured their blood pressure using both the blood pressure cuff device and the smartphone device. Both blood pressure measurements were taken approximately ten seconds after the start of the wallsit.

[0072] Out of the thirty participants, N=6 participants were completely excluded. Of these six excluded participants, one participant was unable to perform any valid force measurements; one participant had low prominence in all measurements; and one participant had a poor Gaussian fit in all measurements. The remaining three excluded participants were excluded on the basis of multiple criteria. The exclusion criteria for participant measurements were as follows:

[0073] • Cuff blood pressure over 160 mmHg: If the reference measurement was over 160 mmHg, the measurement was excluded based on previous research, which showed that fingertip and arm blood pressure values differ at exercise-induced very high BP values.

[0074] • Inconsistent / undetected pulse or excessive noise: The Fast Fourier Transform (FFT) of the PPG signal between zero and 2 Hz (a frequency associated with human pulse) must have a narrow, identifiable peak indicative of pulse rate.

[0075] • Insufficient force applied: The force measurement must have a correlation coefficient greater than 0.88. The first 0.5 seconds of the force signal must have a correlation coefficient less than 0.21 N. Additionally, the minimum force applied must be less than 6 N, and the maximum force value must be greater than 7 N.

[0076] • Poor skewness Gaussian fit: the best fit normalized skewness Gaussian mean absolute error must be less than 0.18.

[0077] • PPG signal saturation: if the PPG signal exceeds 254 pixel intensity units, then the measurement is discarded. This indicates that the finger has been lifted from the smartphone camera during the measurement.

[0078] • PPG prominence invariant: the normalized PPG prominence standard deviation must be greater than 0.17.

[0079] The validity study showed that the MAE for systolic and diastolic BP for mobile device-based oscillometric blood pressure measurements were 9.9 mmHg and 8.9 mmHg, respectively, compared to FDA-approved blood pressure cuffs. The Pearson correlation coefficients for systolic and diastolic measurements were 0.67 and 0.23, respectively. The diastolic blood pressure range measured by blood pressure cuffs was 62 to 107 mmHg and the systolic blood pressure range was 89 to 147 mmHg. See Figure 6 (a) and Figure 6 (b) show the regression and Bland Altman plots for all measurements of systolic (systolic BP) and diastolic (diastolic BP) blood pressure for N = 24 participants, respectively. Because each participant was measured multiple times, a total of 73 measurements were made. Systolic and diastolic BP values were calculated using the same model.

[0080] For screening purposes based on the technology disclosed herein, blood pressure (BP) values can be classified based on AMA / ACC criteria. Figure 7 The results shown in Figure 6 (a) and Figure 6 (b) are shown in a confusion matrix and adjusted to ensure that elevated blood pressure and hypertensive individuals are correctly labeled. Specifically, for the classification prediction, the sensitivity and specificity are adjusted from the raw prediction. The classification prediction threshold is determined solely based on the AMA / ACC established criteria by systolic BP value. Receiver operating characteristic (ROC) analysis shows that the area under the curve (AUC) for detecting hypertensive individuals is 0.88, characterized by systolic pressure greater than 130 mmHg. The ROC curve informs the sensitivity and specificity of the prediction. As Figure 7 The confusion matrix shown in

[0081] These results demonstrate that blood pressure screening using only a downloadable smartphone software application has great potential. The technology can be used as a low- threshold blood pressure screening measurement approach suitable for widespread use.

[0082] In addition, the mobile device-based oscillometric blood pressure technology disclosed herein can provide a calibration mechanism. For example, methods of monitoring relative changes in blood pressure using wearable devices such as smartwatches or smartglasses are becoming more common; these devices typically rely on pulse transit time (PTT), pulse wave analysis (PWA), or pulse arrival time (PAT) methods. One key limitation of this type of blood pressure measurement is the need for calibration using absolute blood pressure measurements, typically via a blood pressure cuff device on a daily or weekly basis. The absolute blood pressure measurements on a mobile device disclosed herein can provide an efficient, low-cost method to calibrate wearable or other continuous BP relative measurements.

[0083] Furthermore, the mobile device-based oscillometric blood pressure technology disclosed herein can be combined with other cardiovascular measurements to analyze heart health. For example, the systems disclosed herein can be combined with another optical system using a camera to perform PWA, heart rate variability (HRV) measurements, or pulse measurements to provide additional information on blood pressure.

[0084] The technical improvements provided in the embodiments disclosed herein include methods for performing absolute blood pressure measurements that can be applied to different types (e.g., models) of mobile devices without requiring hardware modifications or accessories, and in fact can be implemented through downloadable applications. The proliferation of sensor-based measurements on a wide variety of mobile devices is a well-recognized problem that severely limits the impact of work in the mobile health field. Cross-device compatibility research results demonstrate that the force-damping technology disclosed herein allows blood pressure measurements to be run on a range of mobile devices. Data processing and modeling show that many different types of mobile devices are capable of performing oscillometric blood pressure measurements after calibration based on the mobile device type with respect to force measurements. Each popular type of mobile device has tens of millions of users, and a single factory calibration can enable millions of mobile devices to perform oscillometric blood pressure measurements.

[0085] Examples

[0086] The following examples illustrate several embodiments in accordance with the present technology. Other exemplary embodiments of the present technology can be presented before or after the following examples.

[0087] 1. A mobile device comprising: a vibration motor, an inertial motion unit (IMU), a camera, a processor, and a non-transitory computer-readable memory storing instructions that, when executed by the processor, cause the processor to at least: cause the vibration motor to vibrate; acquire, by an inertial measurement unit of the mobile device, IMU data indicative of vibration damping caused by a force exerted by a body part of a user on the camera; determine the exerted force based on a force model and the IMU data; capture, by the camera of the mobile device, photoplethysmography (PPG) data of a set of blood vessels in the body part while the force is being exerted; and determine a blood pressure of the user using an oscillometric technique based on the PPG data and the exerted force.

[0088] 2. The mobile device of any one or more of the solutions described herein, wherein the force model comprises a machine learning algorithm trained to correlate (1) IMU data measured by the mobile device indicative of vibration damping of the mobile device with (2) a force exerted to the mobile device and causing the vibration damping.

[0089] 3. The mobile device of any one or more of the solutions described herein, wherein the IMU comprises an accelerometer and a gyroscope, and the IMU data comprises multi-axis accelerometer data and multi-axis gyroscope data.

[0090] 4. A method for measuring blood pressure using a mobile device, the method comprising: causing a vibration motor of the mobile device to vibrate; acquiring, using an inertial measurement unit (IMU) of the mobile device, IMU data indicative of vibration damping caused by a force exerted by a body part of a user on a camera of the mobile device; determining the exerted force based on a force model and the IMU data; capturing, using the camera of the mobile device, photoplethysmography (PPG) data of a set of blood vessels in the body part while the force is being exerted; and determining a blood pressure of the user using an oscillometric technique based on the PPG data and the exerted force.

[0091] 5. The method of any one or more of the solutions described herein, wherein the IMU data comprises multi-axis accelerometer data and multi-axis gyroscope data.

[0092] 6. The method of any one or more of the solutions described herein, wherein determining the exerted force from the IMU data comprises processing the IMU data based on at least one of filtering, a filter bank, averaging, standard deviation, wavelet decomposition, spectral analysis, or empirical mode decomposition (EMD).

[0093] 7. The method of any one or more of the solutions described herein, wherein each axis of the multi-axis accelerometer data and the multi-axis gyroscope data includes a signal, determining the applied force from the IMU data includes, for each axis of the multi-axis accelerometer data and the multi-axis gyroscope data, decomposing the signal of the axis into a plurality of intrinsic mode functions (IMFs) representing different frequency components of the signal, and the applied force is determined based on at least a portion of the IMFs corresponding to the respective axes of the multi-axis accelerometer data and the multi-axis gyroscope data.

[0094] 8. The method of any one or more of the solutions described herein, wherein determining the applied force from the IMU data further includes, for each axis of the multi-axis accelerometer data and the multi-axis gyroscope data, identifying a first IMF from the plurality of IMFs of the signal of the axis corresponding to a highest frequency, and identifying an upper envelope of the first IMF of the axis, and inputting the upper envelopes of the first IMFs corresponding to all axes of the multi-axis accelerometer data and the multi-axis gyroscope data, respectively, into the force model.

[0095] 9. The method of any one or more of the solutions described herein, wherein the PPG data relates to blood flow within the set of blood vessels, the blood flow varying with an amplitude of the applied force.

[0096] 10. The method of any one or more of the solutions described herein, wherein the PPG data records intensity in a red channel of the camera.

[0097] 11. The method of any one or more of the solutions described herein, wherein the mobile device is placed on a support surface when the user applies the force, the method further including performing a surface calibration by characterizing a material of the support surface, and determining the applied force further based on the surface calibration.

[0098] 12. The method of any one or more of the solutions described herein, further including providing visual guidance to be presented on a display of the mobile device, wherein the visual guidance includes a force plot showing the applied force in substantially real time.

[0099] 13. The method of any one or more of the solutions described herein, wherein the visual guidance further includes a force guide line superimposed with the applied force.

[0100] 14. The method of any one or more of the solutions described herein, further including providing visual guidance to be presented on a display of the mobile device, wherein the visual guidance includes a marker to guide the user to place a surface of the body part onto the camera.

[0101] 15. The method of any one or more of the solutions described herein, further comprising: measuring an area of a surface of the body part, wherein the force is applied by pressing the surface of the body part against the camera; and determining the applied pressure based on the applied force and the area, and determining the blood pressure includes: generating an oscillogram based on the PPG data and the applied pressure; and deriving the blood pressure from the oscillogram.

[0102] 16. The method of any one or more of the solutions described herein, further comprising: causing the oscillogram to be presented on a display of the mobile device.

[0103] 17. The method of any one or more of the solutions described herein, wherein: the body part of the user is a tip portion of a finger, and the set of blood vessels includes at least one blood vessel in a palmar arch branch of a digital artery in the tip portion of the finger.

[0104] 18. The method of any one or more of the solutions described herein, wherein: determining the blood pressure of the user using the oscillography technique includes: generating an oscillogram based on the PPG data and the applied force; extracting characteristic features from the oscillogram, the characteristic features including at least one of: a peak value of the oscillogram, an extreme value of a gradient of the oscillogram corresponding to the peak value, an extreme gradient of the oscillogram, or a variable parameter of the extreme gradient; and generating the blood pressure based on the characteristic features and a blood pressure model.

[0105] 19. The method of any one or more of the solutions described herein, further comprising: obtaining the force model from a model library based on a type of the mobile device, wherein the type of the mobile device is associated with at least one of: a configuration of the IMU, a configuration of the vibration motor, or a configuration of the camera.

[0106] 20. One or more non-transitory computer-readable media storing instructions that, when executed by one or more processors of a mobile device, cause the mobile device to perform any one or more of the solutions described herein.

[0107] Embodiments of the subject matter and the operations described in this patent document can be implemented in a variety of systems, digital electronic circuitry, or in computer software, firmware, or hardware, including the structures disclosed in this specification and their structural equivalents, or in combinations of one or more of them. Embodiments of the subject matter described in this specification can be implemented as one or more computer program products, i.e., one or more modules of computer program instructions encoded on a tangible and non-transitory computer readable medium for execution by, or to control the operation of, data processing apparatus. The computer readable medium can be a machine-readable storage device, a machine-readable storage substrate, a memory device, a composition of matter effecting a machine-readable propagated signal, or a combination of one or more of them. The term“data processing apparatus” encompasses all apparatus, devices, and machines for processing data, including by way of example a programmable processor, a computer, or multiple processors or computers. The apparatus can include, in addition to hardware, code that creates an execution environment for the computer program in question, e.g., code that constitutes processor firmware, a protocol stack, a database management system, an operating system, or a combination of one or more of them.

[0108] A computer program (also known as a program, software, software application, script, or code) can be written in any form of programming language, including compiled or interpreted languages, and it can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment. A computer program does not necessarily correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub programs, or portions of code). A computer program can be deployed to be executed on one computer or on multiple computers that are located at one site or distributed across multiple sites and are interconnected by a communication network.

[0109] The processes and logic flows described in this specification can be performed by one or more programmable processors executing one or more computer programs to perform functions by operating on input data and generating output. The processes and logic flows can also be performed by special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application specific integrated circuit), and that apparatuses can also be implemented as such special purpose logic circuitry.

[0110] Processors suitable for the execution of a computer program include, by way of example, both general and special purpose microprocessors, and any one or more processors of any kind of digital computer. Generally, a processor will receive instructions and data from a read-only memory or a random access memory or both. The essential elements of a computer are a processor for executing instructions and one or more memory devices for storing instructions and data. Generally, a computer will also include, or be operatively coupled to receive data from or transfer data to, or in either or both modes, one or more mass storage devices for storing data, e.g., magnetic, magneto-optical disks, or optical disks. However, a computer need not have such devices. Computer readable media suitable for storing computer program instructions and data include all forms of non-volatile memory, media and memory devices, including by way of example semiconductor memory devices, e.g., EPROM, EEPROM, and flash memory devices. The processor and the memory can be supplemented by, or incorporated in, special purpose logic circuitry.

[0111] The specification and drawings are, of course, to be regarded in an illustrative rather than a restrictive sense, in that illustrative examples are described herein in order to enable a person of ordinary skill in the art to make and use the application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. Further, the use of "or" is intended to include "and / or", unless the context clearly indicates otherwise.

[0112] While this patent document contains many specific details, these should not be construed as limitations on the scope of any invention, but rather as descriptions of features that can be specific to particular embodiments of the invention. Certain features described herein in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features described herein in the context of a single embodiment can also be implemented separately or in any suitable subcombination. Moreover, although features can be described above as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination can in some cases be excised from the combination and the claimed combination can be directed to a subcombination or variation of a subcombination.

[0113] Similarly, while operations are depicted in the drawings in a particular order, this should not be understood as requiring such order, nor that all illustrated operations be performed, to achieve desirable results. In addition, the separation of various system components in the embodiments described herein should not be understood as requiring such separation in all embodiments.

[0114] The various embodiments described herein are described in the general context of methods or processes, which can be implemented in one embodiment by a computer program product, embodied in a computer-readable medium, including computer-executable instructions, such as program code, executed by computers in networked environments. A computer-readable medium can include removable and non-removable storage devices including, but not limited to, Read Only Memory (ROM), Random Access Memory (RAM), compact discs (CDs), digital versatile discs (DVDs), Blu-rays, and others. Accordingly, computer-readable media described in this application includes non-transitory storage media. Generally, program modules can include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. Computer-executable instructions, associated data structures, and program modules represent examples of program code for executing steps of the methods disclosed herein. The particular sequence of such executable instructions or associated data structures represents examples of corresponding acts for implementing the functions described in such steps or processes.

[0115] For example, one aspect of the disclosed embodiments relates to a computer program product tangibly embodied on a non-transitory computer readable medium. The computer program product includes program code for carrying out any one and / or all of the operations of the disclosed embodiments.

[0116] Only a few implementations and examples are described and other implementations, enhancements and variations can be made based on what is described and illustrated in this patent document.

Claims

1. A method for measuring blood pressure using a mobile device, the method comprising: The vibration motor of the mobile device is made to vibrate; The mobile device uses an inertial measurement unit (IMU) to acquire IMU data indicating vibration damping caused by forces exerted by the user's body parts on the camera of the mobile device. The applied force is determined based on the force model and the IMU data; Using the camera of the mobile device, photoplethysmography (PPG) data of a group of blood vessels in the body part is captured while the force is applied; and The user's blood pressure is determined using oscillometric techniques based on the PPG data and the applied force.

2. The method according to claim 1, wherein the IMU data includes multi-axis accelerometer data and multi-axis gyroscope data.

3. The method of claim 2, wherein determining the applied force based on the IMU data comprises processing the IMU data based on at least one of filtering, filter bank, averaging, standard deviation, wavelet decomposition, spectral analysis, or empirical mode decomposition (EMD).

4. The method according to claim 3, wherein: Each axis of the multi-axis accelerometer data and the multi-axis gyroscope data includes a signal. Based on the IMU data, the applied force is determined to include each axis of the multi-axis accelerometer data and the multi-axis gyroscope data, and the signal of that axis is decomposed into multiple intrinsic mode functions (IMFs) representing different frequency components of the signal. The applied force is determined based on at least a portion of the IMF corresponding to each axis of the multi-axis accelerometer data and the multi-axis gyroscope data.

5. The method of claim 4, wherein determining the applied force based on the IMU data further comprises: For each axis of the multi-axis accelerometer data and the multi-axis gyroscope data Identify the first IMF corresponding to the highest frequency from the plurality of IMFs of the signal of the axis; as well as Identify the upper envelope of the first IMF of the axis; as well as The upper envelope of the first IMF, which corresponds to all axes of the multi-axis accelerometer data and the multi-axis gyroscope data, is input into the force model.

6. The method of claim 1, wherein the PPG data relates to blood flow in the set of blood vessels, the blood flow varying with the magnitude of the applied force.

7. The method of claim 1, wherein the PPG data records the intensity in the red channel of the camera.

8. The method of claim 1, wherein the moving device is placed on the support surface when the user applies the force, the method further comprising: Surface calibration is performed by characterizing the material of the supporting surface; as well as The applied force is further determined based on the surface calibration.

9. The method of claim 1, further comprising: Provide visual guidance to be presented on the display of the mobile device, wherein the visual guidance includes a force diagram showing the applied force substantially in real time.

10. The method of claim 9, wherein the visual guidance further comprises a force guide line superimposed on the applied force.

11. The method of claim 1, further comprising: Provide visual guidance to be displayed on the mobile device, wherein the visual guidance includes markings to guide the user to place the surface of the body part onto the camera.

12. The method of claim 1, further comprising: The area of ​​the surface of the body part is measured, wherein the force is applied by pressing the surface of the body part against the camera; as well as The applied pressure is determined based on the applied force and the area, wherein determining the blood pressure includes: An oscilloscope pattern is generated based on the PPG data and the applied pressure; and The blood pressure is derived from the oscillogram.

13. The method of claim 12, further comprising: The oscilloscope graph is displayed on the mobile device's screen.

14. The method of claim 1, wherein determining the user's blood pressure using the oscilloscope technology comprises: An oscilloscope pattern is generated based on the PPG data and the applied force; Characteristic features are extracted from the oscilloscope, the characteristic features including at least one of the following: a peak value of the oscilloscope, an applied force corresponding to the peak value, an extreme value of the gradient of the oscilloscope, a prominence at the extreme gradient of the oscilloscope, or a variable parameter of the extreme gradient; and The blood pressure is generated based on the aforementioned characteristics and the blood pressure model.

15. The method according to claim 1, wherein: The user's body part is the tip of their finger, and The group of blood vessels includes at least one vessel from the palmar transverse arch branch of the digital artery in the tip portion of the finger.

16. The method of claim 1, further comprising: The force model is obtained from a model library based on the type of the mobile device, wherein the type of the mobile device is associated with at least one of the configuration of the IMU, the configuration of the vibration motor, or the configuration of the camera.

17. A mobile device comprising: The system includes a vibration motor, an inertial motion unit (IMU), a camera, a processor, and a non-transitory computer-readable storage device that stores instructions that, when executed by the processor, cause the processor to perform at least the following operations: The vibration motor is made to vibrate; IMU data is acquired through the IMU of the mobile device, and the IMU data indicates the vibration damping caused by the force applied to the camera by the user's body parts; The applied force is determined based on the force model and the IMU data; The camera of the mobile device captures photoplethysmography (PPG) data of a group of blood vessels in the body part while the force is applied; and The user's blood pressure is determined using oscilloscope technology based on the PPG data and the applied force.

18. The mobile device of claim 17, wherein the force model comprises a machine learning algorithm trained to correlate (1) the IMU data, measured by the mobile device, indicating the vibration damping of the mobile device, with (2) the force applied to the mobile device and causing the vibration damping.

19. The mobile device of claim 17, wherein the IMU includes an accelerometer and a gyroscope, and the IMU data includes multi-axis accelerometer data and multi-axis gyroscope data.

20. One or more non-transitory computer-readable media storing instructions that, when executed by one or more processors of a mobile device, cause the mobile device to perform operations comprising: The vibration motor of the mobile device is made to vibrate; The mobile device uses an inertial measurement unit (IMU) to acquire IMU data indicating vibration damping caused by forces exerted by the user's body parts on the camera of the mobile device. The applied force is determined based on the force model and the IMU data; Using the camera of the mobile device, photoplethysmography (PPG) data of a group of blood vessels in the body part is captured while the force is applied; and The user's blood pressure is determined using oscillometric techniques based on the PPG data and the applied force.

21. The one or more non-transitory computer-readable media of claim 20, wherein the operation further comprises: The force model is obtained from a model library based on the type of the mobile device, wherein the type of the mobile device is associated with at least one of the configuration of the IMU, the configuration of the vibration motor, or the configuration of the camera.

Images