A method for measuring physiological signals and related devices

Abstract

An embodiment of the present application provides a physiological signal measurement method and related devices. By determining the initial physiological signal value and probability density distribution data corresponding to each photoplethysmogram (PPG) data segment based on each PPG data segment, where each PPG data segment is obtained by dividing the PPG data, and the probability density distribution data is used to represent the probabilities corresponding to different initial physiological signal values. According to the matching relationship between the initial physiological signal value and the probability density distribution data corresponding to each PPG data segment, the effective initial physiological signal value is determined from multiple initial physiological signal values. Based on the effective initial physiological signal value, the physiological signal measurement result is determined. The physiological signal measurement method provided by the embodiment of the present application can effectively improve the accuracy of the physiological signal value by determining the physiological signal value based on the initial physiological signal value with a relatively high confidence level.

Description

Technical Field

[0001] This application relates to the field of electronic technologies, and in particular, to a method for measuring physiological signals and related devices. Background Art

[0002] Photoplethysmography (PPG) is a non-invasive physiological signal measurement technology. It irradiates the human skin with a light-emitting diode (LED), and uses a photodiode to measure the change in the intensity of the reflected light caused by blood flow, obtaining a periodic waveform containing pulsatile change information of the blood volume within the cardiac cycle.

[0003] The method of calculating systolic blood pressure and diastolic blood pressure based on the periodic waveform obtained by PPG to obtain the physiological signal measurement result has the problem of low accuracy of the physiological signal measurement result due to individual differences.

[0004] Therefore, there is an urgent need for a solution to solve the above technical problems. Summary of the Invention

[0005] A method for measuring physiological signals and related devices provided by this application aims to improve the accuracy of physiological signal measurement results.

[0006] To achieve the above objective, this application adopts the following technical solutions:

[0007] In a first aspect: An embodiment of this application provides a method for measuring physiological signals. By determining, for each PPG data segment, a corresponding initial physiological signal value and probability density distribution data; each PPG data segment is obtained by dividing PPG data, and the probability density distribution data is used to represent the probabilities corresponding to different initial physiological signal values; based on the matching relationship between the initial physiological signal value corresponding to each PPG data segment and the probability density distribution data, an effective initial physiological signal value is determined from multiple initial physiological signal values; based on the effective initial physiological signal value, a physiological signal measurement result is determined.

[0008] In the embodiment of this application, according to the matching relationship between the initial physiological signal value corresponding to each PPG data segment and the probability density distribution data, initial physiological signal values with low confidence can be filtered out, and an effective initial physiological signal value is determined from multiple initial physiological signal values. The method for measuring physiological signals provided by the embodiment of this application determines the physiological signal value based on the initial physiological signal value with high confidence, which can effectively improve the accuracy of the physiological signal value.

[0009] In a possible implementation, according to the matching relationship between the initial physiological signal value corresponding to each PPG data segment and the probability density distribution data, an effective initial physiological signal value is determined from multiple initial physiological signal values, including: according to the matching relationship between each initial physiological signal value corresponding to each PPG data segment and the probability density distribution data, determining the probability interval corresponding to each initial physiological signal value; when the probability interval corresponding to an initial physiological signal value is the first interval or the second interval, determining the initial physiological signal value as an effective initial physiological signal value, and when the probability interval corresponding to an initial physiological signal value is the third interval, determining the initial physiological signal value as an ineffective initial physiological signal value; the integral value of the probability density distribution data in the first interval is the highest, the integral value in the second interval is less than the integral value in the first interval and greater than the integral value in the third interval. In the embodiments of the present application, the effective initial physiological signal value is determined by dividing the probability interval, reducing the difficulty of the classification task.

[0010] In a possible implementation, all effective initial physiological signal values correspond to the second interval. Based on the effective initial physiological signal values, a physiological signal measurement result is determined, including: based on the probability density distribution data, determining the probability of each effective initial physiological signal value; multiplying each effective initial physiological signal value by the probability of each effective initial physiological signal value to determine the weighted physiological signal value corresponding to each effective initial physiological signal value; based on the mean value of each weighted physiological signal value, determining the physiological signal measurement result, which can effectively improve the accuracy of the physiological signal measurement result.

[0011] In a possible implementation, the effective initial physiological signal values include the effective initial physiological signal values corresponding to the first interval and the effective initial physiological signal values corresponding to the second interval. Based on the effective initial physiological signal values, a physiological signal measurement result is determined, including: for the effective initial physiological signal values corresponding to the second interval, multiplying each effective initial physiological signal value by the probability of the effective initial physiological signal value to determine the weighted physiological signal value corresponding to each effective initial physiological signal value; the probability of the effective initial physiological signal value is determined based on the probability density distribution data; based on the weighted physiological signal values of the effective initial physiological signal values corresponding to the second interval and the mean value of the effective initial physiological signal values corresponding to the first interval, determining the physiological signal measurement result, which can effectively improve the accuracy of the physiological signal measurement result.

[0012] In a possible implementation, according to the matching relationship between the initial physiological signal value corresponding to each PPG data segment and the probability density distribution data, an effective initial physiological signal value is determined from multiple initial physiological signal values, including: determining the probability of the initial physiological signal value according to the matching relationship between each initial physiological signal value corresponding to each PPG data segment and the probability density distribution data; when the probability of an initial physiological signal value is greater than the first threshold, determining the initial physiological signal value as an effective initial physiological signal value, and when the probability of an initial physiological signal value is less than or equal to the first threshold, determining the initial physiological signal value as an ineffective initial physiological signal value.

[0013] In a possible implementation, based on each photoplethysmography (PPG) data segment, an initial physiological signal value and probability density distribution data corresponding to each PPG data segment are determined, including: performing numerical regression of the physiological signal on each PPG data segment based on a regression model to determine the initial physiological signal value corresponding to each PPG data segment; performing classification of the physiological signal on each PPG data segment based on a classification model to determine the probability density distribution data corresponding to each PPG data segment; the regression model and the classification model are obtained by training based on training set data; the generation process of the training set data includes: obtaining PPG data and standard data collected in multiple time periods; dividing the PPG data collected in each time period into a first data segment and a second data segment; mapping the first data segment of the Nth time period to the standard data of the (N - 1)th time period, and mapping the second data segment of the Nth time period to the standard data of the Nth time period to obtain the training set data, where N is a positive integer greater than 0. In the embodiments of the present application, based on the regression model and the classification model, the initial physiological signal value and the probability density distribution data are determined, which can effectively improve the efficiency and accuracy of determining the physiological signal measurement result.

[0014] In a possible implementation, the probability density distribution data follows a Gaussian distribution, and the training steps of the classification model include: using the training set data as input and outputting the probability density distribution data corresponding to the training set data, and the mean value of the probability density distribution data is determined based on the standard data in the training set data. In the embodiments of the present application, the classification of the physiological signal is implemented based on the probability density distribution function of the Gaussian distribution, which reduces the difficulty of the classification task and improves the accuracy of the classification.

[0015] In a possible implementation, when the physiological signal is blood pressure, the initial value of the physiological signal includes the initial value of systolic blood pressure and the initial value of diastolic blood pressure. The probability density distribution data includes the first distribution data corresponding to the initial value of systolic blood pressure and the second distribution data corresponding to the initial value of diastolic blood pressure. Determining the effective initial value of the physiological signal from multiple initial values of the physiological signal according to the matching relationship between the initial value of the physiological signal corresponding to each PPG data segment and the probability density distribution data includes: determining the effective initial value of systolic blood pressure from multiple initial values of systolic blood pressure according to the initial value of systolic blood pressure corresponding to each PPG data segment and the first distribution data, and determining the effective initial value of diastolic blood pressure from multiple initial values of diastolic blood pressure according to the initial value of diastolic blood pressure corresponding to each PPG data segment and the second distribution data; determining the physiological signal measurement result based on the effective initial value of the physiological signal includes: determining the mean value of the effective initial values of systolic blood pressure as the systolic blood pressure value, and determining the effective initial value of diastolic blood pressure as the diastolic blood pressure value; determining the physiological signal measurement result based on the systolic blood pressure value and the diastolic blood pressure value.

[0016] In a possible implementation, before determining the initial value of the physiological signal and the probability density distribution data corresponding to each PPG data segment based on each photoplethysmography (PPG) data segment, it further includes: receiving the PPG data sent by the wearable device; after determining the physiological signal measurement result based on the effective initial value of the physiological signal, it further includes: sending the physiological signal value to the wearable device to display the physiological signal measurement result on the display interface of the wearable device for the user to view.

[0017] In a possible implementation, determining the physiological signal measurement result based on the effective initial value of the physiological signal includes: when there are multiple effective initial values of the physiological signal, determining the mean value of the multiple effective initial values of the physiological signal as the physiological signal value; determining the level corresponding to the physiological signal value based on the physiological signal value; determining the physiological signal value and / or the level as the physiological signal measurement result. In the embodiments of the present application, the physiological signal measurement result can be the physiological signal value and / or the level, which is convenient for the user to view and improves the user experience.

[0018] Second aspect: An embodiment of the present application provides an electronic device, which includes: a processor and a memory;

[0019] The memory is used to store program code and transmit the program code to the processor;

[0020] The processor is used to execute the steps of a physiological signal measurement method as described above according to the instructions in the program code.

[0021] Third aspect: The present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of a physiological signal measurement method as described above are implemented. Description of the Drawings

[0022] Figure 1 A schematic diagram of a physiological signal measurement provided by an embodiment of the present application;

[0023] Figure 2 A schematic diagram of a periodic waveform provided by an embodiment of the present application;

[0024] Figure 3 A schematic diagram of an application scenario of a physiological signal measurement method provided by an embodiment of the present application;

[0025] Figure 4 A flowchart of a physiological signal measurement method provided by an embodiment of the present application;

[0026] Figure 5 A schematic diagram of data division provided by an embodiment of the present application;

[0027] Figure 6 Another schematic diagram of data division provided by an embodiment of the present application;

[0028] Figure 7 A schematic diagram of data acquisition provided by an embodiment of the present application;

[0029] Figure 8 Another schematic diagram of data acquisition provided by an embodiment of the present application;

[0030] Figure 9 A schematic diagram of data classification based on hard labels provided by an embodiment of the present application;

[0031] Figure 10 A schematic diagram of data classification based on soft labels provided by an embodiment of the present application;

[0032] Figure 11 A schematic diagram of data classification based on Gaussian labels provided by an embodiment of the present application;

[0033] Figure 12 A schematic diagram of a blood pressure measurement method provided by an embodiment of the present application;

[0034] Figure 13 A schematic diagram of a physiological signal measurement algorithm provided by an embodiment of the present application;

[0035] Figure 14 A schematic diagram of the hardware structure of an electronic device provided by an embodiment of the present application;

[0036] Figure 15 A software structure block diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0037] The terms "first", "second", "third", etc. in the description, claims and drawings of this application are used to distinguish different objects, rather than to limit a specific order.

[0038] In the embodiments of this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design solution described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Rather, the use of words such as "exemplary" or "for example" is intended to present relevant concepts in a specific manner.

[0039] Photoplethysmography (PPG) is a non-invasive physiological signal measurement technology. As Figure 1 shown, when the user's finger touches the PPG sensor, the light emitted by the light-emitting diode (LED) in the PPG sensor can pass through the tissues and arteries and veins in the skin, and be absorbed and reflected back to the photodetector (PD) of the photodiode. In the case where there is no large movement at the measurement site of the user, the absorption of light by muscles, bones, veins and other connective tissues at the measurement site is basically unchanged. However, the blood is different. Due to the flow of blood in the artery, the absorption of light by the blood will change.

[0040] After converting the optical signal into an electrical signal, since the absorption of light by the artery changes while the absorption of light by other tissues remains basically unchanged, the electrical signal can be divided into a DC signal and an AC signal. Among them, the AC signal can characterize the characteristics of blood flow. By extracting the AC signal from the electrical signal, a periodic waveform containing pulsatile change information of blood volume is obtained based on the AC signal. As Figure 2 shown, the figure shows a periodic waveform containing pulsatile change information of blood volume obtained based on PPG. The trough of this periodic waveform indicates that the heart is in a diastolic state, and the peak indicates that the heart is in a systolic state.

[0041] In a first aspect, although the systolic blood pressure and diastolic blood pressure can be calculated based on the periodic waveform obtained through PPG to obtain the physiological signal measurement result, this method is greatly affected by individual differences, and there is a problem of low accuracy of the physiological signal measurement result.

[0042] In a second aspect, in order to obtain a more accurate physiological signal measurement result, currently, after a meta-model is trained based on a data set, the meta-model can be personalized calibrated based on the user's physiological signal data to obtain a physiological signal prediction model for measuring the physiological signal of this user, which has the problem of poor generality.

[0043] Based on this, an embodiment of the present application provides a physiological signal measurement method. By dividing the PPG data, multiple PPG data segments are obtained. Based on each PPG data segment, the initial physiological signal value and probability density distribution data corresponding to each PPG data segment are determined. Among them, the probability density distribution data is used to represent the probabilities corresponding to different initial physiological signal values. On this basis, in the embodiment of the present application, according to the matching relationship between the initial physiological signal value corresponding to each PPG data segment and the probability density distribution data, the initial physiological signal values with low confidence can be filtered out, and the effective initial physiological signal values are determined from multiple initial physiological signal values. Based on the effective initial physiological signal values, the physiological signal measurement result is determined. The physiological signal measurement method provided by the embodiment of the present application can effectively improve the accuracy of the physiological signal value by determining the physiological signal value based on the initial physiological signal value with high confidence.

[0044] The following introduces the application scenario of a physiological signal measurement method provided by an embodiment of the present application with reference to the accompanying drawings.

[0045] Taking the electronic device as an electronic watch and the physiological signal as blood pressure as an example. As Figure 3 shown, after the user wears the electronic watch, by clicking the blood pressure option 1301 in the first interface 1300 of the electronic watch, the electronic watch can respond to the user's operation, obtain photoplethysmogram (PPG) data, and send the PPG data to the cloud server.

[0046] The cloud server receives the PPG data sent by the electronic device, divides the PPG data to obtain multiple PPG data segments. Based on each PPG data segment, the initial blood pressure value and probability density distribution data corresponding to each PPG data segment are determined, where the probability density distribution data is used to represent the probabilities corresponding to different initial blood pressure values.

[0047] After determining the initial blood pressure value and probability density distribution data corresponding to each PPG data segment, the cloud server can determine the effective initial blood pressure value from multiple initial blood pressure values according to the matching relationship between the initial blood pressure value corresponding to each PPG data segment and the probability density distribution data, and determine the blood pressure measurement result based on the effective initial blood pressure value.

[0048] As Figure 3 shown, during the process of the cloud server determining the blood pressure measurement result based on the PPG data, the words "Blood pressure measurement in progress, please stay still" can be displayed on the second interface 1302 of the electronic watch to prompt the user to stay still to improve the accuracy of the blood pressure measurement result.

[0049] After determining the blood pressure measurement result, the cloud server sends the blood pressure measurement result to the electronic watch, and the corresponding blood pressure measurement result can be displayed on the third interface 1303 of the electronic watch. Among them, the blood pressure measurement result can include at least one of a blood pressure value and a blood pressure level.

[0050] It can be understood that in the figure, only the example where the blood pressure measurement result includes a blood pressure value is taken, and the blood pressure value includes a low blood pressure value (diastolic blood pressure value) and a high blood pressure value (systolic blood pressure value).

[0051] Next, a physiological signal measurement method provided by an embodiment of the present application will be introduced. As Figure 4 shown, this figure is a flowchart of a physiological signal measurement method provided by an embodiment of the present application, including S401 - S407.

[0052] S401. The electronic device acquires photoplethysmography (PPG) data.

[0053] In an embodiment of the present application, after the user wears the electronic device, the electronic device can respond to the user's measurement operation and acquire photoplethysmography (PPG) data.

[0054] S402. The electronic device sends the photoplethysmography data to the cloud server.

[0055] After obtaining the PPG data, the electronic device can send the PPG data to the cloud server so that the cloud server can process and analyze the PPG data to obtain a physiological signal measurement result. In an embodiment of the present application, the physiological signal can include but is not limited to blood pressure, heart rate, blood oxygen saturation, etc.

[0056] S403. The cloud server divides the photoplethysmography data to obtain multiple photoplethysmography data segments.

[0057] In an embodiment of the present application, the cloud server can divide a segment of PPG data sent by the electronic device to obtain multiple PPG data segments, so that the initial physiological signal values of the multiple PPG data segments can be screened subsequently to determine the effective initial physiological signal values and obtain an accurate physiological signal measurement result.

[0058] In one example, the PPG data can be divided according to the duration. As Figure 5 shown, dividing a segment of PPG data with a duration of 1 minute (min) and a frequency of 100 Hz can obtain 6 PPG data segments, and the duration of each PPG data segment is 10 seconds.

[0059] In another example, the PPG data can be divided according to the cardiac cycle. As Figure 6As shown in the figure, every two cardiac cycles are determined as a PPG data segment, and multiple PPG data segments can be obtained. For example, by dividing a segment of PPG data containing 10 cardiac cycles, 5 PPG data segments can be obtained, and each PPG data segment contains two cardiac cycles.

[0060] It can be understood that in the embodiments of the present application, the duration of each PPG data segment or the number of cardiac cycles included in the PPG data segment is not specifically limited, and the above is only an example.

[0061] S404. The cloud server determines the initial physiological signal value and probability density distribution data corresponding to each photoplethysmogram (PPG) data segment based on each PPG data segment.

[0062] Among them, the probability density distribution data is used to represent the probabilities corresponding to different initial physiological signal values.

[0063] In the embodiments of the present application, the physiological signal may include but is not limited to blood pressure, heart rate, etc. When the physiological signal is blood pressure, the initial physiological signal value corresponding to each PPG data segment is the blood pressure value corresponding to each PPG data segment; when the physiological signal is heart rate, the initial physiological signal value corresponding to each PPG data segment is the heart rate value corresponding to each PPG data segment.

[0064] In a possible implementation manner, the cloud server may perform physiological signal numerical regression on each PPG data segment based on a regression model to determine the initial physiological signal value corresponding to each PPG data segment; and perform physiological signal classification on each PPG data segment based on a classification model to determine the probability density distribution data corresponding to each PPG data segment.

[0065] In the embodiments of the present application, the regression model and the classification model can be trained based on training set data.

[0066] In a possible implementation manner, PPG data and standard data can be collected, and the best PPG data segments are selected from the PPG data, and the best PPG data segments and the corresponding standard data are used as the training set.

[0067] Among them, the standard data is the data measured by using traditional measurement tools, and it can be used as accurate physiological signal data to be matched with the PPG data. For example, in the case where the physiological signal is blood pressure, the standard data may be the blood pressure value measured by using a traditional mercury sphygmomanometer.

[0068] Such as Figure 7As shown, during each round of data acquisition, 1 minute of PPG data can be acquired first, and then standard data can be measured using traditional measurement tools. After that, to improve the accuracy of the next round of data acquisition, a 30-second break can be taken after measuring the standard data, and the next round of data acquisition can start after 30 seconds. In this way, multiple PPG data and standard data can be obtained.

[0069] To improve the accuracy of the training dataset, each segment of PPG data can be divided to obtain multiple PPG data segments. For example, Figure 5 As shown, dividing 1 minute of PPG data into windows of 10 seconds can obtain 6 PPG data segments. Based on the data quality of the PPG data segments, 5 of the best PPG data segments can be selected from the 6 PPG data segments. The best PPG data segments obtained through screening and the corresponding standard data are used as training data.

[0070] It can be understood that in the embodiments of the present application, the duration of PPG data acquisition, the rest duration, the number of rounds of data acquisition, the number of PPG segments, and the number of the best PPG data segments selected from each PPG data are not specifically limited, and the above are only examples.

[0071] In another possible implementation, since PPG data is affected by factors such as the environment, the user's wearing situation, and the exercise situation, its quality is unstable. For example, in 1 minute of PPG data, the data in the last 30 seconds is usually more consistent with the actually measured standard data.

[0072] To improve the accuracy of the training set data, in the embodiments of the present application, the training set data can be generated through steps 411 - 413 described below.

[0073] Step 411: Obtain PPG data and standard data collected in multiple time periods.

[0074] In the embodiments of the present application, multiple rounds of data acquisition can be performed during the data acquisition stage. For example, Figure 8 Taking four rounds of data acquisition as an example for introduction.

[0075] Each round of data acquisition process includes 1 minute of PPG data acquisition, standard data acquisition, and a 30-second rest time. In this way, PPG data and standard data collected in multiple time periods can be obtained.

[0076] It can be understood that in the embodiments of the present application, the duration of PPG data acquisition, the rest duration, and the number of rounds of data acquisition are not specifically limited, and the above are only examples.

[0077] Step 412: Divide the PPG data collected in each time period into a first data segment and a second data segment.

[0078] Among them, the first data segment in the Nth time period is at least one PPG data segment that matches the standard data in the (N - 1)th time period; the second data segment in the Nth time period is at least one PPG data segment that matches the standard data in the Nth time period.

[0079] Step 413: Map the first data segment in the Nth time period to the standard data in the (N - 1)th time period, and map the second data segment in the Nth time period to the standard data in the Nth time period to obtain training set data.

[0080] Where N is a positive integer greater than 0.

[0081] Exemplarily, the 1-minute PPG data is divided into 6 PPG data segments, and the duration of each PPG data segment is 10 seconds. For each PPG data, the PPG data segment from 0 to 10 seconds can be marked as index 0; the PPG data segment from 11 to 20 seconds can be marked as index 1; the PPG data segment from 21 to 30 seconds can be marked as index 2; the PPG data segment from 31 to 40 seconds can be marked as index 3; the PPG data segment from 41 to 50 seconds can be marked as index 4; the PPG data segment from 51 to 60 seconds can be marked as index 5.

[0082] When the physiological signal fluctuation is unchanged or small before and after a 30-second rest, the PPG data segments with index 0, 1, and 2 in each PPG data can be used as the first data segment, and the PPG data segments with index 3, 4, and 5 in each PPG data can be used as the second data segment.

[0083] When N is 3, map the PPG data segments with index 0, 1, and 2 in the 1st time period to the standard data in the 0th time period, and map the PPG data segments with index 3, 4, and 5 in the 1st time period to the standard data in the 1st time period; map the PPG data segments with index 0, 1, and 2 in the 2nd time period to the standard data in the 1st time period, and map the PPG data segments with index 3, 4, and 5 in the 2nd time period to the standard data in the 2nd time period; map the PPG data segments with index 0, 1, and 2 in the 3rd time period to the standard data in the 2nd time period, and map the PPG data segments with index 3, 4, and 5 in the 3rd time period to the standard data in the 3rd time period to obtain training set data.

[0084] In an embodiment of the present application, training set data is obtained by mapping the first data segment of the Nth time period with the standard data of the N-1th time period, and mapping the second data segment of the Nth time period with the standard data of the Nth time period. The training set data collection process is optimized from the time scale to facilitate the subsequent improvement of the accuracy of the physiological signal detection results.

[0085] In a possible implementation, based on the classification model, each PPG data segment is classified as a physiological signal, and the probability density distribution data corresponding to each PPG data segment obtained is Gaussian distributed. In this case, during the training process of the classification model, the training set data can be used as input to output the probability density distribution data corresponding to the training set data. The mean value of the probability density distribution data is determined based on the standard data in the training set data.

[0086] like Figure 9 As shown, data can be classified based on soft label or hard label classification. Taking the physiological signal as blood pressure value as an example, based on the distribution of training set data, the hard label classification method can determine that the probability of the diastolic blood pressure value being within the range of 90mmHg-100mmHg is 1, and the probability of being outside the range is 0.

[0087] Compared with the hard label classification method, the soft label classification method has a certain degree of fault tolerance, such as Figure 10 As shown, based on the distribution of the training set data, the soft label classification method can determine that the probability that the diastolic blood pressure value is in the range of 80 mmHg-90 mmHg is 0.3, the probability that it is in the range of 90 mmHg-100 mmHg is 0.7, and the probability that it is in other ranges is 0.

[0088] However, the methods of classifying physiological signals based on soft labels or hard labels all have the problem of poor continuity in interval distribution. In order to reduce the difficulty of the classification task and improve the accuracy of classification, Gaussian labels are used to classify physiological signals in the embodiments of the present application.

[0089] In the embodiment of the present application, based on the distribution of the training set data, the probability interval window length corresponding to the physiological signal can be determined, and the probability interval of the physiological signal can be divided. Based on the statistical results of the training set data and the blood pressure acceptance criteria, the mean and variance of the Gaussian curve can be determined to generate probability density distribution data with Gaussian distribution, and the steps are as follows:

[0090] Step 421: Define input parameters.

[0091] The input parameters include the probability intervals (intervals), given values ​​(sample point) and standard deviation (std) of the physiological signal.

[0092] Taking the physiological signal of blood pressure as an example, based on the distribution of the training set data, determine the probability interval window length corresponding to the physiological signal, and divide to obtain the probability interval of the physiological signal. Determine the given value according to the standard data in the training set, and set the standard deviation of the Gaussian curve.

[0093] As Figure 11 shown, the probability interval window length corresponding to the physiological signal can be 10 mm Hg, and the probability intervals of the physiological signal obtained by division can include (80, 90), (90, 100), …, (150, 160), etc. The given value is 92 mm Hg, and the standard deviation of the Gaussian curve can be set to 10 / 3.

[0094] Step 422: Initialize the mean.

[0095] Assign the given value to the mean. Exemplarily, according to the standard data in the training set, the mean of the standard data can be determined, used as the given value, and the mean is initialized based on this given value.

[0096] Step 423: Initialize the probability list.

[0097] Create an empty list and name it probabilities.

[0098] Step 424: Traverse the probability intervals of the physiological signal, determine the probabilities corresponding to the probability intervals of each physiological signal, and add them to the probability list.

[0099] For each probability interval (intervals) of the physiological signal, extract the start value (start) and end value (end) of the probability interval of the physiological signal, and calculate the cumulative distribution function (CDF) of the start value and end value, as shown in Equation (1) below:

[0100] cdf_start = norm.cdf(start, loc = mean, scale = std)

[0101] cdf_end = norm.cdf(end, loc = mean, scale = std) (1)

[0102] Among them, the norm.cdf function is used to calculate the value of the cumulative distribution function of the normal distribution. The cumulative distribution function is used to represent the probability that a random variable takes a value at or below a certain specific value; cdf_start is a variable used to store the value of the cumulative distribution function corresponding to the starting value, and this value of the cumulative distribution function is the probability that the random variable is less than or equal to the starting value; cdf_end is a variable used to store the value of the cumulative distribution function corresponding to the ending value, and this value of the cumulative distribution function is the probability that the random variable is less than or equal to the ending value; loc is a keyword parameter in the norm.cdf function, loc = mean, indicating that loc specifies the mean of the normal distribution (mean). In the normal distribution, the mean is the center point of the distribution and determines the position of the distribution; scale is another keyword parameter in the norm.cdf function, scale = std, indicating that scale specifies the standard deviation of the normal distribution. The standard deviation determines the width or dispersion degree of the distribution. The larger the standard deviation, the wider the distribution, and the smaller the standard deviation, the narrower the distribution.

[0103] The probability corresponding to the probability interval of the physiological signal can be expressed by Equation (2) as follows:

[0104] probability = cdf_end - cdf_start (2)

[0105] Among them, probability represents the probability corresponding to the probability interval of the physiological signal. After obtaining the probability corresponding to the probability interval of each physiological signal, the probability corresponding to the probability interval of each physiological signal can be added to the probability list.

[0106] Step 425: Obtain probability density distribution data based on the probability list.

[0107] In the embodiment of the present application, to improve the data processing efficiency, a probability density function with a Gaussian distribution can be obtained by converting probabilities into a NumPy array and returning it.

[0108] In the embodiment of the present application, probability intervals are used for physiological signal classification, and probability density distribution data with a Gaussian distribution is used to represent the probabilities corresponding to different initial physiological signal values, reducing the difficulty of the physiological signal classification task.

[0109] S405. The cloud server determines valid initial physiological signal values from multiple initial physiological signal values according to the matching relationship between the initial physiological signal values corresponding to each photoplethysmography data segment and the probability density distribution data.

[0110] In a possible implementation, the cloud server can determine the probability interval corresponding to each initial physiological signal value based on the matching relationship between each initial physiological signal value corresponding to each PPG data segment and the probability density distribution data.

[0111] Exemplarily, as Figure 11 shown, the central value of the probability density distribution data is 92. If the initial physiological signal value is between 90 and 100, the probability interval corresponding to this initial physiological signal value is (90, 100); if the initial physiological signal value is between 80 and 90, the probability interval corresponding to this initial physiological signal value is (80, 90), and so on.

[0112] When the probability interval corresponding to an initial physiological signal value is the first interval or the second interval, the cloud server can determine this initial physiological signal value as a valid initial physiological signal value. When the probability interval corresponding to an initial physiological signal value is the third interval, the cloud server can determine the initial physiological signal value as an invalid initial physiological signal value.

[0113] Among them, the integral value of the probability density distribution data in the first interval is the highest, the integral value in the second interval is less than that in the first interval, and greater than that in the third interval.

[0114] Exemplarily, as Figure 11 shown, the integral value of the probability density distribution data in the 90 - 100 probability interval is the highest, the integral value in the 80 - 90 probability interval is less than its integral value in the 90 - 100 probability interval, and greater than its integral value in other probability intervals. (90, 100) is the first interval, (80, 90) is the second interval, and the third interval is other probability intervals except the first interval and the second interval. When the probability interval corresponding to the initial physiological signal value is (80, 100), the cloud server can determine this initial physiological signal value as a valid initial physiological signal value; when the probability interval corresponding to the initial physiological signal value is not within the range of (80, 100), the cloud server can determine this initial physiological signal value as an invalid initial physiological signal value.

[0115] Among them, the range of (80, 100) can include at least one of 80 and 100, or can also not include 80 and 100.

[0116] It can be understood that in the above examples, values such as 80, 90, and 100 are boundary values of each interval, and the probability intervals to which each boundary value belongs can be set as needed. For example, it can be set that the first interval includes [90, 100), the second interval includes [80, 90), that is, 80 belongs to the second interval, 90 belongs to the first interval, and 100 belongs to the third interval; or it can be set that the first interval includes (90, 100], the second interval includes (80, 90], that is, 80 belongs to the third interval, 90 belongs to the second interval, and 100 belongs to the first interval; or it can be set that the first interval includes [90, 100], the second interval includes [80, 90], that is, 80 belongs to the second interval, 90 belongs to the first interval and the second interval, and 100 belongs to the first interval.

[0117] In a possible implementation manner, after the cloud server determines the probability of the initial physiological signal according to the matching relationship between each initial physiological signal value corresponding to each PPG data segment and the probability density distribution data, it can determine whether the initial physiological signal value is a valid initial physiological signal value or an invalid initial physiological signal value according to the magnitude relationship between the initial physiological signal value and the first threshold.

[0118] Exemplarily, when the probability of an initial physiological signal value is greater than the first threshold, the cloud server determines the initial physiological signal value as a valid initial physiological signal value; when the probability of an initial physiological signal value is less than or equal to the first threshold, the cloud server determines the initial physiological signal value as an invalid initial physiological signal value.

[0119] It can be understood that in the embodiments of the present application, the magnitude of the first threshold is not specifically limited and can be set according to actual needs.

[0120] S406. The cloud server determines the physiological signal measurement result based on the valid initial physiological signal value.

[0121] In the embodiments of the present application, the physiological signal measurement result may include at least one of specific physiological signal values and levels corresponding to the physiological signal values.

[0122] Exemplarily, the specific physiological signal value may be a blood pressure value or a heart rate value, etc., and the level corresponding to the physiological signal value may be normal blood pressure, high blood pressure, normal heart rate, high heart rate, etc.

[0123] In a possible implementation manner, when there are multiple valid initial physiological signal values, the cloud server may determine the average value of the multiple valid initial physiological signal values as the physiological signal value and determine the physiological signal measurement result based on this physiological signal value.

[0124] In an example, the cloud server may directly use the physiological signal value as the physiological signal measurement result.

[0125] In another example, the cloud server may determine the level corresponding to the physiological signal value based on the physiological signal value, and determine the physiological signal value and / or the level as the physiological signal measurement result.

[0126] In the embodiments of the present application, to improve the accuracy of the physiological signal measurement result, the cloud server may process the initial value of the valid physiological signal for the probability interval corresponding to the initial value of the valid physiological signal, and then determine the physiological signal measurement result.

[0127] In a possible implementation manner, when the initial values of the valid physiological signals all correspond to the first interval, the cloud server may determine the physiological signal measurement result based on the mean value of each initial value of the valid physiological signal.

[0128] In the embodiments of the present application, when the initial value of the valid physiological signal corresponds to the first interval, it indicates that the credibility of the initial value of the valid physiological signal is high, and the physiological signal value may be directly determined based on the initial value of the valid physiological signal in the first interval.

[0129] Exemplarily, among multiple initial values of physiological signals, 3 initial values of valid physiological signals, a1, a2, and a3, are determined, where a1, a2, and a3 all correspond to the first interval. Then, based on a1, a2, and a3, the mean value of each initial value of the valid physiological signal may be used to determine the physiological signal value as (a1, a2, a3) / 3.

[0130] After determining the physiological signal value, the cloud server may directly use the physiological signal value as the physiological signal measurement result; or determine the level corresponding to the physiological signal value based on the physiological signal value, and determine the physiological signal value and / or the level as the physiological signal measurement result.

[0131] In a possible implementation manner, when the initial values of the valid physiological signals all correspond to the second interval, it indicates that the credibility of the initial value of the valid physiological signal is relatively high. To improve the accuracy of the physiological signal measurement signal, the cloud server may determine the probability of each initial value of the valid physiological signal based on the probability density distribution data.

[0132] By multiplying each initial value of the valid physiological signal by the probability of each initial value of the valid physiological signal to obtain the weighted initial value of the valid physiological signal corresponding to each initial value of the valid physiological signal, the cloud server may determine the physiological signal measurement result based on the mean value of each weighted initial value of the valid physiological signal.

[0133] Wherein, the probability of the initial value of the valid physiological signal is the output value obtained by using the initial value of the valid physiological signal as the input value of the probability density distribution function.

[0134] Exemplarily, 3 initial values of valid physiological signals, b1, b2, and b3, are determined from multiple initial values of physiological signals, where b1, b2, and b3 all correspond to the second interval.

[0135] Based on the probability density distribution data, the probability of b1 is determined to be w1, the probability of b2 is w2, and the probability of b3 is w3. Then, the initial weighted physiological signal corresponding to b1 is b1×w1; the initial weighted physiological signal corresponding to b2 is b2×w2; the initial weighted physiological signal corresponding to b3 is b3×w3. After obtaining the initial weighted physiological signals, the physiological signal value can be determined as the mean of each initial weighted physiological signal, that is, (b1×w1 + b2×w2 + b3×w3) / 3.

[0136] After determining the physiological signal value, the cloud server can directly use the physiological signal value as the physiological signal measurement result; or based on the physiological signal value, determine the level corresponding to the physiological signal value, and determine the physiological signal value and / or the level as the physiological signal measurement result.

[0137] In a possible implementation, when the valid initial physiological signals include the valid initial physiological signals corresponding to the first interval and the valid initial physiological signals corresponding to the second interval, for the valid initial physiological signals corresponding to the second interval, the cloud server can determine the product of each valid initial physiological signal and the probability of the valid initial physiological signal as the initial weighted physiological signal corresponding to each valid initial physiological signal. On this basis, the cloud server can determine the physiological signal measurement result based on the initial weighted physiological signals corresponding to the valid initial physiological signals in the second interval and the mean of the valid initial physiological signals in the first interval.

[0138] Exemplarily, three valid initial physiological signals are determined from multiple initial physiological signals as c1, c2, and c3. Among them, c1 corresponds to the first interval, and c2 and c3 correspond to the second interval. Then, based on the probability density distribution function, the probability of c2 can be determined to be w4, and the probability of c3 can be determined to be w5.

[0139] On this basis, the cloud server can determine that the initial weighted physiological signal corresponding to c2 is c2×w4, and the initial weighted physiological signal corresponding to c3 is c3×w5. Thus, the physiological signal value can be determined as (c1 + c2×w4 + c3×w5) / 3.

[0140] After determining the physiological signal value, the cloud server can directly use the physiological signal value as the physiological signal measurement result; or based on the physiological signal value, determine the level corresponding to the physiological signal value, and determine the physiological signal value and / or the level as the physiological signal measurement result.

[0141] S407. The electronic device receives the physiological signal measurement result sent by the cloud server and displays it.

[0142] In the embodiments of the present application, the physiological signal measurement result may include at least one of a physiological signal value and a level corresponding to the physiological signal value. After receiving the physiological signal measurement result sent by the cloud server, the electronic device may display it on the display screen.

[0143] In summary, in the embodiments of the present application, according to the matching relationship between the initial physiological signal value corresponding to each PPG data segment and the probability density distribution data, the initial physiological signal values with low confidence can be filtered out, and the effective initial physiological signal values can be determined from multiple initial physiological signal values. Based on the effective initial physiological signal values, the physiological signal measurement result is determined. The physiological signal measurement method provided by the embodiments of the present application can effectively improve the accuracy of the physiological signal value by determining the physiological signal value based on the initial physiological signal value with high confidence.

[0144] For better understanding, taking the physiological signal as blood pressure as an example, the physiological signal measurement method provided by the embodiments of the present application will be introduced below. As Figures 12 - 13 shown, Figure 12 is a schematic diagram of a blood pressure measurement method provided by the embodiments of the present application, Figure 13 is a schematic diagram of a physiological signal measurement algorithm provided by the embodiments of the present application.

[0145] In the embodiments of the present application, when the physiological signal is blood pressure, the initial physiological signal value, that is, the blood pressure value corresponding to each PPG data segment, includes the initial systolic blood pressure value and the initial diastolic blood pressure value; the physiological signal value includes the systolic blood pressure value and the diastolic blood pressure value; the probability density distribution data includes the first distribution data corresponding to the initial systolic blood pressure value and the second distribution data corresponding to the initial diastolic blood pressure value.

[0146] The electronic device acquires PPG data and sends the PPG data to the cloud server. The cloud server divides the PPG data into multiple PPG data segments.

[0147] It can be understood that the specific method is the same as that of S401-S403 above and will not be repeated here.

[0148] The cloud server inputs each photoplethysmogram data segment into the blood pressure dual-branch model to determine the initial blood pressure value and the probability density distribution data corresponding to each photoplethysmogram data segment. Among them, the blood pressure dual-branch model includes a regression model and a classification model, and the initial blood pressure value includes the initial systolic blood pressure value and the initial diastolic blood pressure value.

[0149] As Figure 2 or Figure 6 shown, in each PPG data segment, the peak represents the heart in a systolic state, and the trough represents the heart in a diastolic state. Based on this, the cloud server can determine the initial systolic blood pressure value and the initial diastolic blood pressure value corresponding to each PPG segment based on the PPG data segment.

[0150] For the initial systolic blood pressure value and the initial diastolic blood pressure value, based on the method in S404 above, the first distribution data corresponding to the initial systolic blood pressure value and the second distribution data corresponding to the initial diastolic blood pressure value can be obtained.

[0151] The cloud server determines the effective initial systolic blood pressure value from multiple initial systolic blood pressure values according to the initial systolic blood pressure value corresponding to each PPG data segment and the first distribution data. The cloud server determines the effective initial diastolic blood pressure value from multiple initial diastolic blood pressure values according to the initial diastolic blood pressure value corresponding to each PPG data segment and the second distribution data.

[0152] The cloud server determines the mean value of the effective initial systolic blood pressure values as the systolic blood pressure value and determines the effective initial diastolic blood pressure value as the diastolic blood pressure value. On this basis, based on the systolic blood pressure value and the diastolic blood pressure value, the physiological signal measurement result can be determined.

[0153] In the embodiments of the present application, the physiological signal measurement result may include but is not limited to at least one of specific blood pressure values and the levels corresponding to the blood pressure values.

[0154] In one example, the cloud server may directly use the systolic blood pressure value and the diastolic blood pressure value as the physiological signal measurement result.

[0155] In another example, the cloud server may determine the level corresponding to the physiological signal value based on the systolic blood pressure value and the diastolic blood pressure value. The systolic blood pressure value and the diastolic blood pressure value and / or their respective corresponding levels are determined as the physiological signal measurement result.

[0156] As shown in Table 1 below, the table shows the levels corresponding to different systolic blood pressure values and diastolic blood pressure values respectively. When the systolic blood pressure value is less than 120 mmHg and the diastolic blood pressure value is less than 80 mmHg, the blood pressure level is normal blood pressure; when the systolic blood pressure value is between 120 mmHg and 129 mmHg and the diastolic blood pressure value is less than 80 mmHg, the blood pressure level is elevated blood pressure; when the systolic blood pressure value is between 130 mmHg and 139 mmHg or the diastolic blood pressure value is between 80 mmHg and 89 mmHg, the blood pressure level is stage 1 hypertension; when the systolic blood pressure value is greater than or equal to 140 mmHg or the diastolic blood pressure value is greater than or equal to 90 mmHg, the blood pressure level is stage 2 hypertension.

[0157] Table 1

[0158]

[0159] It can be understood that the levels corresponding to the above different systolic blood pressure values and diastolic blood pressure values are only examples.

[0160] After the cloud server determines the physiological signal measurement result, it can send the physiological signal measurement result to the electronic device. After receiving the physiological signal measurement result sent by the cloud server, the electronic device can display it.

[0161] In summary, in the embodiments of the present application, according to the matching relationship between the initial systolic blood pressure value and the initial diastolic blood pressure value corresponding to each PPG data segment and the first distribution data and the second distribution data, the initial systolic blood pressure value and the initial diastolic blood pressure value with low confidence can be filtered out, and the effective initial systolic blood pressure value and the effective initial diastolic blood pressure value can be determined from multiple initial systolic blood pressure values and initial diastolic blood pressure values. Based on the effective initial systolic blood pressure value and the effective initial diastolic blood pressure value, the physiological signal measurement result is determined.

[0162] It can be understood that the method provided by the embodiments of the present application can also be applied to the measurement of physiological signals such as heart rate and blood oxygen saturation. The specific method is similar to blood pressure measurement and will not be elaborated here.

[0163] In some embodiments, the cloud server can determine at least one of physiological signals such as blood pressure value, heart rate value, and blood oxygen saturation based on the PPG data, and determine the physiological signal measurement result based on at least one of the physiological signals such as blood pressure value, heart rate value, and blood oxygen saturation.

[0164] For example, when the cloud server determines the blood pressure value and the heart rate value based on the PPG data, the physiological signal measurement result may include the blood pressure value and the heart rate value; or include the blood pressure value, the level of the blood pressure value, the heart rate value, and the level of the heart rate value. Thus, after the electronic device obtains the physiological signal measurement result sent by the cloud server, it can display the relevant measurement results of blood pressure and heart rate at the same time.

[0165] The physiological signal measurement method provided by the embodiments of the present application determines the physiological signal value based on the physiological signal initial value with high confidence, which can effectively improve the accuracy of the physiological signal value, and further improve the accuracy of the physiological signal measurement result.

[0166] The electronic device in the embodiments of the present application may be a wearable device, such as an electronic watch, an electronic bracelet, and other electronic clothing. The embodiments of the present application do not limit the specific technologies and specific device forms adopted by the electronic device.

[0167] In order to better understand the embodiments of the present application, the structure of the electronic device in the embodiments of the present application will be introduced below. Figure 14 The schematic diagram of the hardware structure of the electronic device is shown.

[0168] The electronic device may include a processor 1410, a mobile communication module 1420, a wireless communication module 1421, a sensor module 1430, a button 1440, and a display screen 1450, etc.

[0169] It can be understood that the structure illustrated in the embodiments of the present application does not constitute a specific limitation on the electronic device. In other embodiments of the present application, the electronic device may include more or fewer components than those illustrated, or combine certain components, or split certain components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0170] The processor 1410 may include one or more processing units. For example, the processor 1410 may include an application processor (AP), a modem processor, a graphics processing unit (GPU), an image signal processor (ISP), etc. Among them, different processing units may be independent devices or integrated in one or more processors.

[0171] The controller can generate operation control signals according to the instruction operation code and timing signals to complete the control of fetching and executing instructions.

[0172] A memory may also be provided in the processor 1410 for storing instructions and data. In some embodiments, the memory in the processor 1410 is a cache memory. This memory can save the instructions or data that the processor 1410 has just used or recycled. If the processor 1410 needs to use the instruction or data again, it can be called from the memory. This avoids repeated accesses, reduces the waiting time of the processor 1410, and thus improves the efficiency of the system.

[0173] In the embodiments of the present application, the processor 1410 can be used to obtain PPG data segments and send the PPG data segments to the cloud server. The cloud server divides the PPG data to obtain multiple PPG data segments, and based on each PPG data segment, determines the initial physiological signal value and probability density distribution data corresponding to each PPG data segment; according to the matching relationship between the initial physiological signal value corresponding to each PPG data segment and the probability density distribution data, determines the effective initial physiological signal value from multiple initial physiological signal values; after determining the physiological signal measurement result based on the effective initial physiological signal value, the processor displays the physiological signal measurement result according to the physiological signal measurement result sent by the cloud server.

[0174] The wireless communication function of the electronic device can be implemented through antenna 1, antenna 2, mobile communication module 1420, wireless communication module 1421, modem processor, and baseband processor, etc.

[0175] Antenna 1 and Antenna 2 are used for transmitting and receiving electromagnetic wave signals. The antennas in the electronic device can be used to cover single or multiple communication frequency bands. Different antennas can also be multiplexed to improve the utilization rate of the antennas. For example, Antenna 1 can be multiplexed as the diversity antenna of the wireless local area network. In some other embodiments, the antenna can be used in combination with a tuning switch.

[0176] The mobile communication module 1420 can provide solutions for wireless communications including 2G / 3G / 4G / 5G, etc. applied to the electronic device. The mobile communication module 1420 can include at least one filter, switch, power amplifier, low noise amplifier (LNA), etc. The mobile communication module 1420 can receive electromagnetic waves through Antenna 1, perform filtering, amplification and other processing on the received electromagnetic waves, and transmit them to the modulation and demodulation processor for demodulation.

[0177] The wireless communication module 1421 can provide solutions for wireless communications including wireless local area networks (WLAN) (such as wireless fidelity (Wi-Fi) networks), Bluetooth (BT), global navigation satellite system (GNSS), frequency modulation (FM), near field communication (NFC), infrared technology (IR), etc. applied to the electronic device.

[0178] The wireless communication module 1421 can be one or more devices integrating at least one communication processing module. The wireless communication module 1421 receives electromagnetic waves through Antenna 2, performs frequency modulation and filtering processing on the electromagnetic wave signals, and sends the processed signals to the processor 1410. The wireless communication module 1421 can also receive the signals to be transmitted from the processor 1410, perform frequency modulation and amplification on them, and convert them into electromagnetic waves through Antenna 2 for radiation.

[0179] In some embodiments, antenna 1 of the electronic device is coupled to the mobile communication module 1420, and antenna 2 is coupled to the wireless communication module 1421, enabling the electronic device to communicate with the network and other devices through wireless communication technologies. The wireless communication technologies may include global system for mobile communications (GSM), general packet radio service (GPRS), code division multiple access (CDMA), wideband code division multiple access (WCDMA), time-division code division multiple access (TD-SCDMA), long term evolution (LTE), BT, GNSS, WLAN, NFC, FM, and / or IR technologies, etc. GNSS may include global positioning system (GPS), global navigation satellite system (GLONASS), beidou navigation satellite system (BDS), quasi-zenith satellite system (QZSS), and / or satellite based augmentation systems (SBAS).

[0180] Based on this, in the embodiments of the present application, the electronic device can communicate with the cloud server. After obtaining the PPG data based on the sensor, it can send the PPG data to the cloud server and receive the physiological signal measurement results sent by the cloud server. After obtaining the physiological signal measurement results, it can display the physiological signal measurement results on the display screen 1450 through the GPU, the display screen 1450, and the application processor, etc.

[0181] Exemplarily, the GPU is a microprocessor for image processing, connected to the display screen 1450 and the application processor. The GPU is used to perform mathematical and geometric calculations for graphics rendering. The processor 1410 may include one or more GPUs, which execute program instructions to generate or change the display information.

[0182] The display screen 1450 includes a display panel for displaying images, videos, etc.

[0183] The display panel can be a liquid crystal display (LCD), an organic light-emitting diode (OLED), an active-matrix organic light-emitting diode (AMOLED), a flexible light-emitting diode (FLED), a MiniLED, a MicroLED, a Micro-OLED, a quantum dot light-emitting diode (QLED), etc.

[0184] In some embodiments, the electronic device may include one or N display screens 1450, where N is a positive integer greater than 1.

[0185] The software system of the electronic device can adopt a layered architecture, an event-driven architecture, a microkernel architecture, a microservices architecture, or a cloud architecture, etc. In this embodiment of the application, the Android system with a layered architecture is taken as an example to exemplarily illustrate the software structure of the electronic device.

[0186] Figure 15 This is a software structure block diagram of an electronic device provided in an embodiment of this application.

[0187] The layered architecture divides the software into several layers, and each layer has a clear role and division of labor. The layers communicate with each other through software interfaces. In some embodiments, the Android system is divided into four layers, from top to bottom, namely the application layer, the application framework layer, the Android runtime and system libraries, and the kernel layer.

[0188] The application layer may include a series of application packages.

[0189] Such as Figure 15 shown, the application packages may include applications such as blood pressure measurement, heart rate measurement, blood oxygen saturation measurement, etc.

[0190] The application framework layer provides application programming interfaces (APIs) and programming frameworks for the applications in the application layer. The application framework layer includes some predefined functions. Such as Figure 15 shown, the application framework layer may include a view system, a resource manager, a notification manager, etc.

[0191] The view system includes visual controls, such as controls for displaying text, controls for displaying pictures, etc. The view system can be used to build applications. The display interface can be composed of one or more views. For example, a display interface including a text message notification icon can include a view for displaying text and a view for displaying pictures.

[0192] The resource manager provides various resources for the application, such as localized strings, icons, pictures, layout files, video files, and so on.

[0193] The notification manager enables the application to display notification information in the status bar. It can be used to convey informative messages, which can automatically disappear after a short stay without user interaction. For example, the notification manager is used to inform that the measurement is completed and to prompt the user that a physiological signal test is in progress, etc.

[0194] Android Runtime includes core libraries and a virtual machine. Android runtime is responsible for the scheduling and management of the Android system.

[0195] The core libraries consist of two parts: one part is the functional functions that need to be called by the Java language, and the other part is the core libraries of Android.

[0196] The application layer and the application framework layer run in the virtual machine. The virtual machine executes the Java files of the application layer and the application framework layer as binary files. The virtual machine is used to perform functions such as object lifecycle management, stack management, thread management, security and exception management, and garbage collection.

[0197] The system libraries can include multiple functional modules. For example, a 3D graphics processing library (e.g., OpenGL ES), a 2D graphics engine (e.g., SGL), etc.

[0198] The 3D graphics processing library is used to implement 3D graphics drawing, image rendering, composition, and layer processing, etc.

[0199] The 2D graphics engine is a drawing engine for 2D drawing.

[0200] The kernel layer is the layer between the hardware and the software. The kernel layer at least includes a display driver, a sensor driver, etc.

[0201] It should be noted that although the embodiments of this application are described by taking the Android system as an example, its basic principles also apply to electronic devices based on operating systems such as iOS and Windows.

[0202] This embodiment also provides a computer-readable storage medium, which includes instructions. When the above instructions run on an electronic device, the electronic device is caused to execute the above-related method steps to implement the method in the above embodiment.

[0203] As described above, the foregoing is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application should be covered by the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method for measuring physiological signals, characterized in that, Including: Based on each photoplethysmogram (PPG) data segment, determining an initial physiological signal value and probability density distribution data corresponding to each PPG data segment; each of the PPG data segments is obtained by partitioning the PPG data, and the probability density distribution data is used to represent the probabilities corresponding to different initial physiological signal values; According to the matching relationship between the initial physiological signal value and the probability density distribution data corresponding to each PPG data segment, determining an effective initial physiological signal value from multiple initial physiological signal values; the matching relationship indicates the probability interval and / or probability corresponding to the initial physiological signal value corresponding to the PPG data segment in the probability density distribution data; Based on the effective initial physiological signal value, determining a physiological signal measurement result.

2. The method according to claim 1, characterized in that, The step of determining an effective initial physiological signal value from multiple initial physiological signal values according to the matching relationship between the initial physiological signal value and the probability density distribution data corresponding to each PPG data segment includes: According to the matching relationship between each initial physiological signal value corresponding to each PPG data segment and the probability density distribution data, determining the probability interval corresponding to each initial physiological signal value; When the probability interval corresponding to an initial physiological signal value is a first interval or a second interval, determining the initial physiological signal value as an effective initial physiological signal value, and when the probability interval corresponding to an initial physiological signal value is a third interval, determining the initial physiological signal value as an ineffective initial physiological signal value; the integral value of the probability density distribution data in the first interval is the highest, the integral value in the second interval is less than the integral value in the first interval and greater than the integral value in the third interval.

3. The method according to claim 2, wherein All of the effective initial physiological signal values correspond to the second interval. The step of determining a physiological signal measurement result based on the effective initial physiological signal value includes: Based on the probability density distribution data, determining the probability of each effective initial physiological signal value; Determining the product of each effective initial physiological signal value and the probability of each effective initial physiological signal value as the weighted initial physiological signal value corresponding to each effective initial physiological signal value; Based on the mean value of the weighted initial physiological signal values, determining a physiological signal measurement result.

4. The method according to claim 2, wherein The effective initial physiological signal values include effective initial physiological signal values corresponding to the first interval and effective initial physiological signal values corresponding to the second interval. The step of determining a physiological signal measurement result based on the effective initial physiological signal value includes: For the effective initial physiological signal values corresponding to the second interval, determining the product of each effective initial physiological signal value and the probability of the effective initial physiological signal value as the weighted initial physiological signal value corresponding to each effective initial physiological signal value; the probability of the effective initial physiological signal value is determined based on the probability density distribution data; Based on the weighted initial physiological signal values of the effective initial physiological signal values corresponding to the second interval and the mean value of the effective initial physiological signal values corresponding to the first interval, determining a physiological signal measurement result.

5. The method according to claim 1, wherein The step of determining an effective initial physiological signal value from multiple initial physiological signal values according to the matching relationship between the initial physiological signal value and the probability density distribution data corresponding to each PPG data segment includes: Determine the probability of the initial physiological signal value according to the matching relationship between each initial physiological signal value corresponding to each PPG data segment and the probability density distribution data; When the probability of an initial physiological signal value is greater than the first threshold, determine the initial physiological signal value as a valid initial physiological signal value; when the probability of an initial physiological signal value is less than or equal to the first threshold, determine the initial physiological signal value as an invalid initial physiological signal value.

6. The method according to any one of claims 1-5, characterized in that, The determining of the initial physiological signal value and the probability density distribution data corresponding to each PPG data segment based on each photoplethysmography (PPG) data segment includes: Based on a regression model, perform numerical regression of the physiological signal on each PPG data segment to determine the initial physiological signal value corresponding to each PPG data segment; Based on a classification model, perform classification of the physiological signal on each PPG data segment to determine the probability density distribution data corresponding to each PPG data segment; the regression model and the classification model are trained based on training set data; The generation process of the training set data includes: Obtain PPG data and standard data collected in multiple time periods; Divide the PPG data collected in each time period into a first data segment and a second data segment; Map the first data segment of the Nth time period to the standard data of the (N - 1)th time period, and map the second data segment of the Nth time period to the standard data of the Nth time period to obtain training set data, where N is a positive integer greater than 0.

7. The method according to claim 6, characterized in that, The probability density distribution data follows a Gaussian distribution, and the training steps of the classification model include: Use the training set data as input and output the probability density distribution data corresponding to the training set data, and the mean of the probability density distribution data is determined based on the standard data in the training set data.

8. The method according to any one of claims 1-5, characterized in that, When the physiological signal is blood pressure, the initial physiological signal value includes an initial systolic blood pressure value and an initial diastolic blood pressure value, the probability density distribution data includes first distribution data corresponding to the initial systolic blood pressure value and second distribution data corresponding to the initial diastolic blood pressure value, and the determining of the valid initial physiological signal value from multiple initial physiological signal values according to the matching relationship between the initial physiological signal value corresponding to each PPG data segment and the probability density distribution data includes: Determine the valid initial systolic blood pressure value from multiple initial systolic blood pressure values according to the initial systolic blood pressure value corresponding to each PPG data segment and the first distribution data, and determine the valid initial diastolic blood pressure value from multiple initial diastolic blood pressure values according to the initial diastolic blood pressure value corresponding to each PPG data segment and the second distribution data; The determining of the physiological signal measurement result based on the valid initial physiological signal value includes: Determine the mean of the valid initial systolic blood pressure values as the systolic blood pressure value, and determine the valid initial diastolic blood pressure value as the diastolic blood pressure value; Based on the systolic blood pressure value and the diastolic blood pressure value, determine the physiological signal measurement result.

9. The method according to any one of claims 1-5, characterized in that, Before the determining of the initial physiological signal value and the probability density distribution data corresponding to each PPG data segment based on each PPG data segment, it further includes: Receive PPG data sent by a wearable device; After determining the physiological signal measurement result based on the initial value of the effective physiological signal, the method further includes: Sending the physiological signal measurement result to the wearable device to display the physiological signal measurement result on the display interface of the wearable device.

10. The method according to any one of claims 1 to 5, wherein Determining the physiological signal measurement result based on the initial value of the effective physiological signal includes: When there are multiple initial values of effective physiological signals, determining the mean value of the multiple initial values of effective physiological signals as the physiological signal value; Based on the physiological signal value, determining the level corresponding to the physiological signal value; Determining the physiological signal value and / or the level as the physiological signal measurement result.

11. An electronic device, characterized in that, The electronic device includes: a processor and a memory; The memory is used to store program code and transmit the program code to the processor; The processor is configured to execute the steps of a physiological signal measurement method according to any one of claims 1-10 based on the instructions in the program code.

Images