Electrocardiogram data processing method, wearable device, electronic device and storage medium
By acquiring and processing ECG data in real time and using a time series prediction model, the real-time and interpretability issues of dynamic ECG monitoring are solved, achieving efficient, accurate analysis and real-time feedback of ECG data.
Patent Information
- Application Number
- CN202410269227.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-03-08
- Publication Date
- 2025-09-09
AI Technical Summary
Existing dynamic electrocardiogram monitoring equipment needs to wait until the entire monitoring cycle is completed before feedback is given. It has poor real-time performance, and the high-dimensional and complex data requires professional interpretation, resulting in a poor user experience.
By acquiring the original ECG data of the target object in real time, using time series prediction models such as the SARIMA model to process the data, abnormal events can be predicted and compared in real time to generate abnormal prompt information.
It achieves real-time processing and feedback of ECG data, avoids the delay of waiting for the end of the monitoring cycle, improves user experience and the interpretability of data interpretation, and improves analysis efficiency and accuracy.
Smart Images
Figure CN120605020A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the technical field of wearable devices, and in particular to an electrocardiogram data processing method, a wearable device, an electronic device, and a storage medium. Background Art
[0002] Electrocardiogram (ECG) signals are commonly used in medicine to monitor heart health, but they can only record short-term characteristics of cardiac activity. To more accurately assess heart health, Holter monitoring is often used to record long-term, continuous heart status. However, existing Holter monitoring devices require waiting for the entire monitoring cycle to complete before providing feedback, resulting in poor real-time performance. Furthermore, the large amount of complex, high-dimensional data generated by Holter monitoring requires specialized personnel to interpret, making the results difficult to interpret.
[0003] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute prior art known to ordinary technicians in the field. Summary of the Invention
[0004] The present invention aims to provide a method for processing electrocardiogram data, a wearable device, an electronic device, and a storage medium.
[0005] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by practice of the present disclosure.
[0006] According to a first aspect of the present disclosure, a method for processing electrocardiogram data is provided, the method comprising:
[0007] Obtain the target object's original ECG data in real time;
[0008] Determining predicted ECG data for the current monitoring period based on the original ECG data corresponding to the historical monitoring period and a time series prediction model; the time series prediction model is obtained based on the current state of the target subject and its original ECG data in the current state;
[0009] Determining corresponding abnormal events based on a comparison result between the original ECG data of the current monitoring period and the predicted ECG data;
[0010] Generate abnormal prompt information based on the abnormal event, and send the abnormal prompt information to the target terminal.
[0011] Optionally, before determining predicted ECG data for the current monitoring period based on the original ECG data corresponding to the historical monitoring period and the time series prediction model, the method further includes:
[0012] determining a current state of the target object;
[0013] Determine the model parameters corresponding to the current state based on the original ECG data corresponding to the current state;
[0014] The time series prediction model corresponding to the current state is determined according to the model parameters.
[0015] Optionally, determining the current state of the target object includes:
[0016] In response to receiving monitoring mode information sent by the target terminal, determining a current state of the target object; the monitoring mode information is triggered in response to a mode selection operation on a target control of a target page, the target page being displayed on the target terminal;
[0017] Alternatively, the current state is determined based on a comparison result between the original ECG data within a target time period and a preset ECG template of the target object; the ECG template is obtained based on ECG monitoring results of the target object in different states.
[0018] Optionally, the time series prediction model is a seasonal difference autoregressive moving average model, and the model parameters corresponding to the current state are determined according to the original electrocardiogram data corresponding to the current state, including:
[0019] Determining the seasonal term parameters of the seasonal difference autoregressive moving average model according to the number of monitoring points corresponding to the electrocardiogram monitoring cycle;
[0020] Perform a stability test on the original ECG data corresponding to the current state;
[0021] Determining the difference order of the seasonal difference autoregressive moving average model according to the stationarity test result;
[0022] Determining the moving average order of the seasonal difference autoregressive moving average model according to the autocorrelation function of the original electrocardiogram data corresponding to the current state;
[0023] The autoregressive order of the seasonal difference autoregressive moving average model is determined according to the partial autocorrelation function of the original electrocardiogram data corresponding to the current state.
[0024] Optionally, the method further includes:
[0025] Determining, according to the first information criterion function, a first function value corresponding to each parameter combination; the parameter combination being a combination of an autoregressive order and a moving average order;
[0026] Determining the second function value corresponding to each parameter combination according to the second information criterion function;
[0027] The minimum values of the first function values and the second function values are screened out, and the parameter combination corresponding to the minimum values is used as the autoregressive order and the moving average order of the seasonal difference autoregressive moving average model.
[0028] Optionally, the method further includes:
[0029] storing the raw ECG data of the first phase in a first container, wherein the first phase is a historical monitoring time range corresponding to the current prediction period;
[0030] The predicted ECG data of the second stage is stored in a second container, where the second stage is a prediction time range corresponding to the current prediction period.
[0031] Optionally, the second stage includes multiple ECG monitoring cycles, and the method further includes:
[0032] The predicted ECG data of the current monitoring cycle is determined according to the predicted ECG data corresponding to the plurality of ECG monitoring cycles in the second stage.
[0033] Optionally, the method further includes:
[0034] Determining an abnormal data interval of the current monitoring period based on a comparison result of the original ECG data and the predicted ECG data of the current monitoring period;
[0035] Storing the original ECG data corresponding to the abnormal data interval in the target storage space;
[0036] In response to a designated operation of a target control on a target client, the storage data of the target storage space is sent to a terminal where the target client is located.
[0037] Optionally, the method further includes:
[0038] Marking the abnormal data interval;
[0039] At least one of the marking result, the predicted ECG data, the comparison result, and the original ECG data is sent to the target terminal so that corresponding data is displayed on the target terminal.
[0040] According to a second aspect of the present disclosure, there is provided a wearable device, comprising:
[0041] A data acquisition unit, used for collecting raw ECG data of the target object in real time;
[0042] A data processing unit is used to determine the predicted ECG data of the current monitoring period based on the original ECG data corresponding to the historical monitoring period and a time series prediction model; the time series prediction model is obtained based on the current state of the target object and the original ECG data of the current state; based on the comparison result of the original ECG data of the current monitoring period and the predicted ECG data, the corresponding abnormal event is determined; based on the abnormal event, abnormal prompt information is generated and sent to the target terminal.
[0043] According to a third aspect of the present disclosure, a computer-readable medium is provided, on which a computer program is stored. When the program is executed by a processor, the electrocardiogram data processing method in the above embodiment is implemented.
[0044] According to a fourth aspect of the present disclosure, an electronic device is provided, comprising: one or more processors; a storage device for storing one or more programs, which, when the one or more programs are executed by the one or more processors, enables the one or more processors to implement the electrocardiogram data processing method as in the above-mentioned embodiment.
[0045] As can be seen from the above technical solutions, the electrocardiogram data processing method in the exemplary embodiments of the present disclosure has at least the following advantages and positive effects:
[0046] The ECG data processing method disclosed in the present invention can, on the one hand, obtain the original ECG data of the target object in real time and process and feedback the results thereof, thereby realizing real-time processing and feedback of ECG data over a long period of time, avoiding the situation in which dynamic ECG monitoring needs to wait for the end of the entire monitoring cycle before feedback of the results is provided, and ensuring the real-time feedback of the ECG data results. On the other hand, the ECG data of the target object is predicted by a time series prediction model determined based on the current state of the target object and its original ECG data in the current state, and abnormal events are determined by comparing the original ECG data with the predicted ECG data, thus realizing real-time autonomous discovery and feedback of abnormal events, and sending abnormal prompt information to the target terminal based on the abnormal events, thereby avoiding the problem that users cannot interpret complex ECG data, and the monitoring results are explainable to users, thereby improving the user experience. Furthermore, the data prediction based on the time series prediction model has a small number of parameters and can be deployed on the wearable device side, thereby realizing efficient and accurate data processing and improving the analysis efficiency and accuracy of ECG data.
[0047] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] The accompanying drawings are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification, are used to explain the principles of the present disclosure. Obviously, the drawings described below are only some embodiments of the present disclosure, and those skilled in the art can derive other drawings based on these drawings without inventive effort.
[0049] Figure 1 One of the flowcharts of the electrocardiogram data processing method according to an embodiment of the present disclosure is schematically shown.
[0050] Figure 2 The figure schematically shows an electrocardiogram diagram of an electrocardiogram monitoring cycle according to an embodiment of the present disclosure.
[0051] Figure 3 The figure schematically shows a flow chart of determining a time series prediction model according to an embodiment of the present disclosure.
[0052] Figure 4 The figure schematically shows a flow chart of determining model parameters using a SARIMA model according to an embodiment of the present disclosure.
[0053] Figure 5 A schematic diagram of determining p and q of the ACF and PACF diagrams of the autoregressive model according to an embodiment of the present disclosure is schematically shown.
[0054] Figure 6 A schematic diagram of determining p and q of the ACF and PACF diagrams of the moving average model according to an embodiment of the present disclosure is schematically shown.
[0055] Figure 7 The figure schematically shows a schematic diagram for determining p and q of the ACF and PACF diagrams of the autoregressive moving average model according to an embodiment of the present disclosure.
[0056] Figure 8 The second schematic diagram schematically shows the process of the electrocardiogram data processing method according to an embodiment of the present disclosure.
[0057] Figure 9 The figure schematically shows a block diagram of a wearable device according to an embodiment of the present disclosure.
[0058] Figure 10 The figure schematically shows a module diagram of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION
[0059] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art.
[0060] In addition, the described features, structures or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure can be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. can be adopted. In other cases, well-known methods, devices, implementations or operations are not shown or described in detail to avoid blurring various aspects of the present disclosure.
[0061] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically separate entities. That is, these functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.
[0062] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, while others may be combined or partially combined. Therefore, the actual execution order may vary depending on the actual situation.
[0063] In one embodiment of the present disclosure, a method for processing electrocardiogram data is proposed. Figure 1 Schematic diagram of the process of processing ECG data is shown in FIG. Figure 1 As shown, the ECG data processing method can be applied to wearable devices or various mobile terminals. The method includes at least the following steps:
[0064] Step S110, acquiring the original ECG data of the target object in real time;
[0065] Step S120, determining predicted ECG data for the current monitoring period based on the original ECG data corresponding to the historical monitoring period and a time series prediction model; the time series prediction model is obtained based on the current state of the target object and its original ECG data in the current state;
[0066] Step S130, determining a corresponding abnormal event based on the comparison result of the original ECG data and the predicted ECG data of the current monitoring period;
[0067] Step S140: Generate abnormal prompt information based on the abnormal event, and send the abnormal prompt information to the target terminal.
[0068] The ECG data processing method of the embodiment of the present disclosure can, on the one hand, obtain the original ECG data of the target object in real time and process it and provide feedback on the results, which can achieve real-time processing and feedback of ECG data over a long period of time, avoiding the situation where dynamic ECG monitoring needs to wait for the end of the entire monitoring cycle to provide feedback on the results, and ensuring the real-time feedback of the ECG data results. On the other hand, the ECG data of the target object is predicted by a time series prediction model determined based on the current state of the target object and its original ECG data in the current state. By comparing the original ECG data with the predicted ECG data, abnormal events are determined, and real-time autonomous discovery and feedback of abnormal events are achieved. Abnormal prompt information is sent to the target terminal based on the abnormal events, which can avoid the problem that users cannot interpret complex ECG data. The monitoring results are explainable to users, improving user experience. Furthermore, the data prediction based on the time series prediction model has a small number of parameters and can be deployed on the wearable device side, thereby achieving efficient and accurate data processing and improving the analysis efficiency and accuracy of ECG data.
[0069] In order to make the technical solution of the present disclosure clearer, the following describes the steps of the electrocardiogram data processing method by taking a wearable device as an example.
[0070] In step S110 , the original electrocardiogram data of the target object is acquired in real time.
[0071] In this exemplary embodiment, the target subject may be a subject undergoing ECG function monitoring. Raw ECG data may be collected by wearing a wearable device or a portable ECG monitor with ECG monitoring capabilities. The wearable device may be an ECG monitoring device based on ECG and / or PPG (Photoplethysmograph). The raw ECG data may be collected ECG and / or PPG data, or other forms of raw collected data, which are not limited in this example.
[0072] It should be noted that when the execution subject is a mobile terminal, a network connection can be established between the wearable device and the mobile terminal to transmit the original ECG data collected by the wearable device to the mobile terminal in real time.
[0073] In step S120 , predicted ECG data for the current monitoring period is determined based on the original ECG data corresponding to the historical monitoring period and the time series prediction model.
[0074] In this exemplary embodiment, the historical monitoring cycle refers to a specified number of ECG monitoring cycles before the current time point, for example, the historical monitoring cycle is 10-20 ECG monitoring cycles before the current time point, and the ECG monitoring cycle refers to a complete cardiac cycle of the heart, such as Figure 2 The figure shows the heartbeat waveform (ECG) of one ECG monitoring cycle, which mainly includes the P wave, PR segment, QRS complex, ST segment, and T wave. The current monitoring cycle refers to a specified number of ECG monitoring cycles after the current time point. For example, the current monitoring cycle is 5 ECG monitoring cycles after the current time point. The historical monitoring cycle can be set to at least twice the current monitoring cycle to ensure prediction accuracy.
[0075] In this exemplary embodiment, the time series prediction model is obtained based on the current state of the target object and its original ECG data in the current state. Different time series prediction models can be determined for different states of the target object, that is, one state corresponds to one time series prediction model. The time series prediction model can be obtained by determining the parameters of the initial time series model through the original ECG data. The initial time series model can be various event sequence prediction models. For example, the initial time series model is an ARMA (Auto Regressive Moving Average) model or a SARIMA (Seasonal Auto Regressive Integrated Moving Average Model), or other time series models, which are not limited in this example.
[0076] In some embodiments, prior to performing ECG data prediction, the raw ECG data may be subjected to denoising processing. The denoising process may include filtering to address at least one of noise caused by muscle movement, power supply interference, and respiratory movement. Denoising can ensure data quality and improve subsequent prediction accuracy.
[0077] For example, before predicting the ECG data of the current monitoring period, a time series prediction model may be determined, such as Figure 3 As shown, this can be achieved by following the steps below:
[0078] Step S310: Determine the current state of the target object.
[0079] In this exemplary embodiment, the target object can be divided into states based on the differences in the ECG data of the target object under different situations or at different times. For example, it can be divided into sleeping state, exercise state (which can include states such as jogging and sprinting), walking state (which can include states such as strolling and brisk walking), working state, dining state, etc. This example does not specifically limit the form of state division. The current state can be determined by receiving user input or by currently monitored ECG data. In the case of user input, the state input by the user shall prevail.
[0080] Exemplarily, the current state of the target object is determined in response to receiving monitoring mode information sent by the target terminal; the monitoring mode information is triggered in response to a mode selection operation of a target control on the target page, and the target page is displayed on the target terminal; or, the current state is determined based on the comparison results of the original ECG data within the target time period and the preset ECG template of the target object; the ECG template is obtained based on the ECG monitoring results of the target object in different states.
[0081] In this exemplary embodiment, the target terminal refers to a terminal device bound to the target object's ECG data monitoring device. For example, the target terminal can be the target object's handheld mobile terminal, or it can be the terminal device of the target object's accompanying person or medical staff, or it can be the terminal of other people associated with the target object. This example does not limit this. Monitoring mode information is information related to the monitoring mode, and may include monitoring mode name, identification and other information. The monitoring mode corresponds to the state of the target object. For example, the monitoring mode may include sleep mode, work mode, exercise mode, and so on. Relevant operations for ECG monitoring can be performed in mini-programs or applications related to ECG monitoring on the target terminal to control the working parameters and modes of the monitoring device. For example, the monitoring mode is selected by performing a mode selection operation on the target control on the target page related to ECG monitoring.
[0082] In this exemplary embodiment, the target time period is the time period from the current designated ECG monitoring cycle. For example, the target time period is 3-5 ECG monitoring cycles from the current cycle. The ECG template is obtained based on the ECG monitoring results of the target object in different states, that is, it is generated based on the historical monitoring data of the target object in different states. The ECG template can be a statistical result or an averaged result of the historical monitoring data, which is not limited in this example. In the absence of monitoring mode information, the monitoring device can compare the monitored raw ECG data with the ECG template and use the state corresponding to the matching ECG template as the current state.
[0083] Step S320 , determining the model parameters corresponding to the current state based on the original ECG data corresponding to the current state.
[0084] In this exemplary embodiment, the model parameters of the time series model may include autoregressive order and moving average order, and may also include difference order and / or seasonal term parameters. The model parameters may be obtained by analyzing the original ECG data.
[0085] For example, the time series forecasting model is a seasonal autoregressive moving average model, or SARIMA model. The SARIMA model consists of four components: S (cyclical component), AR (autoregressive model), MA (moving average model), and I (difference). The autoregressive model AR(p) uses a linear combination of random variables at several previous moments to describe a linear regression model of a random variable at a later moment. The autoregressive model reflects the correlation between the current value of the series and p previous values. The moving average model MA(q) uses random interference or forecast errors from the past q periods to linearly express the current forecast value.
[0086] The general expression of the SARIMA model with a period of s is as follows:
[0087]
[0088] In the above formula, Represents a non-periodic autoregressive model (autoregressive operator), (1-α1L s -…-α P L Ps ) represents the periodic autoregressive model (autoregressive characteristic polynomial), Δ, Δ s They represent non-periodic difference and periodic difference of s period, s represents an ECG monitoring cycle; superscript d and D represent non-periodic difference coefficient and periodic difference coefficient respectively, the difference coefficient is used to ensure that the sequence y to be processed t (Original ECG data) is converted into a stationary time series, where the subscripts p and P represent the maximum lag order corresponding to the non-periodic and periodic time series, respectively. (1+θ1L+…+θ q L q ) represents the non-periodic moving average model (moving average operator), (1+β1L s +…+β Q L Qs ) represents the periodic moving average model (moving average characteristic polynomial), u t represents the white noise term, and the subscripts q and Q represent the maximum lag orders of the autoregressive and moving average operators, respectively.
[0089] refer to Figure 4 For the SARIMA model, the model parameters can be determined by the following steps:
[0090] Step S410 : determining the seasonal term parameters of the seasonal difference autoregressive moving average model according to the number of monitoring points corresponding to the ECG monitoring period.
[0091] In this exemplary embodiment, the season item parameter may be a periodic parameter, and the number of monitoring points (number of time points) included in one ECG monitoring cycle (one heart beat cycle of the target object) may be used as the season item parameter (s).
[0092] Step S420: Perform a stability test on the original ECG data corresponding to the current state.
[0093] In this exemplary embodiment, the stationarity of the original ECG data (sequence data) can be detected by a unit root test. If there is no unit root, it means that the sequence is stationary and no differencing is required; if there is a unit root, it indicates that the sequence is non-stationary and sequence differencing is required.
[0094] Step S430: Determine the difference order of the seasonal difference autoregressive moving average model according to the stationarity test result.
[0095] In this exemplary embodiment, a non-stationary sequence can be converted to a stationary sequence through differencing. If the stationary detection result indicates a non-stationary sequence, the original ECG data is differentiated, and then the differentiated sequence is tested for stationarity, and so on, until a stationary sequence data is obtained. The number of differences in this process is called the difference order.
[0096] Step S440 , determining the moving average order of the seasonal difference autoregressive moving average model according to the autocorrelation function of the original ECG data corresponding to the current state.
[0097] Step S450 , determining the autoregressive order of the seasonal difference autoregressive moving average model according to the partial autocorrelation function of the original ECG data corresponding to the current state.
[0098] In this exemplary embodiment, the moving average order and the autoregressive order can be determined by the autocorrelation function ACF diagram and the partial autocorrelation function PACF diagram. Specifically, if the autocorrelation function is tailing (the sequence decreases monotonically or oscillates at an exponential rate), and the partial autocorrelation function is p-order truncated (the sequence value suddenly decays to 0 or close to 0 after a certain moment), it is determined that the sequence data is applicable to the AR(p) autoregressive model (the model contains the AR part but does not contain the MA part), and the autoregressive order is determined to be p and the moving average order is 0, as shown in FIG. Figure 5As shown in the figure, it can be determined that the autoregressive order is p = 2 and the moving average order is q = 0. If the autocorrelation function is truncated at order q and the partial autocorrelation function is trailing, it is determined that the sequence data is suitable for the MA(q) moving average model (the model contains the MA part but not the AR part), that is, the autoregressive order is 0 and the moving average order is q, as shown in Figure 6 As shown in the figure, it can be determined that the autoregressive order is p = 0 and the moving average order is q = 2. If both the autocorrelation function and the partial autocorrelation function have tails, it is determined that the sequence data is suitable for the ARMA model (the model contains AR and MA parts). The tailing order of ACF is the moving average order, and the tailing order of PACF is the autoregressive order, as shown in the figure. Figure 7 As shown, it can be determined from the figure that the autoregressive order is p=2 and the moving average order is q=2.
[0099] above Figure 4 In the illustrated embodiment, the seasonal difference autoregressive moving average model is used to predict ECG data, which can adapt the model to the periodic characteristics of ECG data and improve the model's prediction accuracy for ECG data. At the same time, the number of parameters in the prediction process is small, ensuring real-time performance.
[0100] In some embodiments, when the moving average order and the autoregressive order cannot be determined from the ACF and PACF diagrams, the autoregressive order and the moving average order can also be determined through the following steps.
[0101] The first step is to determine the first function value corresponding to each parameter combination according to the first information criterion function;
[0102] The second step is to determine the second function value corresponding to each parameter combination according to the second information criterion function;
[0103] The third step is to screen out the minimum values of the first function values and the second function values, and use the parameter combination corresponding to the minimum value as the autoregressive order and moving average order of the seasonal difference autoregressive moving average model.
[0104] In this exemplary embodiment, the parameter combination refers to a combination of an autoregressive order and a moving average order; the value ranges of the autoregressive order p and the moving average order q can be set respectively based on experience or ACF and PACF plots, and then different values of p and q are used to form a combination of multiple parameter combinations. For example, if the value ranges of p and q are both 1 to 3, then 9 parameter combinations can be obtained. The first information criterion function can be the Akaike information criterion (AIC), which is a standard for measuring the fitting performance of a statistical model, and its formula is as follows:
[0105] AIC=2k-2ln(T)
[0106] Where k is the number of parameters and T is the likelihood function.
[0107] The second information criterion function may be the Bayesian Information Criterion (BIC), which is used to select a model that best fits the existing data. The formula is:
[0108] BIC=kln(n)-2ln(T)
[0109] Where n is the number of samples.
[0110] The above AIC and BIC formulas introduce a penalty term related to the number of model parameters k, which can prevent the model from overfitting.
[0111] The first function value (AIC value) and the second function value (BIC value) corresponding to different parameter combinations are calculated respectively by the above formula. The minimum value is screened out from all AIC values and BIC values, and the parameter value in the parameter combination corresponding to the minimum value is used as the final autoregressive order and moving average order.
[0112] In this example, when the autoregressive order and moving average order cannot be determined in the ACF and PACF plots, the AIC and BIC criteria are used to determine the parameter values of p and q in the model to improve the model's prediction accuracy.
[0113] Step S330: Determine the time series prediction model corresponding to the current state according to the model parameters.
[0114] In this exemplary embodiment, each model parameter is brought into the initial time series model to obtain the corresponding time series prediction model.
[0115] In step S130 , the corresponding abnormal event is determined based on the comparison result of the original ECG data and the predicted ECG data of the current monitoring period.
[0116] In this exemplary embodiment, it is possible to determine whether the raw ECG data of the current monitoring cycle is abnormal based on the predicted ECG data. In this way, the raw ECG data currently monitored will have the corresponding predicted ECG data as a benchmark for abnormality judgment, thereby achieving real-time rapid discovery of abnormal events and avoiding the waiting period for result feedback. Specifically, the raw ECG data currently monitored can be compared with the predicted ECG data at the corresponding time point. When the difference between the two at a certain time point exceeds a preset threshold, it is determined that the raw ECG data corresponding to the time point is abnormal. Furthermore, the abnormal event can be comprehensively determined based on the position of the abnormal time point within an ECG monitoring cycle, the number of abnormal time points, and the periodic characteristics of the abnormal occurrence.
[0117] In some embodiments, the method further includes: storing the raw ECG data of the first stage in a first container; and storing the predicted ECG data of the second stage in a second container.
[0118] In this example implementation, the first stage is the historical monitoring time range corresponding to the current prediction period (such as the time range corresponding to 10-20 ECG monitoring periods before the current time point); the second stage is the predicted time range of the current prediction period (such as the time range corresponding to 5-10 ECG monitoring periods after the current time point). A first container and a second container can be constructed and maintained on the execution subject side, with the first container storing the original ECG data of the first stage and the second container storing the predicted ECG data of the second stage. As the monitoring process progresses, the data in the two containers can be updated in real time so that the data in the two containers always keep in real-time synchronization with the monitoring process, thereby ensuring the real-time and accuracy of the results. In addition, the two containers only store the corresponding data within the current prediction period, so that the amount of stored data during the processing is small, and a smaller storage space can meet this requirement, so that the execution subject can be a portable device such as a wearable device or a mobile terminal, which makes it easier for users to obtain feedback results in a timely manner and improve the user experience.
[0119] In some embodiments, the second stage includes multiple ECG monitoring cycles, and the method further includes: determining predicted ECG data for the current monitoring cycle based on predicted ECG data corresponding to the multiple ECG monitoring cycles in the second stage.
[0120] In this example embodiment, when updating the predicted ECG data of the second container, the current newly predicted data can be averaged with the predicted data of multiple ECG monitoring cycles stored in the second container. For example, the predicted data of each monitoring cycle can be averaged or weighted averaged, so that the predicted data at each time point after the update includes five prediction results, which can improve the reliability of the prediction results.
[0121] In some embodiments, the method also includes: determining the abnormal data interval of the current monitoring period based on the comparison results of the original ECG data and the predicted ECG data of the current monitoring period; storing the original ECG data corresponding to the abnormal data interval in the target storage space; and sending the storage data of the target storage space to the terminal where the target client is located in response to the specified operation of the target control for the target client.
[0122] In this example embodiment, the interval formed by the monitoring time point where the difference between the original ECG data and the predicted ECG data is greater than a preset threshold can be used as an abnormal data interval, and the original ECG data corresponding to the abnormal data interval can be stored in the target storage space. A storage space specifically for storing abnormal data (target storage space) can be pre-divided in the device, and when abnormal data is found, the abnormal data is stored in the target storage space. An access control (target control) specifically for the target storage space can be configured on the target client (the application client corresponding to the ECG monitoring). In response to a specified operation on the target control on the target client (such as a click, double-click, or other selection operation), the data in the target storage space can be called. The data in the target storage space can be regularly cleaned or updated to save device storage space. For example, the target storage space can be updated in a manner similar to the data update method of the first container and the second container. This example allows other users or doctors to quickly understand the abnormal data of the target object within a certain period through interface operations. At the same time, only the abnormal data of the current monitoring cycle is stored, which can save device storage space.
[0123] In some embodiments, the abnormal data interval is marked, and the marking result is sent to the target terminal for display.
[0124] In this example embodiment, abnormal data intervals can be marked and the marking results (such as an abnormally marked electrocardiogram) can be sent to the target terminal to display the marking results on the target terminal, so that the user can quickly understand the abnormal situation and improve the user experience.
[0125] In step S140, abnormal prompt information is generated based on the abnormal event, and the abnormal prompt information is sent to the target terminal.
[0126] In this exemplary embodiment, corresponding abnormality prompt information can be generated based on the type and cause of the abnormal event to alert the user on the target terminal side of the current abnormal situation. A mapping table of abnormal events and abnormality prompt information can be configured in advance, and abnormality prompt information can be quickly generated based on this mapping table to improve the efficiency of abnormality notification.
[0127] In some embodiments, the method further includes: sending at least one of the predicted ECG data, the comparison result, and the original ECG data to a target terminal, so as to display corresponding data on the target terminal.
[0128] In this example embodiment, at least one of the predicted ECG data, the original ECG data and the comparison result can be displayed on the target terminal, so that the user on the target terminal side can easily grasp the monitoring and comparison status, providing convenience for the user.
[0129] In the present disclosure, all data sent to the target terminal can also be obtained by performing specified operations on the corresponding controls on the corresponding page of the target client, that is, the user can actively obtain the monitoring result data by installing the target client, thereby improving the user experience.
[0130] For example, if Figure 8 As shown, the ECG data processing method disclosed herein can be performed between a wearable device with an ECG monitoring function and a handheld mobile terminal of a target object, and specifically may include the following steps:
[0131] In step S801, the wearable device collects raw ECG data of the target object in real time, and stores the raw ECG data corresponding to a historical monitoring period in a first container.
[0132] Step S802: In response to the user's mode selection operation on the mobile terminal, the current state of the target object is determined, and the mobile terminal sends the current state to the wearable device.
[0133] In step S803, the wearable device determines a corresponding time series prediction model based on the current state of the target object and its original electrocardiogram data in the current state.
[0134] In step S804, the wearable device determines predicted ECG data for the current monitoring period based on the original ECG data corresponding to the historical monitoring period and the time series prediction model, and stores the predicted ECG data in the second container.
[0135] In step S805, the wearable device determines a corresponding abnormal event based on a comparison result between the original ECG data and the predicted ECG data of the current monitoring period.
[0136] In step S806, the wearable device generates abnormal prompt information based on the abnormal event.
[0137] In step S807, the wearable device sends the comparison result, abnormal event, and abnormal prompt information to the mobile terminal.
[0138] Step S808: The mobile terminal displays the comparison results, abnormal events, and abnormal prompt information.
[0139] In the above embodiment, the wearable device can also send the collected original ECG data to the mobile terminal for processing on the mobile terminal side. Figure 8 Data processing in each step.
[0140] The ECG data processing method provided by the present disclosure can realize real-time feedback of dynamic ECG monitoring data; using the time series prediction model, it realizes real-time prediction of monitoring data and performs abnormal analysis based on the prediction results, realizing real-time discovery and feedback of abnormal results; through data prediction and comparative result analysis based on the time series prediction model, it not only ensures the accuracy of the results, but also realizes data processing close to the user's mobile terminal, provides great convenience for users, truly realizes real-time processing and real-time feedback, and greatly improves the user experience.
[0141] The following describes an embodiment of the device disclosed herein, which can be used to execute the ECG data processing method disclosed herein. For details not disclosed in the embodiment of the device disclosed herein, please refer to the embodiment of the ECG data processing method disclosed herein.
[0142] refer to Figure 9 The present disclosure further provides a wearable device 900, which may include:
[0143] The data acquisition unit 910 is used to acquire the original ECG data of the target object in real time;
[0144] The data processing unit 920 is used to determine the predicted ECG data of the current monitoring period based on the original ECG data corresponding to the historical monitoring period and the time series prediction model; the time series prediction model is obtained based on the current state of the target object and its original ECG data in the current state; based on the comparison results of the original ECG data and the predicted ECG data of the current monitoring period, the corresponding abnormal event is determined; based on the abnormal event, abnormal prompt information is generated and the abnormal prompt information is sent to the target terminal.
[0145] In some embodiments of the present disclosure, based on the aforementioned solution, the data processing unit 920 is configured to: determine the current state of the target subject before determining the predicted ECG data for the current monitoring period based on the original ECG data corresponding to the historical monitoring periods and the time series prediction model;
[0146] Determine the model parameters corresponding to the current state based on the original ECG data corresponding to the current state;
[0147] According to the model parameters, the time series prediction model corresponding to the current state is determined.
[0148] In some embodiments of the present disclosure, based on the above solution, the data processing unit 920 is configured to:
[0149] In response to receiving monitoring mode information sent by the target terminal, determining the current state of the target object; the monitoring mode information is triggered in response to a mode selection operation on a target control of a target page, and the target page is displayed on the target terminal;
[0150] Alternatively, the current state is determined based on a comparison result of the original ECG data within the target time period with a preset ECG template of the target object; the ECG template is obtained based on ECG monitoring results of the target object in different states.
[0151] In some embodiments of the present disclosure, based on the aforementioned solution, the time series prediction model is a seasonal difference autoregressive moving average model, and the data processing unit 920 is used to:
[0152] According to the number of monitoring points corresponding to the ECG monitoring cycle, the seasonal term parameters of the seasonal difference autoregressive moving average model are determined;
[0153] Perform a stability test on the original ECG data corresponding to the current state;
[0154] According to the stationarity test results, the difference order of the seasonal difference autoregressive moving average model is determined;
[0155] Determine the moving average order of the seasonal difference autoregressive moving average model according to the autocorrelation function of the original ECG data corresponding to the current state;
[0156] The autoregressive order of the seasonal difference autoregressive moving average model is determined according to the partial autocorrelation function of the original electrocardiogram data corresponding to the current state.
[0157] In some embodiments of the present disclosure, based on the above solution, the data processing unit 920 is further configured to:
[0158] Determining the first function value corresponding to each parameter combination according to the first information criterion function; the parameter combination is a combination of an autoregressive order and a moving average order;
[0159] Determining the second function value corresponding to each parameter combination according to the second information criterion function;
[0160] The minimum values of the first function values and the second function values are screened out, and the parameter combination corresponding to the minimum value is used as the autoregressive order and the moving average order of the seasonal difference autoregressive moving average model.
[0161] In some embodiments of the present disclosure, based on the aforementioned scheme, the wearable device 900 also includes a first container and a second container, the first container is used to store the original ECG data of the first stage, and the first stage is the historical monitoring time range corresponding to the current prediction period; the second container is used to store the predicted ECG data of the second stage, and the second stage is the prediction time range corresponding to the current prediction period.
[0162] In some embodiments of the present disclosure, based on the aforementioned solution, the second stage includes multiple ECG monitoring cycles, and the data processing unit is further configured to:
[0163] The predicted ECG data for the current monitoring period is determined based on the predicted ECG data corresponding to the multiple ECG monitoring periods in the second stage.
[0164] In some embodiments of the present disclosure, based on the above solution, the data processing unit 920 is further configured to:
[0165] Determine the abnormal data interval of the current monitoring period based on the comparison results of the original ECG data and the predicted ECG data of the current monitoring period;
[0166] Storing the original ECG data corresponding to the abnormal data interval in the target storage space;
[0167] In response to a designated operation of a target control on a target client, the storage data of the target storage space is sent to a terminal where the target client is located.
[0168] In some embodiments of the present disclosure, based on the aforementioned solution, the data processing unit 920 is further configured to: mark abnormal data intervals and send the marking results to the target terminal for display.
[0169] In some embodiments of the present disclosure, based on the aforementioned solution, the data processing unit 920 is further configured to send at least one of the predicted ECG data, the comparison result, and the original ECG data to the target terminal so as to display the corresponding data on the target terminal.
[0170] The specific details of the above-mentioned wearable device have been described in detail in the corresponding ECG data processing method, so they will not be repeated here.
[0171] It should be noted that although several modules or units of the device for execution are mentioned in the detailed description above, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.
[0172] In this exemplary embodiment, an electronic device capable of implementing the above method is also provided.
[0173] Those skilled in the art will appreciate that various aspects of the present invention may be implemented as systems, methods, or program products. Therefore, various aspects of the present invention may be implemented in the following forms: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software implementations, which may be collectively referred to herein as "circuits," "modules," or "systems."
[0174] Refer to the following Figure 10The electronic device 1000 according to this embodiment of the present invention will be described. Figure 10 The electronic device 1000 shown is merely an example and should not limit the functions and scope of use of the embodiments of the present invention.
[0175] like Figure 10 As shown, electronic device 1000 is implemented as a general-purpose computing device. Components of electronic device 1000 may include, but are not limited to, the aforementioned at least one processing unit 1010, the aforementioned at least one storage unit 1020, a bus 1030 connecting various system components (including storage unit 1020 and processing unit 1010), and a display unit 1040.
[0176] The storage unit stores program codes, which can be executed by the processing unit 1010, so that the processing unit 1010 performs the steps of various exemplary embodiments of the present invention described in the above-mentioned electrocardiogram data processing method of this specification. For example, the processing unit 1010 can perform the following steps: Figure 1 Step S110 shown in: acquiring the original ECG data of the target object in real time; step S120: determining the predicted ECG data of the current monitoring period based on the original ECG data corresponding to the historical monitoring period and the time series prediction model; the time series prediction model is obtained based on the current state of the target object and its original ECG data in the current state; step S130: determining the corresponding abnormal event based on the comparison result of the original ECG data and the predicted ECG data of the current monitoring period; step S140: generating abnormal prompt information based on the abnormal event, and sending the abnormal prompt information to the target terminal.
[0177] The storage unit 1020 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 10201 and / or a cache memory unit 10202 , and may further include a read-only memory unit (ROM) 10203 .
[0178] The storage unit 1020 may also include a program / utility 10204 having a set (at least one) of program modules 10205, such program modules 10205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.
[0179] Bus 1030 may represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.
[0180] Electronic device 1000 can also communicate with one or more external devices (e.g., a keyboard, pointing device, Bluetooth device, etc.), one or more devices that enable a viewer to interact with electronic device 1000, and / or any device that enables electronic device 1000 to communicate with one or more other computing devices (e.g., a router, modem, etc.). Such communication can occur via input / output (I / O) interface 1050. Furthermore, electronic device 1000 can communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network such as the Internet) via network adapter 1060. As shown, network adapter 1060 communicates with other modules of electronic device 1000 via bus 1030. It should be understood that, although not shown, other hardware and / or software modules can be used in conjunction with electronic device 1000, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.
[0181] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solution according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.
[0182] In this exemplary embodiment, a computer-readable storage medium is also provided, storing a program product capable of implementing the methods described above. In some possible implementations, various aspects of the present invention may also be implemented in the form of a program product comprising program code. When the program product is executed on a terminal device, the program code is configured to cause the terminal device to execute the steps according to various exemplary embodiments of the present invention described in the "Exemplary Methods" section above.
[0183] According to an embodiment of the present invention, a program product for implementing the above-mentioned method can be a portable compact disc read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited thereto. In this document, a readable storage medium can be any tangible medium containing or storing a program, and the program can be used by or in conjunction with an instruction execution system, apparatus, or device.
[0184] The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, or any combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.
[0185] A computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium that can transmit, propagate, or transfer a program for use by or in conjunction with an instruction execution system, apparatus, or device.
[0186] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
[0187] The program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, and the like, as well as conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a stand-alone software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., via the Internet using an Internet service provider).
[0188] Furthermore, the figures above are merely illustrative of the processes included in the methods according to exemplary embodiments of the present disclosure and are not intended to be limiting. It is readily understood that the processes illustrated in the figures above do not indicate or limit the temporal order of these processes. Furthermore, it is readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.
[0189] Other embodiments of the present disclosure will readily occur to those skilled in the art after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow from the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the present disclosure being indicated by the claims.
[0190] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes can be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims
1. A method for processing electrocardiogram data, characterized in that: include: Obtain the target object's original ECG data in real time; Determining predicted ECG data for a current monitoring period based on the original ECG data corresponding to historical monitoring periods and a time series prediction model obtained based on the current state of the target subject and its original ECG data in the current state; Determining corresponding abnormal events based on a comparison result between the original ECG data of the current monitoring period and the predicted ECG data; Generate abnormal prompt information based on the abnormal event, and send the abnormal prompt information to the target terminal.
2. The method according to claim 1, characterized in that Before determining predicted ECG data for the current monitoring period based on the original ECG data corresponding to the historical monitoring period and the time series prediction model, the method further includes: determining a current state of the target object; Determine the model parameters corresponding to the current state based on the original ECG data corresponding to the current state; The time series prediction model corresponding to the current state is determined according to the model parameters.
3. The method according to claim 2, characterized in that Determining the current state of the target object includes: In response to receiving monitoring mode information sent by the target terminal, determining a current state of the target object; the monitoring mode information is triggered in response to a mode selection operation on a target control of a target page, the target page being displayed on the target terminal; Alternatively, the current state is determined based on a comparison result between the original ECG data within a target time period and a preset ECG template of the target object; the ECG template is obtained based on ECG monitoring results of the target object in different states.
4. The method according to claim 2, characterized in that The time series prediction model is a seasonal difference autoregressive moving average model. The model parameters corresponding to the current state are determined based on the original ECG data corresponding to the current state, including: Determining the seasonal term parameters of the seasonal difference autoregressive moving average model according to the number of monitoring points corresponding to the electrocardiogram monitoring cycle; Perform a stability test on the original ECG data corresponding to the current state; Determining the difference order of the seasonal difference autoregressive moving average model according to the stationarity test result; Determining the moving average order of the seasonal difference autoregressive moving average model according to the autocorrelation function of the original electrocardiogram data corresponding to the current state; The autoregressive order of the seasonal difference autoregressive moving average model is determined according to the partial autocorrelation function of the original electrocardiogram data corresponding to the current state.
5. The method according to claim 4, characterized in that The method further comprises: Determining, according to the first information criterion function, a first function value corresponding to each parameter combination; the parameter combination being a combination of an autoregressive order and a moving average order; Determining the second function value corresponding to each parameter combination according to the second information criterion function; The minimum values of the first function values and the second function values are screened out, and the parameter combination corresponding to the minimum values is used as the autoregressive order and the moving average order of the seasonal difference autoregressive moving average model.
6. The method according to claim 1, characterized in that The method further comprises: storing the raw ECG data of the first phase in a first container, wherein the first phase is a historical monitoring time range corresponding to the current prediction period; The predicted ECG data of the second stage is stored in a second container, where the second stage is a prediction time range corresponding to the current prediction period.
7. The method according to claim 6, characterized in that The second stage includes multiple ECG monitoring cycles, and the method further includes: The predicted ECG data of the current monitoring cycle is determined according to the predicted ECG data corresponding to the plurality of ECG monitoring cycles in the second stage.
8. The method according to claim 1, characterized in that The method further comprises: Determining an abnormal data interval of the current monitoring period based on a comparison result of the original ECG data and the predicted ECG data of the current monitoring period; Storing the original ECG data corresponding to the abnormal data interval in the target storage space; In response to a designated operation of a target control on a target client, the storage data of the target storage space is sent to a terminal where the target client is located.
9. The method according to claim 8, characterized in that The method further comprises: Marking the abnormal data interval; At least one of the marking result, the predicted ECG data, the comparison result, and the original ECG data is sent to the target terminal so that corresponding data is displayed on the target terminal.
10. A wearable device, characterized in that: include: A data acquisition unit, used for collecting raw ECG data of the target object in real time; a data processing unit configured to determine predicted ECG data for a current monitoring period based on the original ECG data corresponding to historical monitoring periods and a time series prediction model obtained based on the current state of the target subject and its original ECG data in that state; and to determine a corresponding abnormal event based on a comparison result between the original ECG data of the current monitoring period and the predicted ECG data; Generate abnormal prompt information based on the abnormal event, and send the abnormal prompt information to the target terminal.