Methods for measuring blood pressure without a cuff, methods for calibrating blood pressure without a cuff, and related equipment

Abstract

This application provides a cuffless blood pressure measurement method, a cuffless blood pressure calibration method, and related equipment. The cuffless blood pressure measurement method includes: receiving a blood pressure measurement request sent by a user; measuring the user's physiological signals based on the blood pressure measurement request, the physiological signals including PPG signals and measurement environment signals; determining the user's corresponding calibration level based on historical calibration records, with different calibration levels having different correction degrees for blood pressure values; and calculating the user's corresponding blood pressure value based on the user's corresponding calibration level and the physiological signals. In this method, different calibration levels correspond to different numbers of calibrations performed by the user. The more times the user calibrates, the greater the influence of the calibration data on the final value. This method can improve the reliability of calibration, allowing the calibration data to be better utilized in the blood pressure value calculation process, thereby increasing the accuracy of cuffless blood pressure measurement.

Description

Technical Field

[0001] This application belongs to the field of signal measurement, and in particular relates to a cuffless blood pressure measurement method, a cuffless blood pressure calibration method, and related equipment. Background Technology

[0002] With societal development and increasingly affluent lifestyles, the excessive intake of nutrients in daily diets has led to a growing threat of hypertension, making the need for rapid and accurate blood pressure measurement methods and equipment increasingly urgent. Immediate and accurate blood pressure measurement methods and equipment are both essential and meaningful.

[0003] Existing cuffless blood pressure measurement methods are generally based on PPG signals. Photoplethysmography (PPG) is a non-invasive method that uses photoelectric means to detect changes in blood volume in living tissue. At the start of the measurement, the wristband emits a photoelectric signal towards the skin. A sensor collects the pulse wave carried by the reflected light, and then the blood pressure value is estimated using a specific calculation formula. PPG measurement is highly dependent on the acquired signal waveform, and due to the complexity of the testing environment, the results of non-invasive blood pressure measurements are quite unstable. To address this issue, existing cuffless blood pressure measurement methods have generally added calibration functions.

[0004] Most existing calibration methods record the calibration values ​​input by the user (which are generally measured using precise equipment such as mercury sphygmomanometers). During actual measurement, the measured values ​​are input into the model to obtain the predicted blood pressure value. Different weights are assigned to the predicted value and the calibration value, and the final output value is obtained by weighted calculation. However, this approach has the problem that the calibration value has too constant an influence on the final measurement value, which is not flexible enough in actual testing. Furthermore, a single calibration may result in the calibration data itself being inaccurate, failing to achieve the calibration purpose. Summary of the Invention

[0005] The purpose of this invention is to provide a cuffless blood pressure measurement method and calibration method, aiming to solve the problems of insufficient accuracy and low efficiency in utilizing calibration data in existing calibration methods. The cuffless blood pressure measurement method provided in this application includes: The first aspect of this application provides a blood pressure measurement method, including: Receive blood pressure measurement requests sent by users; Based on the blood pressure measurement request, the current user's physiological signals are measured, including: PPG signal and measurement environment signal; The calibration level of the user is determined based on historical calibration records, and different calibration levels have different degrees of correction for blood pressure values; The blood pressure value of the user is calculated based on the user's corresponding calibration level and the physiological signals.

[0006] Based on the cuffless blood pressure measurement method provided in the first aspect of the embodiments of this application, optionally, the method further includes: Receive blood pressure calibration requests sent by users; Based on the blood pressure calibration request, the calibration physiological signals of the current user are measured, including: PPG signal and measurement environment signal; Calculate the predicted blood pressure value based on the calibrated physiological signal and obtain the calibrated blood pressure value; Determine whether the difference between the calibrated blood pressure value and the predicted blood pressure value is greater than a preset range; If the difference between the calibrated blood pressure value and the predicted blood pressure value is less than or equal to the preset range, then the calibrated blood pressure and the physiological signal are stored as historical calibration records.

[0007] Based on the cuffless blood pressure measurement method provided in the first aspect of the embodiments of this application, optionally, the step of determining the calibration level corresponding to the user based on historical calibration records includes: If the historical calibration record does not exist, the calibration level is determined to be zero; If the number of historical calibration records is less than or equal to the first preset value, then the calibration level is determined to be one; If the number of historical calibration records is greater than a first preset value and less than or equal to a second preset value, then the calibration level is determined to be two. If the number of historical calibration records is greater than the second preset value, then the calibration level is determined to be three.

[0008] Based on the cuffless blood pressure measurement method provided in the first aspect of the embodiments of this application, optionally, the step of calculating the blood pressure value corresponding to the user based on the user's corresponding calibration level and the physiological signal includes: If the calibration level corresponding to the user is zero, the physiological signal is input into the initial calculation model to obtain the blood pressure value corresponding to the user. Based on the cuffless blood pressure measurement method provided in the first aspect of the embodiments of this application, optionally, the step of calculating the blood pressure value corresponding to the user based on the user's corresponding calibration level and the physiological signal includes: If the user's calibration level is one, the physiological signal is input into the initial calculation model to obtain an initial reference value. Based on the initial reference value and the historical calibration value, a weighted calculation is performed to obtain the user's blood pressure value.

[0009] Based on the cuffless blood pressure measurement method provided in the first aspect of the embodiments of this application, optionally, the step of calculating the blood pressure value corresponding to the user based on the user's corresponding calibration level and the physiological signal includes: If the user's corresponding calibration level is two, then the initial calculation model is updated based on the historical calibration value to obtain the first updated model; The physiological signal is processed based on the first update model to obtain a first reference value; The blood pressure value corresponding to the user is obtained by weighted calculation based on the first reference value and the historical calibration value.

[0010] Based on the cuffless blood pressure measurement method provided in the first aspect of the embodiments of this application, optionally, the step of calculating the blood pressure value corresponding to the user based on the user's corresponding calibration level and the physiological signal includes: If the user's corresponding calibration level is three, then the initial calculation model is updated based on the historical calibration value to obtain the second updated model; The physiological signals are processed based on the second update model to obtain the blood pressure value corresponding to the user.

[0011] A second aspect of this application provides a cuffless blood pressure calibration method, including: Receive blood pressure calibration requests sent by users; Based on the blood pressure calibration request, the calibration physiological signals of the current user are measured, including: PPG signal and measurement environment signal; Calculate the predicted blood pressure value based on the calibrated physiological signal and obtain the calibrated blood pressure value; Determine whether the difference between the calibrated blood pressure value and the predicted blood pressure value is greater than a preset range; If the difference between the calibrated blood pressure value and the predicted blood pressure value is less than or equal to the preset range, then the calibrated blood pressure and the physiological signal are stored as historical calibration records.

[0012] The calibration level of the user is determined based on historical calibration records, and different calibration levels have different degrees of correction for blood pressure measurements.

[0013] A third aspect of this application provides a cuffless blood pressure measurement device, comprising: Central processing unit, memory, input / output interfaces, wired or wireless network interfaces, and power supply; The memory is either a short-term storage memory or a persistent storage memory; The central processing unit is configured to communicate with the memory and execute instructions in the memory on the cuffless blood pressure measuring device to perform the method described in any one of the first aspects of the embodiments of this application.

[0014] A fourth aspect of this application provides a computer-readable storage medium including instructions that, when executed on a computer, cause the computer to perform the method as described in any one of the first aspects of this application.

[0015] The fifth aspect of this application provides a computer program product containing instructions that, when run on a computer, cause the computer to perform the method described in any one of the first aspects of this application.

[0016] As can be seen from the above technical solutions, the embodiments of this application have the following advantages: This application provides a cuffless blood pressure calibration method, including: receiving a blood pressure calibration request sent by a user; measuring the calibration physiological signal of the current user based on the blood pressure calibration request, the calibration physiological signal including: PPG signal and measurement environment signal; calculating a predicted blood pressure value based on the calibration physiological signal and obtaining a calibrated blood pressure value; determining whether the difference between the calibrated blood pressure value and the predicted blood pressure value is greater than a preset range; if the difference between the calibrated blood pressure value and the predicted blood pressure value is less than or equal to the preset range, then storing the calibrated blood pressure and the physiological signal as historical calibration records; determining the user's calibration level based on the historical calibration records; and different calibration levels have different correction degrees for the target user's blood pressure value. In this method, different calibration levels correspond to different numbers of calibrations performed by the user. The more times the user calibrates, the greater the influence of the calibration data on the final value. This method can improve the reliability of calibration, allowing the calibration data to be better utilized in the blood pressure value calculation process, thereby increasing the accuracy of cuffless blood pressure measurement. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort. It is understood that the drawings provided in this section are only for better understanding of this solution and do not constitute a limitation on this application.

[0018] Figure 1 This is a schematic flowchart of an embodiment of the cuffless blood pressure measurement method provided in this application.

[0019] Figure 2 This is a schematic flowchart of an embodiment of the cuffless blood pressure calibration method provided in this application.

[0020] Figure 3 This is a schematic diagram of the acquisition unit used in the cuffless blood pressure measurement method of this application.

[0021] Figure 4 This is another schematic flowchart illustrating an embodiment of the cuffless blood pressure measurement method provided in this application.

[0022] Figure 5 This is a schematic diagram of the blood pressure calculation model provided in this application.

[0023] Figure 6 This is a flowchart illustrating the data synchronization method during the cuffless blood pressure measurement and calibration process provided in this application. Figure 7 A schematic diagram of a cuffless blood pressure measurement device embodiment provided in this application; Figure 8 This is a schematic diagram of a structure of an embodiment of the cuffless blood pressure calibration device provided in this application; Figure 9 Another structural schematic diagram of an embodiment of the cuffless blood pressure measuring device provided in this application; Figure 10 This is another structural schematic diagram of an embodiment of the cuffless blood pressure calibration device provided in this application. Detailed Implementation

[0024] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application are clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application. At the same time, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0025] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0026] With societal development and increasingly affluent lifestyles, the excessive intake of nutrients in daily diets has led to a growing threat of hypertension, making the need for rapid and accurate blood pressure measurement methods and equipment increasingly urgent. Immediate and accurate blood pressure measurement methods and equipment are both essential and meaningful.

[0027] Existing cuffless blood pressure measurement methods are generally based on PPG signals. Photoplethysmography (PPG) is a non-invasive method that uses photoelectric means to detect changes in blood volume in living tissue. At the start of the measurement, the wristband emits a photoelectric signal towards the skin. A sensor collects the pulse wave carried by the reflected light, and then the blood pressure value is estimated using a specific calculation formula. PPG measurement is highly dependent on the acquired signal waveform, and due to the complexity of the testing environment, the results of non-invasive blood pressure measurements are quite unstable. To address this issue, existing cuffless blood pressure measurement methods have generally added calibration functions.

[0028] Existing calibration methods mostly involve recording the calibration values ​​input by the user (generally measured using precise equipment such as mercury sphygmomanometers). During actual measurement, the measured values ​​are input into a model to derive a predicted blood pressure value. Different weights are assigned to the predicted and calibrated values, and a weighted calculation is performed to obtain the final output value. However, this approach suffers from several drawbacks: the calibrated values ​​have an overly constant influence on the final measurement, making it inflexible in practical applications. Furthermore, single calibrations can result in inaccurate calibration data, failing to achieve the calibration objective. To address these issues, please refer to [link to relevant documentation]. Figure 1 An embodiment of the cuffless blood pressure calibration method provided in this application includes steps 101-104.

[0029] 101. Receive blood pressure measurement requests sent by users.

[0030] Specifically, this solution receives blood pressure measurement requests from users. In actual implementation, this solution is applied to cuffless blood pressure measurement devices, especially those based on PPG signals. In actual implementation, the device starts working when the user triggers a button.

[0031] 102. Measure the current user's physiological signals based on the blood pressure measurement request.

[0032] Based on the blood pressure measurement request, the current user's physiological signals are measured, including PPG signals and measurement environment signals.

[0033] PPG signal acquisition is typically performed using a photoelectric sensor, placed on the target user's skin, usually on a finger, earlobe, or wrist. The photoelectric sensor emits light and measures its absorption, thus capturing the pulse waveform on the skin surface. This pulse waveform signal is the PPG signal, and its changes during a heartbeat reflect changes in blood flow. In practice, the PPG signal can be generated from various wavelengths of light, such as red or infrared light, depending on the specific circumstances. In this solution, a cuff is not required, ensuring the convenience of blood pressure measurement.

[0034] The environmental measurement signal is the signal generated during the acquisition of the target user's PPG signal. This environmental measurement signal is acquired synchronously with the PPG signal acquisition process, maintaining consistency in acquisition time. To ensure the correlation between physiological signals and PPG signals, the environmental measurement signal acquisition device can be placed near the PPG signal location to ensure that the blood pressure information reflected by both is as similar as possible. The environmental measurement signal includes signals such as temperature, pressure, and humidity. When the photoelectric sensor acquires the PPG signal, the waveform signal it captures may also be affected by other physiological factors. For example, under the same testing conditions, a higher temperature at the testing site will cause vasodilation, resulting in a stronger acquired signal, while a lower temperature will cause vasoconstriction, resulting in a weaker acquired signal. Meanwhile, parameters such as pressure and humidity may also affect the measurement results of PPG signals, leading to deviations in the final calculated blood pressure value. Therefore, it is necessary to measure and acquire parameters such as pressure and humidity during measurement and participate in the subsequent data processing. Understandably, the types of environmental signals measured in the actual implementation process can be determined according to the actual situation, that is, based on the model used in the subsequent data processing. The types of environmental signals collected should be consistent with the data used to train the model. No specific restrictions are made here.

[0035] 103. Determine the calibration level corresponding to the user based on historical calibration records.

[0036] Specifically, the calibration level for the user is determined based on historical calibration records. The user's historical calibration records are queried; these records document the user's past calibrations, ranging from zero to any number. These records are generated based on the user's past calibrations. For example, the user can use a cuff-type blood pressure monitor or a mercury sphygmomanometer to measure accurate blood pressure information and use the device's data acquisition unit to collect physiological information. The measured accurate blood pressure information is then input into the device to complete the calibration process. The specific calibration process can be determined based on actual circumstances and is not limited here. Different numbers of historical calibration records determine the user's corresponding calibration level, and different calibration levels have different degrees of correction for blood pressure values. If no calibration has ever been performed, there are no historical calibration records, and the calibration level is zero, with no correction for blood pressure values. The more historical calibration records, the higher the calibration level, and the more the historical calibration parameters participate in the subsequent blood pressure value calculation process, thus maximizing the effectiveness of the calibration data. The specific calibration level determination can be determined based on actual circumstances and is not limited here.

[0037] 104. The blood pressure value corresponding to the user is calculated based on the user's corresponding calibration level and the physiological signal.

[0038] Specifically, the user's blood pressure value is calculated based on the user's calibration level and the physiological signal. For example, if the calibration level is zero, it means the user has never been calibrated, and historical data is not needed; the existing blood pressure calculation model can be used directly to process the physiological signal. If the calibration level is one, the possible calculation method is to use the blood pressure calculation model to obtain a preliminary value and assign corresponding weights, then assign certain weights to the blood pressure values ​​in the calibration historical data, and finally calculate the weighted average to obtain the final value. If the calibration level is two, the weight ratios may differ. If the calibration level is higher, the calibration historical data will affect the blood pressure calculation model used. The specific calculation method can be determined according to the actual situation and is not limited here.

[0039] As can be seen from the above description, this application provides a cuffless blood pressure measurement method, including: receiving a blood pressure measurement request sent by a user; measuring the current user's physiological signals based on the blood pressure measurement request, the physiological signals including: PPG signal and measurement environment signal; determining the calibration level corresponding to the user based on historical calibration records, wherein different calibration levels have different correction degrees for blood pressure values; and calculating the blood pressure value corresponding to the user based on the user's calibration level and the physiological signals. In this method, different calibration levels correspond to different numbers of calibrations performed by the user. The more times the user calibrates, the greater the influence of the calibration data on the final value. This method can improve the reliability of calibration, allowing the calibration data to be better utilized in the blood pressure value calculation process, thereby increasing the accuracy of cuffless blood pressure measurement.

[0040] In practical implementation, to ensure the accuracy of blood pressure measurements for users, this application also provides a blood pressure calibration method. Please refer to [link / reference]. Figure 2 The blood pressure calibration method provided in this application includes steps 201 to 205.

[0041] 201. Receive blood pressure calibration request sent by the user.

[0042] Specifically, the application environment of this solution may include mobile terminals (users' mobile phones, tablets, personal computers, etc., used to run the blood pressure measurement APP), measurement sensors (used to collect PPG signals and measurement environment signals), and cloud servers (used to support data storage, synchronization, and other tasks). At the start of calibration, the user initiates the calibration through the terminal APP. The APP will then provide the user with information such as the calibration completion rate, the current calibration level, and which scenarios have not yet been calibrated. It will also display a diagram of the measurement actions for the corresponding scenario and provide prompts on how to achieve the corresponding measurement scenario. Specific measurement scenarios can be set based on different environmental factors, with the aim of enriching the types of data collected as much as possible, thereby ensuring the richness, stability, and usability of the collected calibration data.

[0043] Specifically, the application can prompt the user to perform five calibrations when waking up in the morning, five calibrations when going to sleep in the evening, five calibrations with light pressure on the sensor, and five calibrations with heavier pressure on the sensor. The scenarios include: morning, night, light pressure, heavy pressure, etc. The specific scenario settings can be determined according to the actual situation and are not limited here.

[0044] 202. Measure the current user's calibration physiological signal based on the blood pressure calibration request.

[0045] Specifically, the acquisition unit used in this solution can be as follows: Figure 3As shown, it includes three pressure sensors 311 to 313, a PPG sensor 321, a temperature sensor 331, and a humidity sensor 341; The data collection unit is used to collect various data from the user's fingertips. However, it is understood that the data collection site can also be the finger joints, wrist, earlobe, etc., and no specific restrictions are made here.

[0046] The acquisition unit includes a PPG sensor, a temperature sensor, a pressure sensor, and a humidity sensor; multiple pressure sensors are evenly arranged outside the PPG sensor; the temperature sensor and the humidity sensor are arranged close to the PPG sensor.

[0047] In this embodiment, three pressure sensors are evenly arranged on the outside of the PPG sensor 321. These three sensors are used to measure the contact condition of the test site, determine whether the test site is in complete contact, and the contact force and distribution. The PPG sensor 321 is responsible for collecting PPG sensor signals from the test site. It can be understood that the PPG sensor signals include, but are not limited to, red light and infrared wavelengths, and may also include other wavelength signals such as green light and yellow light. The number of channels for acquiring each wavelength signal is not limited to one. The temperature sensor 331 is a humidity sensor used to detect the moisture level of the skin at the test site. This feature has a significant impact on the skin penetration characteristics of the PPG signal. The humidity sensor 341 is used to detect the skin temperature at the test site. This feature has a significant impact on some waveform characteristics of the signal. For example, under the same test conditions, a higher temperature at the test site leads to vasodilation and a stronger signal, while a lower temperature leads to vasoconstriction and a weaker signal. Optionally, the measured environmental signals include any one or more of the following: pressure signal, temperature signal, humidity signal, acceleration signal, and electrocardiogram signal. This embodiment uses pressure signal, temperature signal, and humidity signal as an example for explanation.

[0048] Once the user is ready, place the measurement area (finger tip) as prompted by the app. Figure 3 On the structure shown, the device will collect data from various pressure sensors, temperature sensors, and humidity sensors in real time. At the same time, the application can also determine the actual collection scenario of the user based on the collected data and send the collection scenario to the APP in real time, which will be displayed to the user. The user can choose to adjust or not adjust according to the prompts on the APP, depending on the actual situation, and there is no limit here.

[0049] During the data acquisition process, the temperature, humidity, pressure values, and PPG signal under the current measurement period are extracted and collected for a certain period of time (usually 30 seconds or 1 minute) to support subsequent data analysis.

[0050] 203. Calculate the predicted blood pressure value based on the calibrated physiological signal and obtain the calibrated blood pressure value.

[0051] Specifically, based on the calibrated physiological signals, the predicted blood pressure value is calculated and the calibrated blood pressure value is obtained. Feature extraction is performed on the measured blood pressure parameters, which generally include: waveform features and fusion features of PPG waveform, VPPG waveform (first-order difference of PPG signal), and APPG waveform (second-order difference of PPG signal). Examples include the sampling interval between sampling points with normalized amplitudes of 10%, 25%, 50%, 75%, and 90% before the PPG signal value point and the peak point; the sampling interval between sampling points with normalized amplitudes of 10%, 25%, 50%, 75%, and 90% after the peak point and the peak point; the number of continuous sampling points during the heartbeat cycle; the interval between the maximum value sampling point in the first-order difference of the PPG signal and the peak point; or the area of ​​the graphic formed by the PGG waveform feature points and the time axis; the slope and amplitude ratio of the lines connecting some feature points; and the amplitude ratio and amplitude baseline ratio formed by feature points corresponding to different wavelength signals. For environmental signals, this includes skin temperature, humidity, and three average pressure values. In actual implementation, the type of features collected depends on the blood pressure measurement model used. The same type of data is collected based on the features used to train the blood pressure measurement model. The collected data is then input into the blood pressure measurement model for calculation and processing to obtain the predicted blood pressure value. The model used for calculating the predicted blood pressure value is the initial model, which is a model pre-trained by the staff. The specific model type and parameter settings can be determined according to the actual situation and are not limited here.

[0052] Once the data collection time reaches the preset value under calibration conditions, a window will pop up on the APP allowing the user to input the calibrated blood pressure value. The calibrated blood pressure value is the blood pressure information obtained by the user based on other more reliable devices (such as mercury sphygmomanometers or cuffed sphygmomanometers). The purpose of inputting the calibrated blood pressure value is generally to correct the predicted blood pressure value as close as possible to the calibrated value, thereby improving the accuracy of the blood pressure value obtained without a cuff.

[0053] 204. Determine whether the difference between the calibrated blood pressure value and the predicted blood pressure value is greater than a preset range.

[0054] Specifically, it is determined whether the difference between the calibrated blood pressure value and the predicted blood pressure value is greater than a preset range. In actual implementation, if the difference between the calibrated blood pressure value and the predicted blood pressure value is greater than the preset range, it may be that the blood pressure value entered by the user is incorrect, such as the user mistakenly entering a certain number, causing the two to deviate too much. In this case, the calibrated blood pressure value should not be recorded, and the user should be prompted to re-enter it. In addition, it is also possible to determine (1) whether the percentage of time during which the signal quality meets the requirements is greater than the threshold; (2) whether there is a switch in the measurement scenario, so as to ensure that the obtained signal is stable and the quality meets the preset requirements. It is understandable that in actual implementation, further quality judgment can be made on the calibrated blood pressure value and the predicted blood pressure value to ensure their accuracy and stability.

[0055] 205. Store the calibrated blood pressure and the physiological signals as historical calibration records.

[0056] Specifically, if the difference between the calibrated blood pressure value and the predicted blood pressure value is less than or equal to a preset range, the calibrated blood pressure and the physiological signal are stored as historical calibration records to support their use in subsequent blood pressure measurement processes.

[0057] Understandably, after storage, data can be synchronized with related devices via cloud servers to ensure data consistency.

[0058] After storing historical calibration records, the corresponding calibration level can be determined based on the user's historical calibration records, and different rules can be set for the different effects of historical calibration records on the measured values ​​under different calibration levels, so as to facilitate direct use in subsequent blood pressure measurement processes.

[0059] It is worth noting that this embodiment can be implemented independently as a separate calibration process, or it can be combined with the above-mentioned... Figure 1 The corresponding embodiments are used together to form a complete blood pressure calibration and measurement process, and the specifics are not limited here.

[0060] As can be seen from the above technical solutions, the embodiments of this application have the following advantages: This application provides a cuffless blood pressure calibration method, including: receiving a blood pressure calibration request sent by a user; measuring the calibration physiological signal of the current user based on the blood pressure calibration request, the calibration physiological signal including: PPG signal and measurement environment signal; calculating a predicted blood pressure value based on the calibration physiological signal and obtaining a calibrated blood pressure value; determining whether the difference between the calibrated blood pressure value and the predicted blood pressure value is greater than a preset range; if the difference between the calibrated blood pressure value and the predicted blood pressure value is less than or equal to the preset range, then storing the calibrated blood pressure and the physiological signal as historical calibration records; determining the user's calibration level based on the historical calibration records; and different calibration levels have different correction degrees for the target user's blood pressure value. In this method, different calibration levels correspond to different numbers of calibrations performed by the user. The more times the user calibrates, the greater the influence of the calibration data on the final value. This method can improve the reliability of calibration, allowing the calibration data to be better utilized in the blood pressure value calculation process, thereby increasing the accuracy of cuffless blood pressure measurement.

[0061] 206. Determine the calibration level corresponding to the user based on historical calibration records.

[0062] Specifically, this embodiment uses five calibration levels as an example for explanation: uncalibrated, calibration level 1, calibration level 2, calibration level 3, and fully calibrated.

[0063] If the number of scenarios successfully calibrated by the user does not exceed the threshold If so, the calibration level is determined to be calibration level 1.

[0064] If the number of user-defined completed scenarios exceeds the threshold Less than the threshold If so, the user is in calibration level 2 state.

[0065] If the number of user-defined completed scenarios exceeds the threshold Less than the threshold If so, the user is in calibration level 3 state.

[0066] If the user completes all calibration scenarios, the user is now in a fully calibrated state.

[0067] Specifically, different rules can be applied to the use of calibration data for different calibration levels. Generally speaking, the higher the calibration level, the greater the involvement of calibration data in the determination of blood pressure values; that is, the higher the calibration level, the more correlated the final measurement output value is with the calibration data. Specific rules can be found in [reference needed]. Figure 4 Contents in the corresponding embodiments.

[0068] As can be seen from the above technical solutions, the embodiments of this application have the following advantages: This application provides a cuffless blood pressure calibration method, including: receiving a blood pressure calibration request sent by a user; measuring the calibration physiological signal of the current user based on the blood pressure calibration request, the calibration physiological signal including: PPG signal and measurement environment signal; calculating a predicted blood pressure value based on the calibration physiological signal and obtaining a calibrated blood pressure value; determining whether the difference between the calibrated blood pressure value and the predicted blood pressure value is greater than a preset range; if the difference between the calibrated blood pressure value and the predicted blood pressure value is less than or equal to the preset range, then storing the calibrated blood pressure and the physiological signal as historical calibration records; determining the user's calibration level based on the historical calibration records; and different calibration levels have different correction degrees for the target user's blood pressure value. In this method, different calibration levels correspond to different numbers of calibrations performed by the user. The more times the user calibrates, the greater the influence of the calibration data on the final value. This method can improve the reliability of calibration, allowing the calibration data to be better utilized in the blood pressure value calculation process, thereby increasing the accuracy of cuffless blood pressure measurement.

[0069] Based on the above Figure 1 Figure 2 The data processing rules for different calibration levels are described in detail below for the corresponding embodiments. For specific details, please refer to [link / reference]. Figure 3 One embodiment of the cuffless blood pressure calibration method provided in this application includes steps 301 to 304.

[0070] 301. Receive a blood pressure measurement request sent by the user.

[0071] 302. Measure the current user's physiological signals based on the blood pressure measurement request.

[0072] Specifically, steps 301 to 302 above are the same as those described above. Figure 1 The corresponding implementation is similar; please refer to the counting instructions for details, which will not be repeated here.

[0073] 303. Determine the calibration level corresponding to the user based on historical calibration records.

[0074] This includes: If the historical calibration record does not exist, the calibration level is determined to be zero; If the number of historical calibration records is less than or equal to the first preset value, then the calibration level is determined to be one; If the number of historical calibration records is greater than a first preset value and less than or equal to a second preset value, then the calibration level is determined to be two. If the number of historical calibration records is greater than the second preset value, then the calibration level is determined to be three.

[0075] The calibration levels determined based on historical calibration records include three categories: those that have not been calibrated at all, in which case the blood pressure value obtained after data processing using the initial model can be used as the output value.

[0076] After a certain number of calibrations, the historical calibration records formed at this point can partially influence the calculated values ​​obtained after data processing of the initial model, but will not completely adjust all the parameters of the initial model. In this case, multiple hierarchical adjustment strategies are usually set to meet the level of calibration performed by the user.

[0077] The calibration has been fully performed, and it can be assumed that the user has performed a large number of calibrations. The blood pressure value can be calculated entirely with reference to the calibration value. That is, all parameters of the initial model are adjusted so that the calibration data records are fully utilized.

[0078] Since the initial model used to calculate blood pressure values ​​was trained on a large amount of historical data from other groups who are not the individuals in question, the model may perform well for users whose data are around the average level, but may not perform well for the specific circumstances of the individuals in question. Therefore, by calibrating the model multiple times and updating the initial model with historical calibration data, the trained model can perform better on the blood pressure measurements of the individuals themselves, thus more effectively improving the model's accuracy.

[0079] In this embodiment, five calibration levels are used as an example for illustration. It is understood that the number of calibration levels and the rules for classifying them can be adjusted according to the actual situation in actual implementation, and no specific limitation is made here.

[0080] If the user has not performed any calibration, the calibration level is considered to be uncalibrated.

[0081] If the number of scenarios successfully calibrated by the user does not exceed the threshold If so, the calibration level is determined to be calibration level 1.

[0082] If the number of user-defined completed scenarios exceeds the threshold Less than the threshold If so, the user is in calibration level 2 state.

[0083] If the number of user-defined completed scenarios exceeds the threshold Less than the threshold If so, the user is in calibration level 3 state.

[0084] If the user completes all calibration scenarios, the user is now in a fully calibrated state.

[0085] Specifically, the calibration scenario settings are the same as those mentioned above. Figure 2 The corresponding implementation is similar, and the details can be found in the foregoing content, which will not be repeated here.

[0086] 304. The blood pressure value corresponding to the user is calculated based on the user's corresponding calibration level and the physiological signal.

[0087] Specifically, if a calibration method with calibration levels of zero, one, and two is used, the calculation method for blood pressure values ​​may include: If the calibration level corresponding to the user is zero, the physiological signal is input into the initial calculation model to obtain the blood pressure value corresponding to the user. If the user's calibration level is one, the physiological signal is input into the initial calculation model to obtain an initial reference value. Based on the initial reference value and the historical calibration value, a weighted calculation is performed to obtain the user's blood pressure value.

[0088] If the user's corresponding calibration level is two, then the initial calculation model is partially updated based on the historical calibration values ​​to obtain the first updated model; The physiological signal is processed based on the first update model to obtain a first reference value; The blood pressure value corresponding to the user is obtained by weighted calculation based on the first reference value and the historical calibration value.

[0089] If the user's corresponding calibration level is three, then the initial calculation model is updated based on the historical calibration value to obtain the second updated model; The physiological signals are processed based on the second update model to obtain the blood pressure value corresponding to the user.

[0090] Understandably, in actual implementation, the calibration strategy may be adjusted accordingly for different usage scenarios based on the different calibration level settings. The following is a detailed introduction to the calibration strategies applicable to the five levels described in step 303 above: (1) If the user account is in the calibration level 0 state, the user's blood pressure calculation will be based entirely on the initial preset model when the user account is in the uncalibrated state. The initial preset model was trained by the staff based on historical data.

[0091] (2) If the user account is in calibration level 1, the measured blood pressure is a weighted sum of the initial preset model calculation value and the reference value, and the calculation rules are as follows: Record the user's calibration values ​​and calibration times for the last 10 calibrations, calculate the time interval between each calibration and the most recent calibration, and increment all time intervals by 1 to avoid division by zero, resulting in a time interval array. Calculate the weight array for each calibration value. The weight calculation rules are as follows:

[0092]

[0093] This formula assigns higher weights to calibration values ​​generated from times closer to the current time and lower weights to calibration values ​​generated from times farther away from the current time. The final blood pressure reference value is obtained by weighting and summing the most recent 10 calibration values, using the SBP reference value as an example. For example, the calculation rules are as follows:

[0094]

[0095] in For the first The formula for calculating the final systolic blood pressure output value is as follows: (SBP calibration values ​​are listed below.)

[0096]

[0097] in This is a weighting factor, which is related to the time interval between the previous calibration and the current measurement. The value is calculated using a preset model. It should be noted that if the number of historical calibrations is less than 10, existing calibration data will be used for the above operations. Understandably, this output value depends on historical calibrations. If a user experiences a calibration error, they can delete the erroneous historical calibration record on the app. After confirmation, the app will update the calibration information and send it to the device.

[0098] (3) If the user is in calibration level 2, some model parameters of the initial model will be updated in this case. Specifically, the flowchart of the blood pressure calculation model is shown below. Figure 5As shown, layer 501 is the input layer for signal features. During processing, the collected parameters need to be input into this layer for subsequent processing. Layer 502 is the main body of the blood pressure calculation model, including but not limited to models using DNN (Deep Neural Networks), xgboost (eXtreme GradientBoosting), GBDT (Gradient Boosting Decision Tree), or combinations of these models, connected to the subsequent fully connected layers 1 and 2. The activation function of fully connected layer 1 is ReLU, while fully connected layer 2 has no activation function. Layer 505 is the output layer. After calibration, if the user is in calibration level 2, the app will update the parameters of fully connected layer 2 using historical calibration information from various test scenarios, and then send the retrained and updated parameters to the device and upload them to the server. After the blood pressure calculation model is updated, the collected signals are processed using the updated blood pressure calculation model to obtain the preliminary processed model output value. The model output value is then weighted and calculated with the calibration values ​​included in the historical calibration records to obtain the final output value. The specific weighting calculation process can refer to the weighting calculation method provided by calibration level 1, or the weighting calculation method can be adjusted according to your own needs. The details are not elaborated here.

[0099] (4) If the user is in calibration level 3, the model update parameters are the same as those in calibration level 2. However, the measured value after calibration is the updated model output value. That is, after the model update strategy based on calibration level 2 is completed, the updated model is used to process the collected signal and output the processed value. There is no need to use historical calibration records and model output values ​​for weighted calculation.

[0100] (5) If the user completes all the calibration scenarios, the user is in a fully calibrated state. In this state, the model will update all the parameters of fully connected layer 1 and fully connected layer 2. The calibrated measurement value is the updated model output value. Similarly, after training and updating the model on the APP, the parameters are uploaded to the server and sent to the device.

[0101] It should be noted that this solution also provides a method for synchronizing the equipment involved in the blood pressure calibration process; please refer to [reference needed]. Figure 6 , Figure 6This is a flowchart illustrating the data synchronization method during the cuffless blood pressure measurement and calibration process provided in this application. After each calibration, the calibration information is not only stored in the flash memory of the wearable device, but the device also uploads all calibration information to the mobile terminal APP for synchronization. After processing, the APP uploads the corresponding information to the cloud server via the network. When the user connects to the wearable device using the APP and triggers blood pressure measurement, if the APP can successfully request the server, it will obtain the calibration information from the server and compare it with the local calibration information to determine the latest calibration information. The APP will then use the data set containing the latest calibration information for calibration. If the APP cannot successfully request the server, it will send the latest calibration information from within the APP to the terminal device. After receiving the calibration information sent by the APP, the wearable device will compare it with the locally stored calibration information and use the latest calibration information. This information synchronization mechanism can reduce the problem of inconsistencies in calibration information among the wearable device, the APP (carrier), and the server.

[0102] The foregoing describes the cuffless blood pressure measurement method provided in this application. To support the implementation of the above embodiments, this application also provides a cuffless blood pressure measurement device. Please refer to [link to relevant documentation]. Figure 7 One embodiment of the cuffless blood pressure measurement device provided in this application includes: The receiving unit 701 is used to receive a blood pressure measurement request sent by the user; Measurement unit 702 is used to measure the current user's physiological signals based on the blood pressure measurement request, the physiological signals including: PPG signal and measurement environment signal; The determining unit 703 is used to determine the calibration level corresponding to the user based on historical calibration records, wherein different calibration levels have different degrees of correction for blood pressure values; The calculation unit is used to calculate the blood pressure value corresponding to the user based on the user's corresponding calibration level and the physiological signal.

[0103] Optionally, the device further includes: a calibration unit, used for: Receive blood pressure calibration requests sent by users; Based on the blood pressure calibration request, the calibration physiological signals of the current user are measured, including: PPG signal and measurement environment signal; Calculate the predicted blood pressure value based on the calibrated physiological signal and obtain the calibrated blood pressure value; Determine whether the difference between the calibrated blood pressure value and the predicted blood pressure value is greater than a preset range; If the difference between the calibrated blood pressure value and the predicted blood pressure value is less than or equal to the preset range, then the calibrated blood pressure and the physiological signal are stored as historical calibration records.

[0104] Optionally, the determining unit is specifically used for: If the historical calibration record does not exist, the calibration level is determined to be zero; If the number of historical calibration records is less than or equal to the first preset value, then the calibration level is determined to be one; If the number of historical calibration records is greater than a first preset value and less than or equal to a second preset value, then the calibration level is determined to be two. If the number of historical calibration records is greater than the second preset value, then the calibration level is determined to be three.

[0105] Optionally, the computing unit is specifically used for: If the calibration level corresponding to the user is zero, the physiological signal is input into the initial calculation model to obtain the blood pressure value corresponding to the user. Optionally, the computing unit is specifically used for: If the user's calibration level is one, the physiological signal is input into the initial calculation model to obtain an initial reference value. Based on the initial reference value and the historical calibration value, a weighted calculation is performed to obtain the user's blood pressure value.

[0106] Optionally, the computing unit is specifically used for: If the user's corresponding calibration level is two, then the initial calculation model is updated based on the historical calibration value to obtain the first updated model; The physiological signal is processed based on the first update model to obtain a first reference value; The blood pressure value corresponding to the user is obtained by weighted calculation based on the first reference value and the historical calibration value.

[0107] Optionally, the computing unit is specifically used for: If the user's corresponding calibration level is three, then the initial calculation model is updated based on the historical calibration value to obtain the second updated model; The physiological signals are processed based on the second update model to obtain the blood pressure value corresponding to the user.

[0108] The aforementioned cuffless measuring device can be used to implement the aforementioned Figure 1 , Figure 3 The cuffless blood pressure measurement method described in the corresponding embodiment.

[0109] This application also provides a cuffless blood pressure calibration device, comprising: The receiving unit 801 is used to receive a blood pressure calibration request sent by the user; Measurement unit 802 is used to measure the calibration physiological signal of the current user based on the blood pressure calibration request, the calibration physiological signal including: PPG signal and measurement environment signal; The acquisition unit 803 is used to calculate the predicted blood pressure value and acquire the calibrated blood pressure value based on the calibrated physiological signal; The judgment unit 804 is used to determine whether the difference between the calibrated blood pressure value and the predicted blood pressure value is greater than a preset range; Storage unit 805 is used to store the calibrated blood pressure and the physiological signal as historical calibration records if the difference between the calibrated blood pressure value and the predicted blood pressure value is less than or equal to the preset range.

[0110] The determining unit 806 is used to determine the calibration level corresponding to the user based on historical calibration records, and different calibration levels have different degrees of correction for blood pressure measurement values.

[0111] In this embodiment, the processes executed by each unit in the cuffless blood pressure calibration device are the same as those described above. Figure 2 The method flow described in the corresponding embodiments is similar and will not be repeated here. In actual implementation, the above... Figure 7 and Figure 8 The described cuffless blood pressure measurement device and cuffless blood pressure calibration device can be the same device, that is, the device is used for both measurement and calibration processes. The specific device can be determined according to the actual situation and is not limited here.

[0112] Figure 9 This is a schematic diagram of a cuffless blood pressure measuring device provided in an embodiment of this application. The cuffless blood pressure measuring device 900 may include one or more central processing units 901 and memory 905, and the memory 905 stores one or more application programs or data.

[0113] In this embodiment, the specific functional module division in the central processing unit 901 can be the same as described above. Figure 9 The functional module division of each unit described in the text is similar, so it will not be repeated here.

[0114] The memory 905 can be volatile or persistent storage. The program stored in the memory 905 can include one or more modules, each module including a series of instruction operations on the server. Furthermore, the central processing unit 901 can be configured to communicate with the memory 905 and execute the series of instruction operations in the memory 905 on the cuffless blood pressure measuring device 900.

[0115] The cuffless blood pressure measuring device 900 may also include one or more power supplies 902, one or more wired or wireless network interfaces 903, and one or more sensors 904.

[0116] The central processing unit 901 can perform the aforementioned... Figure 1 , Figure 3 The specific operations performed by the cuffless blood pressure measurement method in the illustrated embodiment will not be described in detail here.

[0117] Figure 10 This is a schematic diagram of a cuffless blood pressure calibration device provided in an embodiment of this application. The cuffless blood pressure calibration device 1000 may include one or more central processing units 1001 and memory 1005, and the memory 1005 stores one or more applications or data.

[0118] In this embodiment, the specific functional module division in the central processing unit 1001 can be the same as described above. Figure 8 The functional module division of each unit described in the text is similar, so it will not be repeated here.

[0119] The memory 1005 can be volatile or persistent storage. The program stored in the memory 1005 can include one or more modules, each module including a series of instruction operations on the server. Furthermore, the central processing unit 1001 can be configured to communicate with the memory 1005 and execute the series of instruction operations stored in the memory 1005 on the cuffless blood pressure calibration device 1000.

[0120] The cuffless blood pressure calibration device 1000 may also include one or more power supplies 1002, one or more wired or wireless network interfaces 1003, and one or more sensors 1004.

[0121] The central processing unit 1001 can perform the aforementioned... Figure 2 The specific operations performed by the cuffless blood pressure measurement method in the illustrated embodiment will not be described in detail here.

[0122] This application also provides a computer storage medium for storing computer software instructions used for the above-described cuffless blood pressure measurement method, including a program designed for executing the cuffless blood pressure measurement method.

[0123] The cuffless blood pressure measurement method can be as described above. Figure 1 or Figure 3 The cuffless blood pressure measurement method described in the document.

[0124] This application also provides a computer program product, which includes computer software instructions that can be loaded by a processor to implement the above-described functionality. Figure 1 , Figure 2 Figure 3The procedure for cuffless blood pressure measurement and calibration of any one of the following methods.

[0125] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, equivalent circuit transformations and unit divisions are only logical functional divisions. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connections shown or discussed may be through some interfaces, or indirect coupling or communication connections between apparatuses or units, and may be electrical, mechanical, or other forms.

[0126] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0127] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0128] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions or improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A cuffless blood pressure measurement method, characterized in that, include: Receive blood pressure measurement requests sent by users; Based on the blood pressure measurement request, the current user's physiological signals are measured, including: PPG signal and measurement environment signal; The calibration level of the user is determined based on historical calibration records, and different calibration levels have different degrees of correction for blood pressure values; The blood pressure value of the user is calculated based on the user's corresponding calibration level and the physiological signals.

2. The cuffless blood pressure measurement method according to claim 1, characterized in that, The method further includes: Receive blood pressure calibration requests sent by users; Based on the blood pressure calibration request, the calibration physiological signals of the current user are measured, including: PPG signal and measurement environment signal; Calculate the predicted blood pressure value based on the calibrated physiological signal and obtain the calibrated blood pressure value; Determine whether the difference between the calibrated blood pressure value and the predicted blood pressure value is greater than a preset range; If the difference between the calibrated blood pressure value and the predicted blood pressure value is less than or equal to the preset range, then the calibrated blood pressure and the physiological signal are stored as historical calibration records.

3. The cuffless blood pressure measurement method according to claim 1, characterized in that, The process of determining the user's calibration level based on historical calibration records includes: If the historical calibration record does not exist, the calibration level is determined to be zero; If the number of historical calibration records is less than or equal to the first preset value, then the calibration level is determined to be one; If the number of historical calibration records is greater than a first preset value and less than or equal to a second preset value, then the calibration level is determined to be two. If the number of historical calibration records is greater than the second preset value, then the calibration level is determined to be three.

4. The cuffless blood pressure measurement method according to claim 3, characterized in that, The calculation of the user's blood pressure value based on the user's corresponding calibration level and the physiological signals includes: If the calibration level corresponding to the user is zero, the physiological signal is input into the initial calculation model to obtain the blood pressure value corresponding to the user.

5. The cuffless blood pressure measurement method according to claim 3, characterized in that, The calculation of the user's blood pressure value based on the user's corresponding calibration level and the physiological signals includes: If the user's calibration level is one, the physiological signal is input into the initial calculation model to obtain an initial reference value. Based on the initial reference value and the historical calibration value, a weighted calculation is performed to obtain the user's blood pressure value.

6. The cuffless blood pressure measurement method according to claim 3, characterized in that, The calculation of the user's blood pressure value based on the user's corresponding calibration level and the physiological signals includes: If the user's corresponding calibration level is two, then the initial calculation model is updated based on the historical calibration value to obtain the first updated model; The physiological signal is processed based on the first update model to obtain a first reference value; The blood pressure value corresponding to the user is obtained by weighted calculation based on the first reference value and the historical calibration value.

7. The cuffless blood pressure measurement method according to claim 3, characterized in that, The calculation of the user's blood pressure value based on the user's corresponding calibration level and the physiological signals includes: If the user's corresponding calibration level is three, then the initial calculation model is updated based on the historical calibration value to obtain the second updated model; The physiological signals are processed based on the second update model to obtain the blood pressure value corresponding to the user.

8. A cuffless blood pressure calibration method, characterized in that, include: Receive blood pressure calibration requests sent by users; Based on the blood pressure calibration request, the calibration physiological signals of the current user are measured, including: PPG signal and measurement environment signal; Calculate the predicted blood pressure value based on the calibrated physiological signal and obtain the calibrated blood pressure value; Determine whether the difference between the calibrated blood pressure value and the predicted blood pressure value is greater than a preset range; If the difference between the calibrated blood pressure value and the predicted blood pressure value is less than or equal to the preset range, then the calibrated blood pressure and the physiological signal are stored as historical calibration records. The calibration level corresponding to the user is determined based on historical calibration records, and different calibration levels have different degrees of correction for blood pressure measurements.

9. A cuffless blood pressure measuring device, characterized in that, include: Central processing unit, memory, input / output interfaces, wired or wireless network interfaces, and power supply; The memory is either a short-term storage memory or a persistent storage memory; The central processing unit is configured to communicate with the memory and execute instructions in the memory on the cuffless blood pressure measuring device to perform the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Includes instructions that, when executed on a computer, cause the computer to perform the method as described in any one of claims 1 to 7.

Images