A method, system, electronic device, and storage medium for processing electrocardiogram and heart sound data.

Abstract

This application provides a method, system, electronic device, and storage medium for processing electrocardiogram (ECG) and heart sound data. A terminal processor simultaneously acquires ECG and heart sound data of a target user through an ECG and heart sound monitoring device and body status data of the target user through a body status monitoring device within a first time period. Based on the body status data, the reliability of the ECG and heart sound data is determined. If the reliability exceeds a preset value, a first analysis strategy for the ECG and heart sound data is calculated based on the current data processing pressure of the terminal processor and the current data processing pressure of the cloud server. The first analysis strategy includes the analysis content of the terminal processor and the analysis content of the cloud server, and the analysis content of the cloud server is uploaded to the cloud server. The cloud server analyzes the analysis content of the cloud server according to the determined analysis method to determine the cloud analysis result. This application can improve the efficiency of ECG and heart sound detection.

Description

Technical Field

[0001] This application relates to the field of medical technology, and more specifically, to a method, system, electronic device, and storage medium for processing electrocardiogram and heart sound data. Background Technology

[0002] With the improvement of living standards, the incidence of various cardiovascular and cerebrovascular diseases such as hypertension, hyperlipidemia, and coronary heart disease is increasing year by year. Traditional hospitals often use electrocardiogram (ECG) analysis to diagnose and treat patients' conditions. This method is cost-effective, but it cannot reflect the whole picture.

[0003] With the advancement of technology, electrocardiogram and heart sound detection technology has emerged, but traditional electrocardiogram and heart sound detection technology only performs joint analysis of electrocardiograms and heart sounds, and its application is not in-depth enough. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a method, system, electronic device and storage medium for processing electrocardiogram and heart sound data, which can improve the efficiency of electrocardiogram and heart sound detection.

[0005] In a first aspect, embodiments of this application provide a method for processing electrocardiogram (ECG) and heart sound data, applied to an ECG and heart sound monitoring system. The monitoring system includes an ECG and heart sound monitoring device, a body status monitoring device, a terminal processor, and a cloud server, all installed at the location of the user being tested. The terminal processor is communicatively connected to the cloud server, the ECG and heart sound monitoring device, and the body status monitoring device.

[0006] The processing method includes a data uploading stage and a data analysis stage;

[0007] The data upload phase includes:

[0008] During the first time period, the terminal processor simultaneously acquires the target user's electrocardiogram and heart sound data through the electrocardiogram and heart sound monitoring device, and acquires the target user's body status data through the body status monitoring device; the electrocardiogram and heart sound data includes electrocardiogram data and heart sound data;

[0009] The terminal processor determines the reliability of the electrocardiogram and heart sound data based on the body status data;

[0010] If the credibility exceeds a preset value, the terminal processor calculates a first analysis strategy for the ECG and heart sound data based on the current data processing pressure of the terminal processor and the current data processing pressure of the cloud server; the first analysis strategy includes the analysis content of the terminal processor and the analysis content of the cloud server.

[0011] The terminal processor uploads the analyzed content from the cloud server to the cloud server;

[0012] The data analysis phase includes:

[0013] The cloud server analyzes the content of the cloud server according to a determined analysis method to determine the cloud analysis results.

[0014] In one optional embodiment of this application, if the confidence level exceeds a preset value, the terminal processor calculates a first analysis strategy for the ECG and heart sound data based on the current data processing pressure of the terminal processor and the current data processing pressure of the cloud server, including:

[0015] If the credibility exceeds the preset value, the terminal processor calculates the estimated computing power consumption based on the total amount of ECG and heart sound data and the pre-input analysis requirements.

[0016] The terminal processor calculates the analysis content for the terminal processor and the analysis content for the cloud server based on the current data processing pressure of the terminal processor, the current data processing pressure of the cloud server, and the estimated computing power consumption.

[0017] In one optional embodiment of this application, the terminal processor calculates the analysis content of the terminal processor and the analysis content of the cloud server based on the current data processing pressure of the terminal processor, the current data processing pressure of the cloud server, and the estimated computing power consumption, including:

[0018] If the current data processing pressure of the terminal processor does not meet the estimated computing power consumption, and the current data processing pressure of the cloud server is greater than a preset value, then the terminal processor will split the analysis requirements to determine the preliminary analysis task and the fine analysis task, and extract the preliminary data and fine analysis data from the ECG and heart sound data based on the preliminary analysis task and the fine analysis task respectively; the preliminary data is carried in the analysis content of the terminal processor.

[0019] The terminal processor generates a preliminary judgment result for the target user based on the preliminary judgment data and the preliminary judgment analysis task;

[0020] If the initial judgment result does not meet the preset conditions, the terminal processor uses the refined judgment data and refined judgment analysis task to generate the analysis content of the cloud server; if the initial judgment result meets the preset conditions, the terminal processor provides feedback to the target user based on the initial judgment result.

[0021] In one optional embodiment of this application, the terminal processor calculates the analysis content of the terminal processor and the analysis content of the cloud server based on the current data processing pressure of the terminal processor, the current data processing pressure of the cloud server, and the estimated computing power consumption, and further includes:

[0022] If the current data processing pressure of the terminal processor meets the estimated computing power consumption requirement, and the current data processing pressure of the cloud server is less than a preset value, then the terminal processor will split the ECG and heart sound data based on the analysis requirements, the current data processing pressure of the terminal processor, and the current data processing pressure of the cloud server to determine the first type of computational data, the calculation rules of the first type of computational data, the second type of computational data, and the calculation rules of the second type of computational data.

[0023] The terminal processor carries the first type of computational data and the computational rules of the first type of computational data in the analysis content of the terminal processor; and carries the second type of computational data and the computational rules of the second type of computational data in the analysis content of the cloud server; the ratio of the computational power consumption of executing the computational rules of the first type of computational data to the computational power consumption of executing the computational rules of the second type of computational data is negatively correlated with the ratio of the current data processing pressure of the terminal processor to the current data processing pressure of the cloud server;

[0024] The method further includes:

[0025] The terminal processor generates terminal analysis results based on the analysis content of the terminal processor;

[0026] The terminal processor combines the terminal analysis results with the cloud analysis results to generate the final analysis result.

[0027] In one optional embodiment of this application, the physical condition data includes: mental state data and skin condition data; the mental state data is determined based on the following data: heart data, facial image features, blood pressure data, and respiratory data;

[0028] The terminal processor determines the reliability of the electrocardiogram and heart sound data based on the body state data, including:

[0029] The terminal processor determines the reliability of the electrocardiogram and heart sound data based on the values ​​of the mental state data;

[0030] The cloud server analyzes the analysis content according to a defined analysis method to determine the cloud analysis results, including:

[0031] The cloud server determines the calibration offset between electrocardiogram data and heart sound data based on mental state data and skin condition data;

[0032] The cloud server generates cloud analysis results based on the calibration offset, the first type of calculation data, and the calculation rules of the first type of calculation data.

[0033] In one optional embodiment of this application, the cloud server determines the calibration offset between electrocardiogram data and heart sound data based on mental state data and skin condition data, including:

[0034] The cloud server determines whether its current data processing load exceeds a preset value.

[0035] If the current data processing pressure of the cloud server is greater than the preset value, the cloud server will call the pre-trained offset calculation model of the target user according to the target user's attribute information.

[0036] The cloud server inputs the target user's mental state data and skin condition data into the pre-trained offset calculation model to determine the calibration offset between the electrocardiogram data and the heart sound data;

[0037] If the current data processing pressure of the cloud server is less than the preset value, the cloud server uses the target user's mental state data and skin condition data to find reference users of the same type as the target user, and uses the reference user's mental state data, skin condition data, reference user's electrocardiogram data and heart sound data to train the initial RNN model, and obtain the trained offset calculation model.

[0038] The cloud server inputs the target user's mental state data and skin condition data into the pre-trained offset calculation model to determine the calibration offset between the electrocardiogram data and the heart sound data.

[0039] Secondly, this application also provides an electrocardiogram and heart sound monitoring system, including: an electrocardiogram and heart sound monitoring device, a body status monitoring device, a terminal processor, and a cloud server, all installed at the location of the user being tested; the terminal processor is communicatively connected to the cloud server, the electrocardiogram and heart sound monitoring device, and the body status monitoring device, respectively.

[0040] During the first time period, the terminal processor simultaneously acquires the target user's electrocardiogram and heart sound data through the electrocardiogram and heart sound monitoring device, and acquires the target user's body status data through the body status monitoring device; the electrocardiogram and heart sound data includes electrocardiogram data and heart sound data;

[0041] The terminal processor determines the reliability of the electrocardiogram and heart sound data based on the body status data;

[0042] If the credibility exceeds a preset value, the terminal processor calculates a first analysis strategy for the ECG and heart sound data based on the current data processing pressure of the terminal processor and the current data processing pressure of the cloud server; the first analysis strategy includes the analysis content of the terminal processor and the analysis content of the cloud server.

[0043] The terminal processor uploads the analyzed content from the cloud server to the cloud server;

[0044] The cloud server analyzes the content of the cloud server according to a determined analysis method to determine the cloud analysis results.

[0045] In one optional embodiment of this application, if the confidence level exceeds a preset value, the terminal processor calculates a first analysis strategy for the ECG and heart sound data based on the current data processing pressure of the terminal processor and the current data processing pressure of the cloud server, including:

[0046] If the credibility exceeds the preset value, the terminal processor calculates the estimated computing power consumption based on the total amount of ECG and heart sound data and the pre-input analysis requirements.

[0047] The terminal processor calculates the analysis content for the terminal processor and the analysis content for the cloud server based on the current data processing pressure of the terminal processor, the current data processing pressure of the cloud server, and the estimated computing power consumption.

[0048] Thirdly, embodiments of this application also provide an electronic device, including: a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform the steps of the method described above.

[0049] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the method described above.

[0050] Compared with the prior art, this application has at least the following technical effects:

[0051] The ECG and heart sound data processing method used in this application is applied to an ECG and heart sound monitoring system. The ECG and heart sound detection device and the body status monitoring device in the detection system can directly collect the ECG and heart sound data and body status data of the target user. In actual operation, the terminal processor can determine whether subsequent processing can be carried out based on the body status data, and determine the appropriate analysis strategy for the system in a timely manner based on the current data processing pressure of the terminal processor and the current data processing pressure of the cloud server. The entire analysis task is completed by the cooperation of the terminal processor and the cloud server, ensuring the efficiency of completion.

[0052] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0053] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 A flowchart illustrating a method for processing electrocardiogram and heart sound data provided in an embodiment of this application;

[0055] Figure 2 A flowchart illustrating another method for processing electrocardiogram and heart sound data provided in an embodiment of this application;

[0056] Figure 3 A schematic diagram showing electrocardiogram data and heart sound data in an image provided for an embodiment of this application;

[0057] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. Based on the embodiments of this application, every other embodiment obtained by those skilled in the art without inventive effort falls within the scope of protection of this application.

[0059] While techniques using electrocardiograms and heart sounds for patient diagnosis already exist, these techniques are still in their early stages and cannot be readily adapted to specific needs. In response to this situation, such as... Figure 1 As shown, this application provides a method for processing electrocardiogram (ECG) and heart sound data, applied to an ECG and heart sound monitoring system. The monitoring system includes an ECG and heart sound monitoring device, a body status monitoring device, a terminal processor, and a cloud server, all installed at the location of the user being tested. The terminal processor is communicatively connected to the cloud server, the ECG and heart sound monitoring device, and the body status monitoring device.

[0060] The processing method includes a data uploading stage and a data analysis stage;

[0061] The data upload phase includes:

[0062] S101, the terminal processor simultaneously acquires the target user's electrocardiogram and heart sound data through the electrocardiogram and heart sound monitoring device, and acquires the target user's body status data through the body status monitoring device during the first time period; the electrocardiogram and heart sound data includes electrocardiogram data and heart sound data;

[0063] S102, the terminal processor determines the reliability of the electrocardiogram and heart sound data based on the body status data;

[0064] S103, if the credibility exceeds a preset value, the terminal processor calculates a first analysis strategy for the ECG and heart sound data based on the current data processing pressure of the terminal processor and the current data processing pressure of the cloud server; the first analysis strategy includes the analysis content of the terminal processor and the analysis content of the cloud server.

[0065] S104, The terminal processor uploads the analysis content from the cloud server to the cloud server;

[0066] The data analysis phase includes:

[0067] S105, the cloud server analyzes the analysis content of the cloud server according to the determined analysis method to determine the cloud analysis result.

[0068] The monitoring system in this solution can be portable, meaning the body condition monitoring device and the ECG / heart sound monitoring device are carried directly on the user's person and will not hinder the user's normal activities; conversely, the monitoring system can also be fixed, meaning the body condition monitoring device and the ECG / heart sound monitoring device are relatively fixed on the ground. When the monitoring system is fixed, the processing power of the terminal processor will inevitably be less than that of the cloud server. When the monitoring system is portable, the processing power of the terminal processor will be further reduced. In other words, to ensure portability, the terminal processor needs to be smaller.

[0069] When the detection system is portable (in actual products, a portable monitoring system is preferred), its main feature is its portability. Users can wear miniaturized monitoring devices on their bodies, thus achieving portability. Generally, ECG and heart sound monitoring devices consist of at least three parts: a monitoring device, a fixing component, and a transmission device. The monitoring device can be composed of electrode catheters and catheter sensors, etc. The catheter sensor, which contacts the human skin, is used to collect ECG and heart sound signals from within the body. The electrode catheter is used to transmit the collected ECG and heart sound signals to the transmission device in the terminal processor, and further transmits the ECG and heart sound signals to the cloud server through the transmission device.

[0070] There are two setup methods for the transmission equipment: direct communication with the cloud server via long-distance communication technology, and indirect communication with the cloud server via short-range communication technology. Specifically, with long-distance communication technology, a long-term connection can be maintained with the cloud server using technologies such as GPRS, and the ECG and heart sound signals are transmitted to the cloud server through this connection. With short-range communication technology, multiple repeaters (routers) need to be set up in the target user's living and working areas. After collecting the ECG and heart sound signals, the ECG and heart sound acquisition device transmits them to nearby routers via short-range communication technology (such as Bluetooth), and the routers then transmit the signals to the cloud server. Of course, if the user is at home, the home router can be used directly as a repeater to relay the ECG and heart sound signals to the cloud server. The fasteners in ECG and heart sound monitoring equipment are used to carry the monitoring and transmission devices on the user's body (such as on clothing). For example, the monitoring device can be hung on the user's clothes by fasteners, which can be structural components such as clips.

[0071] In step S101, the terminal processor acquires ECG and heart sound data and body status data simultaneously within the same time period (the first time period). The ECG and heart sound data includes both ECG data and heart sound data, which are measured synchronously. The body status data primarily reflects the target user's physical and mental state. This is mainly because the accuracy of the ECG and heart sound data is influenced to some extent by the user's physical and mental state. Therefore, measuring the body status data first can determine whether to proceed with subsequent steps. That is, in step S102, the reliability of the ECG and heart sound data can be determined based on the body status data. If it is reliable, step S103 can be executed; otherwise, the process can be terminated, or other methods can be used (such as prompting the target to execute step S101 at a different time).

[0072] Specifically, in step S102, if the user's physical condition data is not stable enough or deviates too much from historical data, the current reliability is considered insufficient. There are two possible reasons for this: user health issues or equipment problems (problems with the physical condition monitoring device). In this case, replacing the device with a new one and re-measuring is recommended. Generally, drastic data fluctuations are mostly due to hardware problems. If the data is due to user health issues, it needs to be handled in two ways: Scenario 1: caused by temporary or stress-related issues such as user rest. In this case, measurement can be performed after the user's condition has improved. Scenario 2: the user has a long-term health problem, which cannot be resolved even after a period of rest (usually a few minutes or hours after the user has calmed down). In this case, the measured data should be considered accurate, and analysis can be performed based on the ECG and heart sound data from step S101.

[0073] Therefore, in practical implementation, after detecting body status data, if the data fluctuates excessively, it can be directly considered that the reliability is too low (exceeding the threshold value). In this case, it can be considered a hardware problem, and a different device should be used for detection. If the reliability exceeds the threshold value but does not reach a reasonable value, it should be remeasured after a period of time, and the judgment should be re-evaluated based on the new measured data (body status data). If the remeasured body status data reaches a reasonable value, the remeasured data (ECG and heart sound data) should be used for analysis. If the reliability reaches a reasonable value, the ECG and heart sound data obtained in the first time period can be directly used for analysis.

[0074] In step S103, after confirming that the credibility has reached a predetermined value, the first analysis strategy can be determined based on the terminal's data processing pressure (the terminal's idle computing power) and the cloud's data processing pressure (the cloud's idle computing power). The first analysis strategy includes the tasks that the terminal processor needs to execute and the tasks that the cloud server needs to execute. These two tasks combined constitute the complete task of analyzing the target user's ECG and heart sound data. The amount of tasks executed here is positively correlated with the size of the idle computing power. Generally, since the computing power of the terminal processor is much smaller than that of the cloud server, the cloud server is allocated more tasks. However, the issue of network transmission bandwidth also needs to be considered. Since the system in this scheme is preferably a portable system, the terminal processor often uses unlimited transmission when transmitting data to the cloud server, thus limiting the transmission bandwidth. Therefore, when dividing tasks, the bandwidth situation (bandwidth occupancy) should be considered to determine the number of tasks allocated to the terminal processor and the cloud server. Specifically, when the computing power occupied by processing tasks is the same, tasks that occupy less bandwidth for sending data are preferentially allocated to the cloud server. Based on this strategy, the analysis content of the terminal processor and the analysis content of the cloud server can be determined. The specific analysis time, analysis method, and other content of the terminal processor and cloud server can be set according to the analysis requirements (the requirements given by the user or doctor).

[0075] Then, in step S104, the analysis content from the cloud server can be uploaded to the cloud server, and the terminal processor can independently process the analysis content to determine the terminal analysis results.

[0076] Finally, the cloud server can analyze the content received from the cloud server according to a preset analysis method to determine the cloud analysis results. Afterwards, the terminal processor can receive the cloud analysis results from the cloud server and merge the terminal analysis results with the cloud analysis results to generate the final analysis results that the target user can view.

[0077] In specific implementation, such as Figure 2 As shown, step S103 can be implemented as follows:

[0078] S1031, if the credibility exceeds the preset value, the terminal processor calculates the estimated computing power consumption based on the total amount of ECG and heart sound data and the pre-input analysis requirements.

[0079] S1032, the terminal processor calculates the analysis content of the terminal processor and the analysis content of the cloud server based on the current data processing pressure of the terminal processor, the current data processing pressure of the cloud server, and the estimated computing power consumption.

[0080] In step S1031, if the confidence level exceeds the preset value (the confidence level has reached a reasonable value), the terminal processor can predict the estimated amount of computing power required to perform the analysis of the ECG and heart sound data according to the analysis requirements.

[0081] Then, in step S1032, following the principle of load balancing, the ratio of the workload allocated to the cloud server and the terminal processor is determined based on the current data processing pressure of the terminal processor (reflecting idle computing power) and the current data processing pressure of the cloud server (reflecting idle computing power). (This ratio is usually a range value, and the specific value can be determined based on bandwidth usage.) Then, based on this ratio and the estimated computing power consumption, the workload allocated to the terminal processor and the cloud server can be determined. Then, all the analysis content (analysis tasks) can be broken down into multiple subtasks. Finally, based on the computing power consumed by each subtask, the analysis content for the terminal processor and the analysis content for the cloud server are calculated separately.

[0082] More specifically, step S1032 can be implemented as follows:

[0083] Step 10321: If the current data processing pressure of the terminal processor does not meet the estimated computing power consumption, and the current data processing pressure of the cloud server is greater than a preset value, then the terminal processor will split the analysis requirements to determine the preliminary analysis task and the fine analysis task, and extract the preliminary data and fine analysis data from the ECG and heart sound data based on the preliminary analysis task and the fine analysis task, respectively; the preliminary data is carried in the analysis content of the terminal processor.

[0084] Step 10322: The terminal processor generates a preliminary judgment result for the target user based on the preliminary judgment data and the preliminary judgment analysis task;

[0085] Step 10323: If the initial judgment result does not meet the preset conditions, the terminal processor generates the analysis content of the cloud server using the refined judgment data and refined judgment analysis task; if the initial judgment result meets the preset conditions, the terminal processor provides feedback to the target user based on the initial judgment result.

[0086] In step 10321, if the current data processing pressure of the terminal processor does not meet the estimated computing power consumption (the terminal processor has limited remaining computing power), and the current data processing pressure of the cloud server is greater than the preset value (the terminal processor has limited remaining computing power), then we should consider how to reduce computing power consumption. This means dividing the analysis process into two stages: preliminary judgment (consuming less computing power, performing critical analysis, but with insufficient precision) and fine judgment (consuming more computing power, performing a more comprehensive analysis with relatively high precision). Of course, considering the difference in computing power (the server's computing power is always greater than the terminal's), the terminal processor should perform the preliminary judgment, while the cloud server should perform the fine judgment. Correspondingly, the complete task (analysis requirements) needs to be broken down to generate preliminary judgment analysis tasks and fine judgment analysis tasks. Based on these tasks, preliminary judgment data and fine judgment data should be extracted from the ECG and heart sound data, respectively.

[0087] In step 10322, the terminal processor generates a preliminary judgment result for the target user based on the preliminary judgment data and the preliminary judgment analysis task. Simultaneously, it determines whether the preliminary judgment result meets expectations (whether it can definitively determine that the target user's analysis needs have been met, such as definitively determining that the user is in a normal state). If the determination is yes, the terminal processor can directly provide feedback to the target user based on the preliminary judgment result. If the determination is no (e.g., it cannot definitively determine that the user is in a normal state), it indicates that a refined judgment is needed. In this case, the terminal processor needs to use the refined judgment data and refined judgment analysis task to generate analysis content for the cloud server, and the cloud server will then perform the refined judgment action.

[0088] It should be noted that the fine-grained judgment task is also composed of multiple sub-tasks. After determining that the initial judgment result does not meet expectations, we can specifically analyze which part of the initial judgment result has a significant anomaly. When generating the analysis content for the cloud server using the fine-grained judgment data and the fine-grained judgment analysis task, we should selectively include the parts that can significantly contribute to the judgment accuracy in the analysis content for the cloud server. In practice, when generating the analysis content for the cloud server, the impact of bandwidth can also be considered. When bandwidth is limited, we should consider filtering the fine-grained judgment data in the analysis content for the cloud server to reduce the amount of data sent to a certain extent.

[0089] Correspondingly, if the current data processing pressure of the terminal processor meets the estimated computing power consumption requirement, and the current data processing pressure of the cloud server is less than the preset value, then step S1032 can be executed as follows:

[0090] Step 10324: If the current data processing pressure of the terminal processor meets the estimated computing power consumption requirement, and the current data processing pressure of the cloud server is less than a preset value, then the terminal processor will split the ECG and heart sound data based on the analysis requirements, the current data processing pressure of the terminal processor, and the current data processing pressure of the cloud server to determine the first type of computational data, the calculation rules of the first type of computational data, the second type of computational data, and the calculation rules of the second type of computational data.

[0091] Step 10325: The terminal processor carries the first type of computing data and the computing rules of the first type of computing data in the analysis content of the terminal processor; and carries the second type of computing data and the computing rules of the second type of computing data in the analysis content of the cloud server; the ratio of the computing power consumption of executing the computing rules of the first type of computing data to the computing power consumption of executing the computing rules of the second type of computing data is negatively correlated with the ratio of the current data processing pressure of the terminal processor to the current data processing pressure of the cloud server;

[0092] The method further includes:

[0093] The terminal processor generates terminal analysis results based on the analysis content of the terminal processor;

[0094] The terminal processor combines the terminal analysis results with the cloud analysis results to generate the final analysis result.

[0095] The conditions in step 10324 indicate that both the terminal and the cloud are in a normal state (the opposite of the state in step 10321). At this point, step S1032 can be executed solely based on the load balancing strategy. That is, the terminal processor, based on the analysis requirements, the current data processing pressure of the terminal processor, and the current data processing pressure of the cloud server, splits the ECG and heart sound data to determine the first type of computational data, the computation rules for the first type of computational data, and the computation rules for the second type of computational data. Simply put, the greater the data processing pressure on both the terminal processor and the cloud server, the less data is processed, and the simpler the computation rules become. The computation rules for the first type of computational data and the second type of data correspond to the preliminary judgment analysis task and the refined judgment analysis task mentioned earlier.

[0096] Then, in step 10325, analysis content for the cloud server and analysis content for the terminal processor are generated respectively. In subsequent steps, the terminal processor can generate terminal analysis results based on its analysis content, and the cloud server can generate cloud analysis results based on its analysis content.

[0097] The terminal processor combines the terminal analysis results with the cloud analysis results to generate the final analysis result.

[0098] In addition to adopting a load balancing strategy, the impact of users' physical condition data on data analysis should also be considered during implementation. Specifically, physical condition data can include the following types: mental state data and skin condition data. Mental state data is determined based on the following data: heart data, facial image features, blood pressure data, and respiratory data.

[0099] Specifically, cardiac data can include heart rate data; heart rate data can be further divided into heart rate variability (the stability of heart rate) and heart rate variability (the degree of deviation from normal values). Blood pressure data can be measured separately as diastolic and systolic blood pressure, and can be judged based on whether the blood pressure meets general data requirements (common examples include systolic blood pressure below 120 mmHg and diastolic blood pressure below 80 mmHg). Similarly, changes in blood pressure can also be detected (e.g., whether there is a significant change in blood pressure within 20 minutes). Respiratory data is similar, mainly respiratory rate and respiratory saturation (the amount of oxygen inhaled per breath); respiratory data can be determined by detecting the number and amplitude of chest rises and falls using a detector placed close to the target user's chest.

[0100] Facial feature data primarily reflects the target user's facial expressions, such as whether they are drowsy (judged by blinking frequency and eye opening / closing), and whether their face has color. Specifically, this can be achieved through image recognition. Positive and negative samples can be prepared in advance to train the model. Positive samples can be photos of people with normal facial features, while negative samples can be photos of people who are drowsy or have pale faces. By pre-training the model, the trained model can be directly retrieved for video testing. Of course, besides using images for training, video can also be used for model training and recognition.

[0101] Skin condition data mainly refers to the skin's moisture content (such as sweat), which can affect the accuracy of the test to some extent, but the impact is not significant.

[0102] Specifically, step S102 may involve the terminal processor determining the reliability of the electrocardiogram and heart sound data based on the values ​​of the mental state data. The aforementioned mental state data can reflect whether the user is in a normal state. If the user is in an abnormal state, the significance of subsequent testing is also very small, and the process can be terminated.

[0103] Furthermore, step S105 can be performed as follows:

[0104] Step 1051: The cloud server determines the calibration offset between the electrocardiogram data and the heart sound data based on the mental state data and skin condition data.

[0105] Step 1052: The cloud server generates cloud analysis results based on the calibration offset, the first type of calculation data, and the calculation rules of the first type of calculation data.

[0106] The preceding steps S10321-10323 introduced the principle and process by which a cloud server can perform precise analysis using analysis content carrying precise analysis data and tasks. Steps 1051-1052 also address the issue of offset, and it should be noted that offset directly affects the accuracy of the final analysis. Step 1051 primarily discloses the determination of the calibration offset between ECG data and heart sound data based on mental state data and skin condition data. Specifically, ECG data and heart sound data are correlated to a certain extent, such as... Figure 3 The image shows a schematic diagram displaying ECG and heart sound data in a single image. The first row of waveforms represents the ECG data, while the second to fourth rows represent heart sound data at low, medium, and high frequencies. Auxiliary lines (vertical lines) are displayed at the top (left) of both the first and second rows of waveforms. These vertical lines mark abnormal points or areas that the user or doctor needs to focus on. Analysis of these areas requires collaborative analysis of both ECG and heart sound data to arrive at the final result. Offset refers to the correlation between ECG and heart sound data over time, or in other words, it reflects which time point of ECG data is correlated with which time point of heart sound data. Mental state data and skin condition data are two parameters that can affect the offset. Specifically, excessive changes in mental state data can distort the measured data, thus changing the correspondence between ECG and heart sound data. Skin condition data has a smaller impact, but can sometimes significantly affect the results and should therefore be monitored.

[0107] After determining the mental state data and skin condition data, the calibration offset between the electrocardiogram data and heart sound data can be determined based on the mental state data and skin condition data. Then, in step 1052, a preliminary analysis result can be generated based on the first type of calculated data and the calculation rules of the first type of calculated data (the preliminary analysis result mainly reflects the result calculated according to the case where the offset is 0, but the result has a deviation). The calibration offset is then used to correct the preliminary analysis result, thereby obtaining the cloud analysis result.

[0108] Furthermore, the above method for determining the offset is relatively coarse, mainly because it treats all users as the same when calculating the offset. To improve calculation accuracy, users can be categorized, and then the offset can be calculated specifically according to the type. Specifically, step 1051 can be executed as follows:

[0109] Step 10511: The cloud server determines whether the current data processing pressure on the cloud server is greater than a preset value;

[0110] Step 10512: If the current data processing pressure of the cloud server is greater than the preset value, the cloud server calls the pre-trained offset calculation model of the target user according to the target user's attribute information.

[0111] Step 10513: The cloud server inputs the target user's mental state data and skin condition data into the pre-trained offset calculation model to determine the calibration offset between the electrocardiogram data and the heart sound data.

[0112] Step 10514: If the current data processing pressure of the cloud server is less than the preset value, the cloud server uses the target user's mental state data and skin condition data to find reference users of the same type as the target user, and uses the reference offset between the reference user's mental state data, skin condition data, ECG data and heart sound data to train the initial RNN model, and obtains the trained offset calculation model.

[0113] Step 10515: The cloud server inputs the target user's mental state data and skin condition data into the pre-trained offset calculation model to determine the calibration offset between the electrocardiogram data and the heart sound data.

[0114] The execution method of step 10511 is the same as that of step S103; both involve checking the data processing pressure on the cloud server. If the pressure is high, step 10512 is executed, which uses the pre-trained offset calculation model to calculate the offset. It should be noted that this processing method also employs a classification approach, although this classification is not the most targeted, but the cost is controllable. Because the model training is not performed when the cloud server's data processing pressure is high, it will not affect the current operation of the cloud server.

[0115] Correspondingly, if the data processing pressure is relatively low, step 10541 is executed, which involves reclassifying the user's mental state and skin condition data, and then temporarily training the model. This results in more targeted analysis using the trained model. The samples used for training the model should ideally be prepared in advance, generally requiring manual labeling of positive and negative samples. The samples used for training the model in step 10514 can be the same as those used in training the pre-trained offset calculation model in step 10512. In other words, the samples used for pre-training and temporary model training can be the same, but the classification method or the specific training samples differ.

[0116] Furthermore, the solution provided in this application considers operational pressure in both the process of calculating the calibration offset and in determining the analysis content of the terminal processor and cloud server. However, the optimal values ​​for considering pressure differ between these two steps. The process of calculating the calibration offset primarily affects accuracy (according to current statistics, for male target users with relatively stable historical health, even recent data mutations will not significantly impact the results; however, other user types will be somewhat affected, but the overall impact is relatively controllable). In contrast, determining the analysis content of the terminal processor and cloud server directly affects whether the terminal processor and cloud server can operate normally (whether they will crash or fail to provide timely feedback to the user). Therefore, the priority (importance) of considering data processing pressure when determining the analysis content of the terminal processor and cloud server is greater than that of considering data processing pressure in the process of calculating the calibration offset.

[0117] Based on the same inventive concept, this application also provides an electrocardiogram and heart sound data processing system corresponding to the electrocardiogram and heart sound data processing method. Since the principle of the system in this application is similar to the electrocardiogram and heart sound data processing method described above in this application, the implementation of the system can refer to the implementation of the method, and the repeated parts will not be described again.

[0118] Specifically, the ECG and heart sound monitoring system includes: an ECG and heart sound monitoring device, a body status monitoring device, a terminal processor, and a cloud server, all installed at the user's location; the terminal processor is communicatively connected to the cloud server, the ECG and heart sound monitoring device, and the body status monitoring device, respectively.

[0119] During the first time period, the terminal processor simultaneously acquires the target user's electrocardiogram and heart sound data through the electrocardiogram and heart sound monitoring device, and acquires the target user's body status data through the body status monitoring device; the electrocardiogram and heart sound data includes electrocardiogram data and heart sound data;

[0120] The terminal processor determines the reliability of the electrocardiogram and heart sound data based on the body status data;

[0121] If the credibility exceeds a preset value, the terminal processor calculates a first analysis strategy for the ECG and heart sound data based on the current data processing pressure of the terminal processor and the current data processing pressure of the cloud server; the first analysis strategy includes the analysis content of the terminal processor and the analysis content of the cloud server.

[0122] The terminal processor uploads the analyzed content from the cloud server to the cloud server;

[0123] The cloud server analyzes the content of the cloud server according to a determined analysis method to determine the cloud analysis results.

[0124] In one optional embodiment of this application, if the confidence level exceeds a preset value, the terminal processor calculates a first analysis strategy for the ECG and heart sound data based on the current data processing pressure of the terminal processor and the current data processing pressure of the cloud server, including:

[0125] If the credibility exceeds the preset value, the terminal processor calculates the estimated computing power consumption based on the total amount of ECG and heart sound data and the pre-input analysis requirements.

[0126] The terminal processor calculates the analysis content for the terminal processor and the analysis content for the cloud server based on the current data processing pressure of the terminal processor, the current data processing pressure of the cloud server, and the estimated computing power consumption.

[0127] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 4 As shown, the electronic device 400 includes a processor 401, a memory 402, and a bus 403.

[0128] The memory 402 stores machine-readable instructions executable by the processor 401. When the electronic device 400 is running, the processor 401 communicates with the memory 402 via the bus 403. When the machine-readable instructions are executed by the processor 401, they can perform the operations described above. Figure 1 as well as Figure 2 The steps of the ECG and heart sound data processing method in the illustrated method embodiment can be found in the method embodiment for specific implementation, and will not be repeated here.

[0129] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, can perform the above-described actions. Figure 1 as well as Figure 2 The steps of the ECG and heart sound data processing method in the illustrated method embodiment can be found in the method embodiment for specific implementation, and will not be repeated here.

[0130] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0131] 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. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the shown or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0132] 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 network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0133] In addition, 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.

[0134] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0135] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The scope of protection of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for processing electrocardiogram and heart sound data, characterized in that, An electrocardiogram and heart sound monitoring system is used, the monitoring system including an electrocardiogram and heart sound monitoring device, a body status monitoring device, a terminal processor, and a cloud server, all installed at the user being tested; the terminal processor is communicatively connected to the cloud server, the electrocardiogram and heart sound monitoring device, and the body status monitoring device, respectively. The processing method includes a data uploading stage and a data analysis stage; The data upload phase includes: During the first time period, the terminal processor simultaneously acquires the target user's electrocardiogram and heart sound data through the electrocardiogram and heart sound monitoring device, and acquires the target user's body status data through the body status monitoring device; the electrocardiogram and heart sound data includes electrocardiogram data and heart sound data; The terminal processor determines the reliability of the electrocardiogram and heart sound data based on the body status data; If the credibility exceeds a preset value, the terminal processor calculates a first analysis strategy for the ECG and heart sound data based on the current data processing pressure of the terminal processor and the current data processing pressure of the cloud server; the first analysis strategy includes the analysis content of the terminal processor and the analysis content of the cloud server. The terminal processor uploads the analyzed content from the cloud server to the cloud server; The data analysis phase includes: The cloud server analyzes the analysis content of the cloud server according to a determined analysis method to determine the cloud analysis results; If the reliability exceeds a preset value, the terminal processor calculates a first analysis strategy for the ECG and heart sound data based on the current data processing pressure of the terminal processor and the current data processing pressure of the cloud server, including: If the credibility exceeds the preset value, the terminal processor calculates the estimated computing power consumption based on the total amount of ECG and heart sound data and the pre-input analysis requirements. The terminal processor calculates the analysis content of the terminal processor and the analysis content of the cloud server based on the current data processing pressure of the terminal processor, the current data processing pressure of the cloud server, and the estimated computing power consumption. The terminal processor calculates the analysis content for both the terminal processor and the cloud server based on the current data processing pressure of the terminal processor, the current data processing pressure of the cloud server, and the estimated computing power consumption, including: If the current data processing pressure of the terminal processor meets the estimated computing power consumption requirement, and the current data processing pressure of the cloud server is less than a preset value, then the terminal processor will split the ECG and heart sound data based on the analysis requirements, the current data processing pressure of the terminal processor, and the current data processing pressure of the cloud server to determine the first type of computational data, the calculation rules of the first type of computational data, the second type of computational data, and the calculation rules of the second type of computational data. The terminal processor carries the first type of computational data and the computational rules of the first type of computational data in the analysis content of the terminal processor; and carries the second type of computational data and the computational rules of the second type of computational data in the analysis content of the cloud server; the ratio of the computational power consumption of executing the computational rules of the first type of computational data to the computational power consumption of executing the computational rules of the second type of computational data is negatively correlated with the ratio of the current data processing pressure of the terminal processor to the current data processing pressure of the cloud server; The method further includes: The terminal processor generates terminal analysis results based on the analysis content of the terminal processor; The terminal processor combines the terminal analysis results with the cloud analysis results to generate the final analysis result.

2. The method according to claim 1, characterized in that, The terminal processor calculates the analysis content for the terminal processor and the analysis content for the cloud server based on the current data processing pressure of the terminal processor, the current data processing pressure of the cloud server, and the estimated computing power consumption, and also includes: If the current data processing pressure of the terminal processor does not meet the estimated computing power consumption, and the current data processing pressure of the cloud server is greater than a preset value, then the terminal processor will split the analysis requirements to determine the preliminary analysis task and the fine analysis task, and extract the preliminary data and fine analysis data from the ECG and heart sound data based on the preliminary analysis task and the fine analysis task respectively; the preliminary data is carried in the analysis content of the terminal processor. The terminal processor generates a preliminary judgment result for the target user based on the preliminary judgment data and the preliminary judgment analysis task; If the initial judgment result does not meet the preset conditions, the terminal processor uses the refined judgment data and refined judgment analysis task to generate the analysis content of the cloud server; if the initial judgment result meets the preset conditions, the terminal processor provides feedback to the target user based on the initial judgment result.

3. The method according to claim 1, characterized in that, Physical condition data includes: mental state data and skin condition data; mental state data is determined based on the following data: cardiac data, facial image features, blood pressure data, and respiratory data; The terminal processor determines the reliability of the electrocardiogram and heart sound data based on the body state data, including: The terminal processor determines the reliability of the electrocardiogram and heart sound data based on the values ​​of the mental state data; The cloud server analyzes the analysis content according to a defined analysis method to determine the cloud analysis results, including: The cloud server determines the calibration offset between electrocardiogram data and heart sound data based on mental state data and skin condition data; The cloud server generates cloud analysis results based on the calibration offset, the first type of calculation data, and the calculation rules of the first type of calculation data.

4. The method according to claim 3, characterized in that, The cloud server determines the calibration offset between electrocardiogram (ECG) data and heart sound data based on mental state data and skin condition data, including: The cloud server determines whether its current data processing load exceeds a preset value. If the current data processing pressure of the cloud server is greater than the preset value, the cloud server will call the pre-trained offset calculation model of the target user according to the target user's attribute information. The cloud server inputs the target user's mental state data and skin condition data into the pre-trained offset calculation model to determine the calibration offset between the electrocardiogram data and the heart sound data; If the current data processing pressure of the cloud server is less than the preset value, the cloud server uses the target user's mental state data and skin condition data to find reference users of the same type as the target user, and uses the reference user's mental state data, skin condition data, reference user's electrocardiogram data and heart sound data to train the initial RNN model, and obtain the trained offset calculation model. The cloud server inputs the target user's mental state data and skin condition data into the pre-trained offset calculation model to determine the calibration offset between the electrocardiogram data and the heart sound data.

5. A cardiac sound monitoring system, characterized in that, include: ECG and heart sound monitoring equipment, physical condition monitoring equipment, terminal processor, and cloud server are installed at the user's location; The terminal processor is communicatively connected to the cloud server, the electrocardiogram and heart sound monitoring device, and the body status monitoring device, respectively. During the first time period, the terminal processor simultaneously acquires the target user's electrocardiogram and heart sound data through the electrocardiogram and heart sound monitoring device, and acquires the target user's body status data through the body status monitoring device; the electrocardiogram and heart sound data includes electrocardiogram data and heart sound data; The terminal processor determines the reliability of the electrocardiogram and heart sound data based on the body status data; If the credibility exceeds a preset value, the terminal processor calculates a first analysis strategy for the ECG and heart sound data based on the current data processing pressure of the terminal processor and the current data processing pressure of the cloud server; the first analysis strategy includes the analysis content of the terminal processor and the analysis content of the cloud server. The terminal processor uploads the analyzed content from the cloud server to the cloud server; The cloud server analyzes the analysis content of the cloud server according to a determined analysis method to determine the cloud analysis results; If the reliability exceeds a preset value, the terminal processor calculates a first analysis strategy for the ECG and heart sound data based on the current data processing pressure of the terminal processor and the current data processing pressure of the cloud server, including: If the credibility exceeds the preset value, the terminal processor calculates the estimated computing power consumption based on the total amount of ECG and heart sound data and the pre-input analysis requirements. The terminal processor calculates the analysis content of the terminal processor and the analysis content of the cloud server based on the current data processing pressure of the terminal processor, the current data processing pressure of the cloud server, and the estimated computing power consumption. The terminal processor calculates the analysis content for both the terminal processor and the cloud server based on the current data processing pressure of the terminal processor, the current data processing pressure of the cloud server, and the estimated computing power consumption, including: If the current data processing pressure of the terminal processor meets the estimated computing power consumption requirement, and the current data processing pressure of the cloud server is less than a preset value, then the terminal processor will split the ECG and heart sound data based on the analysis requirements, the current data processing pressure of the terminal processor, and the current data processing pressure of the cloud server to determine the first type of computational data, the calculation rules of the first type of computational data, the second type of computational data, and the calculation rules of the second type of computational data. The terminal processor carries the first type of computational data and the computational rules of the first type of computational data in the analysis content of the terminal processor; and carries the second type of computational data and the computational rules of the second type of computational data in the analysis content of the cloud server; the ratio of the computational power consumption of executing the computational rules of the first type of computational data to the computational power consumption of executing the computational rules of the second type of computational data is negatively correlated with the ratio of the current data processing pressure of the terminal processor to the current data processing pressure of the cloud server; The terminal processor is also used for: Generate terminal analysis results based on the analysis content of the terminal processor; The terminal analysis results and the cloud analysis results are combined to generate the final analysis result.

6. An electronic device, characterized in that, include: The device includes a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is in operation, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform the steps of the method as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of the method as described in any one of claims 1 to 4.

Images