Remote electrocardiogram data monitoring and analyzing method, device and equipment for department of cardiology and medium

Abstract

The invention relates to a remote electrocardiogram data monitoring and analyzing method, device and equipment for the department of cardiology and a medium. The method comprises the following steps: acquiring electrocardiogram data, and performing lead channel matching on the electrocardiogram data to obtain lead channel marked electrocardiogram data; inputting the lead channel marked electrocardio data of each lead channel into each lead electrocardio anomaly identification model corresponding to the lead channel in a lead electrocardio anomaly identification model group to obtain electrocardio anomaly identification information of the lead channel with abnormal electrocardiogram data; remote electrocardio preliminary diagnosis information and remote electrocardio weight coefficient information corresponding to the remote electrocardio preliminary diagnosis information are generated based on the electrocardio anomaly identification information of all the lead channels; and according to the remote electrocardiogram weight coefficient information, adjusting data transmission parameters of the lead channel marked electrocardiogram data and acquisition precision parameters of the electrocardiogram data. By adopting the method, collaborative optimization of diagnosis precision, transmission efficiency and equipment efficiency can be realized through lead cascade processing and resource regulation and control.

Description

Technical Field

[0001] The present invention belongs to the technical field of medical equipment, and in particular relates to a cardiology remote electrocardiogram data monitoring and analysis method, device, equipment and medium. Background Art

[0002] In recent years, cardiovascular disease has become one of the leading causes of death due to its rapid onset, high mortality, numerous complications, and high recurrence rates. Therefore, the prevention and treatment of cardiovascular disease has become extremely important and challenging. Electrocardiogram (ECG) testing plays a crucial role in the diagnosis of cardiovascular disease. However, traditional ECG testing performed in hospital emergency departments and outpatient clinics lacks real-time and timeliness, making it difficult to overcome the challenges of remote monitoring and diagnosis. Therefore, 24 / 7 Holter monitoring and real-time online ECG data diagnosis for suspected and confirmed cardiovascular patients are urgently needed.

[0003] Furthermore, traditional remote ECG monitoring equipment is susceptible to external interference when collecting data, affecting its accuracy and usability. Furthermore, due to the numerous lead wires, electrodes can easily fall off, potentially resulting in the loss of critical ECG information and hindering doctors' accurate interpretation of the ECG. Furthermore, when relying on a network for data transmission, unstable network signals can cause delays or even loss of ECG data, further impacting the timeliness and reliability of remote diagnosis.

[0004] In the prior art with authorization announcement number CN108512629B, a method and system for sending, receiving and controlling ECG monitoring data are provided. However, this prior art can only realize the transmission rate control of single-lead ECG data, and lacks abnormality identification, weight calculation and differentiated parameter adjustment mechanism for multiple lead channels. Summary of the Invention

[0005] Based on this, it is necessary to provide a cardiology remote ECG data monitoring and analysis method, device, equipment and medium that can achieve lead-level collaborative optimization of diagnostic accuracy, transmission efficiency and equipment effectiveness in response to the above technical problems.

[0006] In a first aspect, the present application provides a cardiology remote electrocardiogram (ECG) data monitoring and analysis method, comprising:

[0007] Acquire electrocardiogram data, and perform lead channel matching on the electrocardiogram data to obtain lead channel marked electrocardiogram data;

[0008] Inputting the lead channel marked ECG data of each lead channel into each lead ECG abnormality recognition model corresponding to the lead channel in the lead ECG abnormality recognition model group to obtain ECG abnormality recognition information of the lead channel with abnormal ECG data;

[0009] Generate remote ECG preliminary diagnosis information and remote ECG weight coefficient information corresponding to the remote ECG preliminary diagnosis information based on ECG abnormality recognition information of each lead channel;

[0010] The data transmission parameters of the lead channel marked electrocardiogram data and the acquisition accuracy parameters of the electrocardiogram data are adjusted according to the remote electrocardiogram weight coefficient information. The data transmission parameters are used to regulate the data quality and / or transmission efficiency of the lead channel marked electrocardiogram data.

[0011] In one embodiment, the cardiology remote ECG data monitoring and analysis method further includes:

[0012] Conducting lead channel demand analysis on the remote ECG preliminary diagnosis information to generate lead channel guidance status information, which is used to represent the working status of ECG sensing devices of each lead channel required for ECG data monitoring and analysis corresponding to the remote ECG preliminary diagnosis information;

[0013] Performing lead channel status recognition on the lead channel marked ECG data to generate lead channel actual status information, which is used to represent the actual working status of the ECG sensing device of each lead channel;

[0014] generating lead channel state abnormality information and lead channel state abnormality level information corresponding to the lead channel state abnormality information based on the lead channel guidance state information and the lead channel actual state information;

[0015] If the abnormal level information exceeds the preset lead channel status abnormal level threshold, the corresponding lead channel status adjustment guidance information is generated based on the lead channel status abnormal information. The lead channel status adjustment guidance information is used to guide the remote ECG data monitoring operator of the cardiology department to adjust the actual working status of the ECG sensor equipment of the lead channel.

[0016] In one embodiment, generating remote ECG preliminary diagnosis information and remote ECG weight coefficient information corresponding to the remote ECG preliminary diagnosis information based on ECG abnormality identification information of each lead channel includes:

[0017] Acquire lead channel template electrocardiogram data of each lead channel in which electrocardiogram data of the electrocardiogram abnormality identification information is abnormal;

[0018] Based on the comparison of the lead channel template electrocardiogram data with the lead channel marked electrocardiogram data, the electrocardiogram data of each lead channel with abnormal electrocardiogram data is generated, and the lead electrocardiogram template conformity score data is compared;

[0019] According to the ECG data, the lead ECG template conformity score data is compared with the ECG abnormality recognition information to calculate the comprehensive conformity score and remote ECG weight coefficient information;

[0020] If the comprehensive compliance score is greater than or equal to the comprehensive compliance threshold, the ECG abnormality identification information is used as the remote ECG preliminary diagnosis information;

[0021] If the comprehensive compliance score is less than the comprehensive compliance threshold, each ECG abnormality identification information is input into the joint lead ECG abnormality identification model according to the corresponding lead channel where the ECG data has abnormalities, and remote ECG preliminary diagnosis information and joint remote ECG weight coefficient information are generated. The joint remote ECG weight coefficient information is used to correct and update the remote ECG weight coefficient information.

[0022] In one embodiment, the ECG anomaly identification information includes ECG anomaly severity score data, and the calculation formula for the remote ECG weight coefficient information is:

[0023]

[0024] Where, RCM is the remote ECG weight coefficient information, N is the total number of lead channels with abnormal ECG data, ΔQ is the lead device data accuracy weight coefficient of the i-th lead channel where the ECG data is abnormal. i D is the data quality offset coefficient of the i-th lead channel where the ECG data has abnormalities. i and are the actual working state label information and the actual working state fuzzy scoring function of the ECG sensor device for the i-th lead channel where the ECG data has abnormalities, R is the weighted sum of the ECG data of the i-th lead channel with abnormal ECG data, the ECG data of the lead channel template ECG data of all lead channels with abnormal ECG data, and the ECG template conformity score data. i N is the ECG abnormality severity score data of the i-th lead channel with abnormal ECG data, * is the total number of lead channel template ECG data that is equal to the total number N of lead channels with abnormal ECG data. and They are respectively the ECG data comparison between the i-th lead channel with abnormal ECG data and the lead channel template ECG data of the j-th lead channel with abnormal ECG data, the ECG data comparison between the lead ECG template conformity score data, and the weights of the ECG data comparison between the lead ECG template conformity score data.

[0025] In one embodiment, the lead ECG anomaly recognition model group includes a lead ECG anomaly classification sub-model group and a lead ECG anomaly recognition sub-model group. The lead channel labeled ECG data of each lead channel is input into each lead ECG anomaly recognition model corresponding to the lead channel in the lead ECG anomaly recognition model group to obtain ECG anomaly recognition information of the lead channel with abnormal ECG data, including:

[0026] Acquire photoplethysmography data;

[0027] Based on the photoplethysmography data, the lead channel labeled electrocardiogram data is time-series calibrated and image segmented to obtain a lead channel labeled single electrocardiogram dataset for each lead channel;

[0028] Inputting the lead channel labeled single ECG dataset into the lead ECG abnormality classification sub-model corresponding to each lead channel in the lead ECG abnormality classification sub-model group, performing ECG abnormality image screening, and obtaining an abnormal single ECG dataset and a normal single ECG dataset;

[0029] The abnormal single electrocardiogram data set is input into the lead electrocardiogram abnormality analysis sub-model in the lead electrocardiogram abnormality recognition sub-model group that matches the lead channel corresponding to the abnormal single electrocardiogram data set to generate electrocardiogram abnormality recognition information.

[0030] In one embodiment, the cardiology remote ECG data monitoring and analysis method further includes:

[0031] Input the normal single ECG data set of each lead channel into the normal ECG image fusion model group to generate the normal ECG fusion image of each lead channel;

[0032] Verify the normal single ECG data set based on the normal ECG fusion image to identify the ECG data that deviates from the normal state. The ECG data that deviates from the normal state is used to represent the single ECG data in the normal single ECG data set that has statistically significant differences from the normal ECG fusion image.

[0033] removing the deviated-normal electrocardiogram data from the normal single electrocardiogram data set to obtain a calibrated normal single electrocardiogram data set, and constructing a deviated-normal electrocardiogram data set based on the deviated-normal electrocardiogram data;

[0034] Generate and update cardiology department remote electrocardiogram data monitoring medical history information based on abnormal single electrocardiogram data set, calibrated normal single electrocardiogram data set and deviated normal electrocardiogram data set.

[0035] In one embodiment, the cardiology remote ECG data monitoring and analysis method further includes:

[0036] Obtain and initialize the initial digital twin model of the heart;

[0037] If the remote ECG weight coefficient information exceeds a preset remote ECG weight coefficient threshold, selecting ECG abnormality core identification information that matches the remote ECG preliminary diagnosis information from the ECG abnormality identification information of each lead channel;

[0038] Generate visual annotation information of cardiac abnormalities based on the core recognition information of ECG abnormalities;

[0039] The visual annotation information of cardiac abnormalities is annotated on the initial digital twin model of the heart, the initial digital twin model of the heart is updated, and a digital twin model of cardiac abnormalities annotated is generated.

[0040] In a second aspect, the present application also provides a cardiology remote electrocardiogram data monitoring and analysis device, comprising:

[0041] ECG data acquisition module, used to acquire ECG data and perform lead channel matching on the ECG data to obtain lead channel marked ECG data;

[0042] The ECG abnormality recognition module is used to input the lead channel marked ECG data of each lead channel into each lead ECG abnormality recognition model corresponding to the lead channel in the lead ECG abnormality recognition model group to obtain ECG abnormality recognition information of the lead channel with abnormal ECG data;

[0043] A remote preliminary diagnosis module, configured to generate remote ECG preliminary diagnosis information and remote ECG weight coefficient information corresponding to the remote ECG preliminary diagnosis information based on ECG abnormality identification information of each lead channel;

[0044] The remote parameter adjustment module is used to adjust the data transmission parameters of the lead channel marked electrocardiogram data and the acquisition accuracy parameters of the electrocardiogram data according to the remote electrocardiogram weight coefficient information. The data transmission parameters are used to regulate the data quality and / or transmission efficiency of the lead channel marked electrocardiogram data.

[0045] In a third aspect, the present application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method described in any one of the first aspects of the present application when executing the computer program.

[0046] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any one of the first aspects of the present application.

[0047] The above-mentioned cardiology remote ECG data monitoring and analysis method, device, computer equipment and storage medium can accurately locate the source of abnormalities and improve the accuracy of remote ECG monitoring through lead channel matching; by constructing independent abnormality recognition models for lead channels, it can capture the specific myocardial electrical activity of different lead channels, thereby effectively avoiding the misjudgment of complex arrhythmias; through the dynamic weight control mechanism, it can dynamically control the data quality and transmission efficiency of ECG monitoring data, thereby giving priority to ensuring the integrity and accuracy of key lead data under limited transmission conditions, and supporting real-time decision-making in remote emergency scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0049] Figure 1 A schematic diagram of an application environment for a remote ECG data monitoring and analysis method for cardiology provided in one embodiment of the present application;

[0050] Figure 2 A flowchart of a remote ECG data monitoring and analysis method for cardiology provided in one embodiment of the present application;

[0051] Figure 3 A flowchart of another method for remote ECG data monitoring and analysis in cardiology provided in one embodiment of the present application;

[0052] Figure 4 A flowchart of a remote preliminary diagnosis method provided in one embodiment of the present application;

[0053] Figure 5 A flowchart of a method for identifying electrocardiogram abnormalities provided in one embodiment of the present application;

[0054] Figure 6 A schematic structural diagram of a remote ECG data monitoring and analysis device for cardiology provided in one embodiment of the present application. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0056] The remote ECG data monitoring and analysis method for cardiology provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the sensing device 102 and the diagnosis and treatment terminal 103 can communicate with the server 104 through the network. The database 104 can store the data that the server 101 needs to process. The database 104 can be integrated on the server 101, or it can be placed on the cloud or other network servers. The sensing device 102 can collect the user's remote ECG data and send the data to the server 101 through the network. The server 101 can store the user's remote ECG data collected by the sensing device 102 in the database, and analyze the user's remote ECG data collected by the sensing device 102 through the cardiology remote ECG data monitoring and analysis application installed on the server 101 to generate a cardiology remote ECG data monitoring and analysis report. The diagnosis and treatment terminal 103 can obtain the cardiology remote ECG data monitoring and analysis report generated by the server 101 through the network. The diagnosis and treatment terminal 103 may be, but is not limited to, various desktop computers, all-in-one computers, laptops, smartphones, tablet computers, and medical computing platforms. The sensor device 102 may be, but is not limited to, an IoT sensor device and a portable wearable device. The portable wearable device may be a smart watch, a smart bracelet, a smart ECG monitoring device, etc. The server 101 may be implemented as a standalone server or a server cluster consisting of multiple servers.

[0057] In an exemplary embodiment of the present application, Figure 2 As shown, a remote ECG data monitoring and analysis method for cardiology is provided. Figure 1 The server 101 in the example is used to illustrate the process, which includes the following steps S201 to S204.

[0058] Step S201 : acquiring electrocardiogram data, and performing lead channel matching on the electrocardiogram data to obtain lead channel labeled electrocardiogram data.

[0059] Specifically, the server 101 may obtain remote electrocardiogram data of the user collected by the sensor device via the network, and perform lead channel matching on the electrocardiogram data using a lead channel matching application installed on the server 101 to obtain lead channel labeled electrocardiogram data. The lead channels may include, but are not limited to, standard limb leads, unipolar chest leads, and unipolar limb leads.

[0060] Schematically, lead I of the standard limb lead obtains the ECG signal from the left and right upper limbs, mainly reflecting the depolarization process of the left ventricle; lead II of the standard limb lead obtains the signal from the right upper limb and the left lower limb, which can better reflect the electrical activity of the left ventricle; lead III of the standard limb lead obtains the signal from the left upper limb and the left lower limb, which has a certain reflection of the electrical activity of the left ventricle. Due to the relationship between its projection angle, the waveform characteristics are different from those of lead II; unipolar chest lead The V1 lead of the unipolar chest lead is close to the right ventricle and can reflect the depolarization and repolarization process of the right ventricle. The QRS complex of the V1 lead is mainly rS type, and the R wave amplitude is relatively low; the V2 lead of the unipolar chest lead is located at the fourth intercostal space on the left side of the sternum and can reflect the electrical activity of the left and right ventricles. The QRS complex of the V2 lead is usually mainly rS type or Rs type; the V3 lead of the unipolar chest lead is located at the midpoint of the line connecting the V2 lead and the V4 lead and is more sensitive to the electrical activity of the left ventricle. The QRS complex of the 3 leads has various shapes, which can be, but not limited to, rS, Rs, qR, etc.; the V4 lead of the unipolar chest lead is located at the fifth intercostal space on the left midclavicular line, which can clearly show the electrical activity of the left ventricle. The QRS complex of the V4 lead is usually mainly R-type, and the R wave amplitude is relatively high; the V5 lead of the unipolar chest lead is located at the fifth intercostal space on the left anterior axillary line, which mainly reflects the electrical activity of the left ventricle. The QRS complex of the V5 lead is usually mainly R-type, and the R wave amplitude is relatively high. High; the V6 lead of the unipolar chest lead is located in the fifth intercostal space of the left mid-axillary line, which can reflect the electrical activity of the left ventricle. The QRS complex of the V6 lead is usually R-type, and the R wave amplitude is relatively high; the aVR lead of the unipolar limb lead is the right arm unipolar lead, which can reflect the electrical activity of the right ventricle; the aVL lead of the unipolar limb lead is the left arm unipolar lead, which can reflect the electrical activity of the left ventricle; the aVF lead of the unipolar limb lead is the left leg unipolar lead, which is more sensitive to the observation of the inferior wall myocardium.

[0061] Step S202: inputting the lead channel marked ECG data of each lead channel into each lead ECG abnormality recognition model corresponding to the lead channel in the lead ECG abnormality recognition model group to obtain ECG abnormality recognition information of the lead channel with abnormal ECG data.

[0062] Specifically, the server 101 can input the lead channel marked ECG data of each lead channel into each lead channel ECG abnormality recognition model corresponding to the lead channel in the lead channel ECG abnormality recognition model group carried by the server 101. If the lead channel marked ECG data of the lead channel carried by the server 101 is recognized to be abnormal, the server 101 can generate ECG abnormality recognition information of the lead channel where the ECG data is abnormal. If the lead channel marked ECG data of the lead channel carried by the server 101 is recognized to be normal, the server 101 can store the lead channel marked ECG data in the database.

[0063] Optionally, the lead ECG abnormality recognition model can be trained based on a joint neural network model constructed by combining an improved recurrent convolutional neural network with an improved autoencoder.

[0064] Step S203 : generating remote ECG preliminary diagnosis information and remote ECG weight coefficient information corresponding to the remote ECG preliminary diagnosis information based on the ECG abnormality identification information of each lead channel.

[0065] Specifically, the server 101 may generate remote ECG preliminary diagnosis information based on the ECG abnormality identification information of the abnormal lead channel generated by the lead ECG abnormality identification model according to a preset remote ECG preliminary diagnosis strategy, and calculate the remote ECG weight coefficient information corresponding to the remote ECG preliminary diagnosis information in combination with the remote ECG weight coefficient information generation algorithm. The remote ECG weight coefficient information may include, but is not limited to, a remote ECG critical weight coefficient for characterizing the severity of the disease corresponding to the ECG data abnormality and a remote ECG lead weight coefficient for characterizing the importance of the data of each lead channel corresponding to the remote ECG preliminary diagnosis information.

[0066] Optionally, the remote ECG preliminary diagnostic information may include, but is not limited to, atrial fibrillation, intermittent bundle branch block, intermittent preexcitation, intraventricular aberrant conduction, and ventricular escape.

[0067] Step S204 : adjusting the data transmission parameters of the lead channel marked electrocardiogram data and the acquisition accuracy parameters of the electrocardiogram data according to the remote electrocardiogram weight coefficient information.

[0068] Specifically, the server 101 may adjust the data transmission parameters of the lead-channel-marked ECG data for all lead channels and the acquisition accuracy parameters of the ECG data for all lead channels based on the remote ECG critical weight coefficient. The server 101 may also adjust the data transmission parameters of the lead-channel-marked ECG data for a specific lead channel and the acquisition accuracy parameters of the ECG data for the specific lead channel based on the remote ECG lead weight coefficient.

[0069] Illustratively, data transmission parameters can be used to regulate the data quality and / or transmission efficiency of ECG data labeled on lead channels. ECG data acquisition accuracy parameters can be used to improve acquisition accuracy and obtain more detailed, accurate, and reliable ECG information when suspected abnormalities occur on lead channels.

[0070] Optionally, data transmission parameters of the lead channel labeled ECG data may include, but are not limited to, sampling rate, sampling size, packet size, signal-to-noise ratio, carrier frequency, data rate, bandwidth, and transmission power.

[0071] In the above-mentioned remote ECG data monitoring and analysis method for cardiology, by acquiring ECG data and matching the lead channels, the information of different lead channels can be integrated to achieve multimodal data fusion, which can help to comprehensively evaluate the electrical activity of the heart and improve the accuracy of diagnosis; by combining the ECG abnormality recognition model with the labeled ECG data of each lead channel, the ECG abnormality characteristics can be accurately identified, and the information of different lead channels can be integrated to comprehensively evaluate the electrical activity of the heart, reduce the possibility of missed diagnosis and misdiagnosis, and achieve accurate diagnosis; by adjusting the data transmission parameters of the lead channel labeled ECG data according to the remote ECG weight coefficient information, more transmission resources can be allocated to the lead channels with higher weights to ensure high-quality transmission of key data, which can help to improve the efficiency and reliability of data transmission and reduce the risk of loss and damage of important data; by adjusting the acquisition accuracy parameters of the ECG data, the ECG data acquisition process can be dynamically optimized according to the remote ECG weight coefficient information, thereby improving the reliability of diagnosis.

[0072] In an optional embodiment of the present application, please refer to Figure 3 , the cardiology department remote electrocardiogram data monitoring and analysis method may also include:

[0073] Step S305 , performing lead channel demand analysis on the remote ECG preliminary diagnosis information, and generating lead channel guidance status information.

[0074] Specifically, the lead channel guidance status information may be used to represent the working status of the ECG sensing devices of each lead channel required for monitoring and analyzing ECG data corresponding to the remote ECG preliminary diagnosis information.

[0075] Step S306: performing lead channel status recognition on the lead channel marked ECG data to generate lead channel actual status information.

[0076] Specifically, the lead channel actual state information is used to represent the actual working state of the electrocardiographic sensor device of each lead channel.

[0077] Step S307 : generating lead channel state abnormality information and lead channel state abnormality level information corresponding to the lead channel state abnormality information based on the lead channel guidance state information and the lead channel actual state information.

[0078] Step S308: If the abnormality level information exceeds the preset lead channel state abnormality level threshold, corresponding lead channel state adjustment guidance information is generated according to the lead channel state abnormality information.

[0079] Specifically, the lead channel state adjustment guidance information is used to guide a remote ECG data monitoring operator of the cardiology department to adjust the actual working state of the ECG sensor device of the lead channel.

[0080] For example, using a lead electrode detachment as an example, the lead channel status abnormality information may include, but is not limited to, lead electrode detachment abnormality identification information for the lead channel with the detached lead electrode. In this case, the lead channel status adjustment guidance information can be used to guide the operator in placing the electrode of the detached lead channel in a specified manner and at a specified location to ensure normal operation of the lead channel.

[0081] In the above-mentioned remote ECG data monitoring and analysis method for cardiology, by performing status recognition on the ECG data marked on the lead channels, the working status of each lead channel and its corresponding sensor device can be accurately detected, and then the abnormal working status of the sensor device can be discovered in time and targeted adjustments can be made to ensure that the ECG sensor device can operate in accordance with the expected working status, guarantee the accuracy and completeness of the remote ECG data, and avoid misdiagnosis and missed diagnosis due to poor equipment status.

[0082] Furthermore, when an abnormality occurs in the lead channel, the above-mentioned remote ECG data monitoring and analysis method for cardiology can quickly and accurately locate the fault information, provide operators with clear fault handling directions, reduce the interruption time of patient monitoring, and enhance the stability and reliability of the remote ECG monitoring system.

[0083] In an optional embodiment of the present application, Figure 4 As shown, generating remote ECG preliminary diagnosis information and remote ECG weight coefficient information corresponding to the remote ECG preliminary diagnosis information based on ECG abnormality identification information of each lead channel may include:

[0084] Step S401 : obtaining lead channel template electrocardiogram data of a lead channel having abnormal electrocardiogram data in electrocardiogram abnormality identification information.

[0085] Step S402 : comparing the lead channel labeled ECG data with the lead channel template ECG data to generate ECG data comparison lead ECG template conformity score data for each lead channel having abnormal ECG data of each ECG abnormality identification information.

[0086] Specifically, the server can compare the lead channel marked ECG data based on the lead channel template ECG data in combination with preset comparison rules to generate ECG data comparison lead ECG template conformity score data for each lead channel where the ECG data of each ECG abnormality identification information has abnormalities.

[0087] Optionally, the server can input the lead channel template ECG data and the lead channel marked ECG data into the lead ECG data comparison model mounted on the server to generate ECG data comparison lead ECG template conformity score data for each lead channel where the ECG data of each ECG abnormality identification information has abnormalities.

[0088] Step S403 , a comprehensive conformity score and remote ECG weight coefficient information are calculated based on the ECG data comparison lead ECG template conformity score data and the ECG abnormality recognition information.

[0089] Specifically, the server can construct a conformity scoring matrix based on the ECG data comparison lead ECG template conformity scoring data, and generate a conformity scoring weight matrix of the conformity scoring matrix based on the ECG abnormality identification information. The server can calculate the comprehensive conformity score based on the Hadamard product of the conformity scoring matrix and the conformity scoring weight matrix. The server can also calculate the remote ECG weight coefficient information based on the ECG data comparison lead ECG template conformity scoring data and ECG abnormality identification information combined with the channel data quality information and the channel device working status information.

[0090] For example, taking the case where the number of lead channels with abnormal electrocardiogram data is three as an example, the conformity scoring matrix Conformity scoring weight matrix and overall compliance score The expression can be:

[0091]

[0092] Where, is the conformity scoring matrix, is the conformity score weight matrix, N is the total number of lead channels with abnormal ECG data, N * is the total number of lead channel template ECG data that is equal to the total number N of lead channels with abnormal ECG data. The ECG data of the i-th lead channel with abnormal ECG data is compared with the lead channel template ECG data of the j-th lead channel with abnormal ECG data, and the ECG data is compared with the lead ECG template conformity score data. is the weight of the ECG data comparison between the lead channel template ECG data of the i-th lead channel with abnormal ECG data and the lead ECG template conformity score data of the j-th lead channel with abnormal ECG data, Based on the compliance scoring matrix and conformity score weight matrix The comprehensive conformity score is obtained, ⊙ is the Hadamard product, ||·|| F is the Frobenius norm.

[0093] Step S404: If the comprehensive conformity score is greater than or equal to the comprehensive conformity threshold, the ECG abnormality identification information is used as remote ECG preliminary diagnosis information.

[0094] Step S405: If the comprehensive conformity score is less than the comprehensive conformity threshold, each ECG abnormality identification information is input into the joint lead ECG abnormality identification model according to the lead channel where the corresponding ECG data has abnormalities, and remote ECG preliminary diagnosis information and joint remote ECG weight coefficient information are generated.

[0095] Specifically, the combined remote ECG weight coefficient information can be used to correct and update the remote ECG weight coefficient information.

[0096] Optionally, a joint lead ECG abnormality recognition model can be constructed based on a transformer model.

[0097] In the above-mentioned remote ECG data monitoring and analysis method for cardiology, by generating conformity score data, the difference between the ECG abnormality identification information and the preset template can be accurately quantified, which can not only improve the accuracy of diagnosis, but also provide a reliable quantitative basis for subsequent diagnostic decisions; by calculating the comprehensive conformity score and remote ECG weight coefficient information, a comprehensive evaluation of ECG abnormalities can be achieved, which helps to identify the most representative ECG abnormalities in multiple lead channels, thereby improving the reliability of diagnosis; by introducing a joint lead ECG abnormality recognition model, the ECG abnormality identification information can be further analyzed, and the joint remote ECG weight coefficient information can be generated, and the remote ECG weight coefficient information can be corrected and updated based on the joint remote ECG weight coefficient information, thereby ensuring that accurate diagnostic information can be provided even in complex or atypical ECG abnormality situations; through precise comparison and comprehensive evaluation mechanisms, the diagnostic performance of remote ECG monitoring can be effectively improved, thereby providing strong technical support for telemedicine in cardiology.

[0098] In an optional embodiment of the present application, the ECG abnormality identification information includes ECG abnormality severity score data, and the calculation formula of the remote ECG weight coefficient information can be:

[0099]

[0100] Where, RCM is the remote ECG weight coefficient information, N is the total number of lead channels with abnormal ECG data, ΔQ is the lead device data accuracy weight coefficient of the i-th lead channel where the ECG data is abnormal. i D is the data quality offset coefficient of the i-th lead channel where the ECG data has abnormalities. i and are the actual working state label information and the actual working state fuzzy scoring function of the ECG sensor device for the i-th lead channel where the ECG data has abnormalities, R is the weighted sum of the ECG data of the i-th lead channel with abnormal ECG data, the ECG data of the lead channel template ECG data of all lead channels with abnormal ECG data, and the ECG template conformity score data. i N is the ECG abnormality severity score data of the i-th lead channel with abnormal ECG data, * is the total number of lead channel template ECG data that is equal to the total number N of lead channels with abnormal ECG data. and They are respectively the ECG data comparison between the i-th lead channel with abnormal ECG data and the lead channel template ECG data of the j-th lead channel with abnormal ECG data, the ECG data comparison between the lead ECG template conformity score data, and the weights of the ECG data comparison between the lead ECG template conformity score data.

[0101] Optionally, when the available computing power of the server meets the preset computing power threshold condition, the server can compare the weighted sum of the ECG template conformity score data of the lead ECG data of the i-th lead channel with abnormal ECG data with the influence coefficient of the weighted sum of the lead ECG template conformity score data. and ECG abnormality severity score data impact coefficient The remote ECG weight coefficient information is further corrected. At this time, the calculation formula of the remote ECG weight coefficient information can be:

[0102]

[0103] In the formula, the influence coefficient of the weighted sum of the ECG data of the i-th lead channel with abnormal ECG data and the ECG template conformity score data is and ECG abnormality severity score data impact coefficient The weighted sum of the ECG data comparison of the lead channel template ECG data of the lead channel with abnormal ECG data and the lead ECG template conformity score data can be calculated based on the ECG data comparison of the lead channel template ECG data of the lead channel with abnormal ECG data. and ECG data with abnormal ECG severity score data R for the i-th lead channel i The influence coefficient calculation rule is generated by combining the preset influence coefficient calculation rule. The influence coefficient calculation rule can be but is not limited to being constructed based on a fuzzy control algorithm.

[0104] In an optional embodiment of the present application, the lead ECG abnormality recognition model group includes a lead ECG abnormality classification sub-model group and a lead ECG abnormality recognition sub-model group, please refer to Figure 5Inputting the lead channel marked ECG data of each lead channel into each lead ECG abnormality recognition model corresponding to the lead channel in the lead ECG abnormality recognition model group to obtain ECG abnormality recognition information of the lead channel with abnormal ECG data may include:

[0105] Step S501: Acquire photoplethysmography data.

[0106] Schematically, photoplethysmography (PPG) data is a non-invasive detection data used to detect changes in blood volume.

[0107] Step S502 : performing time sequence calibration and image segmentation on the lead channel labeled electrocardiogram data based on the photoplethysmography data to obtain a lead channel labeled single electrocardiogram data set for each lead channel.

[0108] Step S503: input the lead channel labeled single ECG dataset into the lead ECG abnormality classification sub-model corresponding to each lead channel in the lead ECG abnormality classification sub-model group, perform ECG abnormality image screening, and obtain an abnormal single ECG dataset and a normal single ECG dataset.

[0109] Optionally, the server may construct a lead ECG abnormality classification sub-model based on a clustering algorithm model. The clustering algorithm model may include, but is not limited to, a K-means clustering model, a hierarchical clustering model, a spectral clustering model, a random forest model, a self-organizing map network model, and a Gaussian mixture model.

[0110] Step S504 : inputting the abnormal single ECG data set into the lead ECG abnormality analysis sub-model in the lead ECG abnormality identification sub-model group that matches the lead channel corresponding to the abnormal single ECG data set to generate ECG abnormality identification information.

[0111] Schematically, the lead ECG abnormality analysis sub-model can be trained based on a joint neural network model constructed by combining an improved recurrent convolutional neural network with an improved autoencoder.

[0112] In the above-mentioned remote ECG data monitoring and analysis method for cardiology, by combining photoelectric volumetric pulse data and lead channel labeled ECG data, multimodal fusion of remote monitoring data can be achieved, among which photoelectric volumetric pulse data can provide hemodynamic information, which can help reduce data noise and interference, improve the accuracy of ECG abnormality identification, and thus more comprehensively evaluate the function and state of the heart, and improve the accuracy of remote diagnosis; the lead channel labeled single ECG data set is screened for ECG abnormality images through the lead ECG abnormality classification sub-model, which can effectively distinguish abnormal ECG data from normal ECG data, which can help reduce false alarms and missed alarms, and improve the reliability of abnormality identification; the abnormal single ECG data set is input into the lead ECG abnormality identification sub-model group for targeted ECG abnormality analysis, which can more accurately identify the type and degree of ECG abnormality, and can help provide more detailed diagnostic information, thereby supporting doctors to make more accurate remote diagnostic decisions.

[0113] In an optional embodiment of the present application, please refer to Figure 5 , the remote ECG data monitoring and analysis method of cardiology department also includes:

[0114] Step S505 : inputting the normal single electrocardiogram data set of each lead channel into the normal electrocardiogram image fusion model group to generate a normal electrocardiogram fusion image of each lead channel.

[0115] Step S506: Verify the normal single ECG data set based on the normal ECG fusion image to identify the ECG data that deviates from the normal state.

[0116] Specifically, the deviated normal electrocardiogram data may be used to represent single electrocardiogram data in a normal single electrocardiogram data set that has statistically significant differences from a normal electrocardiogram fusion image.

[0117] Step S507 , removing the deviated-normal electrocardiogram data from the normal single electrocardiogram data set to obtain a calibrated normal single electrocardiogram data set, and constructing a deviated-normal electrocardiogram data set based on the deviated-normal electrocardiogram data set.

[0118] Step S508 : Generate and update the cardiology department remote electrocardiogram data monitoring medical history information based on the abnormal single electrocardiogram data set, the calibrated normal single electrocardiogram data set, and the deviated normal electrocardiogram data set.

[0119] Illustratively, the abnormal single ECG data set may include a real-time abnormal ECG data set and a historical abnormal ECG data set; the calibrated normal single ECG data set may include a real-time calibrated normal single ECG data set and a historical calibrated normal single ECG data set; the deviated normal ECG data set may include a real-time deviated normal ECG data set and a historical deviated normal ECG data set.

[0120] In the above-mentioned cardiology remote ECG data monitoring and analysis method, by verifying the normal single ECG data set based on the normal ECG fusion image, it is possible to accurately identify the deviated normal ECG data that has statistically significant differences from the normal ECG fusion image, which helps to further improve the purity and consistency of the data and provide a more reliable data basis for subsequent analysis; by constructing a calibrated normal single ECG data set, it is possible to effectively reduce the risk of misjudgment due to data anomalies, thereby improving the accuracy of diagnosis; by generating and updating the cardiology remote ECG data monitoring medical history information based on the abnormal single ECG data set, the calibrated normal single ECG data set and the deviated normal ECG data set, the patient's medical history information can be made more comprehensive and accurate, and can provide doctors with more accurate and comprehensive diagnostic information, thereby providing strong support for the patient's long-term cardiology remote monitoring and cardiology treatment.

[0121] In an optional embodiment of the present application, please refer to Figure 3 , the cardiology department remote electrocardiogram data monitoring and analysis method may also include:

[0122] Step S309: Acquire and initialize the initial digital twin model of the heart.

[0123] Step S310 : If the remote ECG weight coefficient information exceeds a preset remote ECG weight coefficient threshold, select ECG abnormality core identification information that matches the remote ECG preliminary diagnosis information from the ECG abnormality identification information of each lead channel.

[0124] Step S311 : Generate cardiac abnormality visualization annotation information based on the ECG abnormality core identification information.

[0125] Step S312: annotate the visual annotation information of the heart abnormality on the initial digital twin model of the heart, update the initial digital twin model of the heart, and generate a digital twin model of the heart abnormality annotated.

[0126] The aforementioned remote ECG data monitoring and analysis method for cardiology can intuitively display the heart's anatomical structure and electrophysiological information, helping doctors quickly identify abnormal areas and make more accurate diagnoses. Furthermore, doctors can use the model to intuitively explain the condition to patients and their families, improving their understanding of the disease and their compliance with treatment.

[0127] In an exemplary embodiment of the present application, Figure 3 As shown, another cardiology remote ECG data monitoring and analysis method is provided, including the following steps S301 to S312.

[0128] Step S301: Acquire electrocardiogram data, and perform lead channel matching on the electrocardiogram data to obtain lead channel labeled electrocardiogram data.

[0129] Step S302: inputting the lead channel marked ECG data of each lead channel into each lead ECG abnormality recognition model corresponding to the lead channel in the lead ECG abnormality recognition model group to obtain ECG abnormality recognition information of the lead channel with abnormal ECG data.

[0130] Step S303 : generating remote ECG preliminary diagnosis information and remote ECG weight coefficient information corresponding to the remote ECG preliminary diagnosis information based on the ECG abnormality identification information of each lead channel.

[0131] Step S304 : adjusting the data transmission parameters of the lead channel marked electrocardiogram data and the acquisition accuracy parameters of the electrocardiogram data according to the remote electrocardiogram weight coefficient information.

[0132] Step S305 , performing lead channel demand analysis on the remote ECG preliminary diagnosis information, and generating lead channel guidance status information.

[0133] Step S306: performing lead channel status recognition on the lead channel marked ECG data to generate lead channel actual status information.

[0134] Step S307 : generating lead channel state abnormality information and lead channel state abnormality level information corresponding to the lead channel state abnormality information based on the lead channel guidance state information and the lead channel actual state information.

[0135] Step S308: If the abnormality level information exceeds the preset lead channel state abnormality level threshold, corresponding lead channel state adjustment guidance information is generated according to the lead channel state abnormality information.

[0136] Step S309: Acquire and initialize the initial digital twin model of the heart.

[0137] Step S310 : If the remote ECG weight coefficient information exceeds a preset remote ECG weight coefficient threshold, select ECG abnormality core identification information that matches the remote ECG preliminary diagnosis information from the ECG abnormality identification information of each lead channel.

[0138] Step S311 : Generate cardiac abnormality visualization annotation information based on the ECG abnormality core identification information.

[0139] Step S312: annotate the visual annotation information of the heart abnormality on the initial digital twin model of the heart, update the initial digital twin model of the heart, and generate a digital twin model of the heart abnormality annotated.

[0140] The above-mentioned remote ECG data monitoring and analysis method for cardiology can achieve accurate collection, efficient transmission, intelligent analysis and intuitive display of ECG data through technical means such as multi-lead collaborative diagnosis, deep learning models, dynamic data transmission adjustment, equipment status monitoring and adjustment, and visualization of cardiac digital twin models. It can effectively improve the accuracy of remote diagnosis, monitoring efficiency and medical service quality of cardiology, and thus support the formulation of personalized medical plans and telemedicine collaboration, and optimize the allocation of telemedicine resources.

[0141] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0142] Based on the same inventive concept, the present application also provides a cardiology remote ECG data monitoring and analysis device for implementing the aforementioned cardiology remote ECG data monitoring and analysis method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more embodiments of the cardiology remote ECG data monitoring and analysis device provided below can be found in the limitations of the cardiology remote ECG data monitoring and analysis method described above, and will not be repeated here.

[0143] In an exemplary embodiment, Figure 6 As shown, a cardiology remote electrocardiogram data monitoring and analysis device 600 is provided, comprising:

[0144] The ECG data acquisition module 601 may be used to acquire ECG data and perform lead channel matching on the ECG data to obtain lead channel labeled ECG data.

[0145] The ECG anomaly recognition module 602 can be used to input the lead channel marked ECG data of each lead channel into each lead ECG anomaly recognition model corresponding to the lead channel in the lead ECG anomaly recognition model group to obtain ECG anomaly recognition information of the lead channel with abnormal ECG data.

[0146] The remote preliminary diagnosis module 603 may be configured to generate remote ECG preliminary diagnosis information and remote ECG weight coefficient information corresponding to the remote ECG preliminary diagnosis information based on ECG abnormality identification information of each lead channel.

[0147] The remote parameter adjustment module 604 can be used to adjust the data transmission parameters of the lead channel marked ECG data and the ECG data acquisition accuracy parameters according to the remote ECG weight coefficient information. The data transmission parameters are used to regulate the data quality and / or transmission efficiency of the lead channel marked ECG data.

[0148] In an optional embodiment of the present application, the cardiology remote ECG data monitoring and analysis device 600 may also be used for:

[0149] Conducting lead channel demand analysis on the remote ECG preliminary diagnosis information to generate lead channel guidance status information, which is used to represent the working status of ECG sensing devices of each lead channel required for ECG data monitoring and analysis corresponding to the remote ECG preliminary diagnosis information;

[0150] Performing lead channel status recognition on the lead channel marked ECG data to generate lead channel actual status information, which is used to represent the actual working status of the ECG sensing device of each lead channel;

[0151] generating lead channel state abnormality information and lead channel state abnormality level information corresponding to the lead channel state abnormality information based on the lead channel guidance state information and the lead channel actual state information;

[0152] If the abnormal level information exceeds the preset lead channel status abnormal level threshold, the corresponding lead channel status adjustment guidance information is generated based on the lead channel status abnormal information. The lead channel status adjustment guidance information is used to guide the remote ECG data monitoring operator of the cardiology department to adjust the actual working status of the ECG sensor equipment of the lead channel.

[0153] In an optional embodiment of the present application, the remote preliminary diagnosis module 603 may also be used to:

[0154] Acquire lead channel template electrocardiogram data of each lead channel in which electrocardiogram data of the electrocardiogram abnormality identification information is abnormal;

[0155] Based on the comparison of the lead channel template electrocardiogram data with the lead channel marked electrocardiogram data, the electrocardiogram data of each lead channel with abnormal electrocardiogram data is generated, and the lead electrocardiogram template conformity score data is compared;

[0156] According to the ECG data, the lead ECG template conformity score data is compared with the ECG abnormality recognition information to calculate the comprehensive conformity score and remote ECG weight coefficient information;

[0157] If the comprehensive compliance score is greater than or equal to the comprehensive compliance threshold, the ECG abnormality identification information is used as the remote ECG preliminary diagnosis information;

[0158] If the comprehensive compliance score is less than the comprehensive compliance threshold, the ECG abnormality identification information of each lead channel according to the corresponding ECG data with abnormalities is input into the joint lead ECG abnormality recognition model to generate remote ECG preliminary diagnosis information and joint remote ECG weight coefficient information. The joint remote ECG weight coefficient information can be used to correct and update the remote ECG weight coefficient information.

[0159] In an optional embodiment of the present application, the lead ECG anomaly recognition model group includes a lead ECG anomaly classification sub-model group and a lead ECG anomaly recognition sub-model group. The ECG anomaly recognition module 602 may also be used to:

[0160] Acquire photoplethysmography data;

[0161] Based on the photoplethysmography data, the lead channel labeled electrocardiogram data is time-series calibrated and image segmented to obtain a lead channel labeled single electrocardiogram dataset for each lead channel;

[0162] Inputting the lead channel labeled single ECG dataset into the lead ECG abnormality classification sub-model corresponding to each lead channel in the lead ECG abnormality classification sub-model group, performing ECG abnormality image screening, and obtaining an abnormal single ECG dataset and a normal single ECG dataset;

[0163] The abnormal single electrocardiogram data set is input into the lead electrocardiogram abnormality analysis sub-model in the lead electrocardiogram abnormality recognition sub-model group that matches the lead channel corresponding to the abnormal single electrocardiogram data set to generate electrocardiogram abnormality recognition information.

[0164] In an optional embodiment of the present application, the cardiology remote ECG data monitoring and analysis device 600 may also be used for:

[0165] Input the normal single ECG data set of each lead channel into the normal ECG image fusion model group to generate the normal ECG fusion image of each lead channel;

[0166] Verify the normal single ECG data set based on the normal ECG fusion image to identify the ECG data that deviates from the normal state. The ECG data that deviates from the normal state is used to represent the single ECG data in the normal single ECG data set that has statistically significant differences from the normal ECG fusion image.

[0167] removing the deviated-normal electrocardiogram data from the normal single electrocardiogram data set to obtain a calibrated normal single electrocardiogram data set, and constructing a deviated-normal electrocardiogram data set based on the deviated-normal electrocardiogram data;

[0168] Generate and update cardiology department remote electrocardiogram data monitoring medical history information based on abnormal single electrocardiogram data set, calibrated normal single electrocardiogram data set and deviated normal electrocardiogram data set.

[0169] In an optional embodiment of the present application, the cardiology remote ECG data monitoring and analysis device 600 may also be used for:

[0170] Obtain and initialize the initial digital twin model of the heart;

[0171] If the remote ECG weight coefficient information exceeds a preset remote ECG weight coefficient threshold, selecting ECG abnormality core identification information that matches the remote ECG preliminary diagnosis information from the ECG abnormality identification information of each lead channel;

[0172] Generate visual annotation information of cardiac abnormalities based on the core recognition information of ECG abnormalities;

[0173] The visual annotation information of cardiac abnormalities is annotated on the initial digital twin model of the heart, the initial digital twin model of the heart is updated, and a digital twin model of cardiac abnormalities annotated is generated.

[0174] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the aforementioned method for remote ECG data monitoring and analysis in cardiology are implemented.

[0175] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0176] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the partial description of the method embodiments. The device embodiments described above are merely illustrative, wherein the components described as separate parts may or may not be physically separated, and the parts displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed solution. A person of ordinary skill in the art can understand and implement it without expending creative work.

[0177] The above-described embodiments merely represent several implementation methods of the embodiments of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that a person skilled in the art may make various modifications and improvements without departing from the concept of the embodiments of the present application, and these modifications and improvements fall within the scope of protection of the embodiments of the present application.

Claims

1. A remote ECG data monitoring and analysis method for cardiology, characterized in that: The method comprises: Acquiring electrocardiogram data, and performing lead channel matching on the electrocardiogram data to obtain lead channel labeled electrocardiogram data; Inputting the lead channel marked ECG data of each lead channel into each lead ECG abnormality recognition model corresponding to the lead channel in the lead ECG abnormality recognition model group to obtain ECG abnormality recognition information of the lead channel where the ECG data is abnormal; generating remote ECG preliminary diagnosis information and remote ECG weight coefficient information corresponding to the remote ECG preliminary diagnosis information based on the ECG abnormality identification information of each lead channel; The data transmission parameters of the lead channel marked electrocardiogram data and the acquisition accuracy parameters of the electrocardiogram data are adjusted according to the remote electrocardiogram weight coefficient information. The data transmission parameters are used to regulate the data quality and / or transmission efficiency of the lead channel marked electrocardiogram data.

2. The method according to claim 1, characterized in that The method further comprises: Performing a lead channel demand analysis on the remote ECG preliminary diagnosis information to generate lead channel guidance status information, wherein the lead channel guidance status information is used to represent the working status of the ECG sensing device of each lead channel required for ECG data monitoring and analysis corresponding to the remote ECG preliminary diagnosis information; Performing lead channel status recognition on the lead channel marked ECG data to generate lead channel actual status information, wherein the lead channel actual status information is used to represent the actual working status of the ECG sensing device of each lead channel; generating, based on the lead channel guidance state information and the lead channel actual state information, lead channel state abnormality information and lead channel state abnormality level information corresponding to the lead channel state abnormality information; If the abnormal level information exceeds the preset lead channel status abnormal level threshold, corresponding lead channel status adjustment guidance information is generated based on the lead channel status abnormal information. The lead channel status adjustment guidance information is used to guide the remote ECG data monitoring operator of the cardiology department to adjust the actual working status of the ECG sensing device of the lead channel.

3. The method according to claim 1, characterized in that The generating of remote ECG preliminary diagnosis information and remote ECG weight coefficient information corresponding to the remote ECG preliminary diagnosis information based on the ECG abnormality identification information of each lead channel includes: Acquire the lead channel template electrocardiogram data of the lead channel in which the electrocardiogram data of each of the electrocardiogram data of the electrocardiogram abnormality identification information is abnormal; Comparing the lead channel template electrocardiogram data with the lead channel labeled electrocardiogram data, generating electrocardiogram data comparison lead electrocardiogram template conformity score data for the lead channel where the electrocardiogram data of each of the electrocardiogram abnormality identification information is abnormal; Comparing the lead ECG template conformity score data with the ECG data and combining it with the ECG abnormality identification information to obtain a comprehensive conformity score and the remote ECG weight coefficient information; If the comprehensive compliance score is greater than or equal to the comprehensive compliance threshold, the ECG abnormality identification information is used as the remote ECG preliminary diagnosis information; If the comprehensive compliance score is less than the comprehensive compliance threshold, each of the ECG abnormality identification information is input into the joint lead ECG abnormality identification model according to the lead channel where the corresponding ECG data has abnormalities, and the remote ECG preliminary diagnosis information and the joint remote ECG weight coefficient information are generated. The joint remote ECG weight coefficient information is used to correct and update the remote ECG weight coefficient information.

4. The method according to claim 3, characterized in that The ECG abnormality identification information includes ECG abnormality severity score data, and the calculation formula of the remote ECG weight coefficient information is: Where, RCM is the remote ECG weight coefficient information, N is the total number of the lead channels with abnormal ECG data, is the lead device data accuracy weight coefficient of the i-th lead channel where the electrocardiogram data has abnormalities, ΔQ i is the data quality offset coefficient of the i-th lead channel where the electrocardiogram data has abnormalities, D i and are respectively the actual working state label information and the actual working state fuzzy scoring function of the electrocardiogram sensing device of the i-th lead channel where the electrocardiogram data has an abnormality, R is the weighted sum of the ECG data of the i-th lead channel with abnormal ECG data and the ECG data of the lead channel template ECG data of all the lead channels with abnormal ECG data and the ECG template conformity score data, i N is the ECG abnormality severity score data of the i-th lead channel where the ECG data is abnormal. * is the total number of the lead channel template electrocardiogram data, which is equal to the total number N of the lead channels with abnormal electrocardiogram data, and The electrocardiogram data comparison lead electrocardiogram template conformity score data and the weights of the electrocardiogram data comparison lead electrocardiogram template conformity score data of the electrocardiogram data of the i-th lead channel where the electrocardiogram data has an abnormality are compared with the lead channel template electrocardiogram data of the j-th lead channel where the electrocardiogram data has an abnormality.

5. The method according to claim 1, characterized in that The lead ECG anomaly recognition model group includes a lead ECG anomaly classification sub-model group and a lead ECG anomaly recognition sub-model group. The step of inputting the lead channel labeled ECG data of each lead channel into each lead ECG anomaly recognition model corresponding to the lead channel in the lead ECG anomaly recognition model group to obtain ECG anomaly recognition information of the lead channel where the ECG data is abnormal includes: Acquire photoplethysmography data; Performing time sequence calibration and image segmentation on the lead channel labeled electrocardiogram data based on the photoplethysmography data to obtain a lead channel labeled single electrocardiogram data set for each lead channel; Inputting the lead channel labeled single ECG dataset into the lead ECG abnormality classification sub-model corresponding to each lead channel in the lead ECG abnormality classification sub-model group, performing ECG abnormality image screening, and obtaining an abnormal single ECG dataset and a normal single ECG dataset; The abnormal single electrocardiogram data set is input into the lead electrocardiogram abnormality analysis sub-model in the lead electrocardiogram abnormality identification sub-model group that is consistent with the lead channel corresponding to the abnormal single electrocardiogram data set to generate the electrocardiogram abnormality identification information.

6. The method according to claim 5, characterized in that The method further comprises: Inputting the normal single electrocardiogram data set of each lead channel into a normal electrocardiogram image fusion model group to generate a normal electrocardiogram fusion image of each lead channel; verifying the normal single ECG data set based on the normal ECG fusion image to identify ECG data that deviates from normal, wherein the ECG data that deviates from normal is used to represent single ECG data in the normal single ECG data set that has statistically significant differences from the normal ECG fusion image; removing the deviated-normal electrocardiogram data from the normal single electrocardiogram data set to obtain a calibrated normal single electrocardiogram data set, and constructing a deviated-normal electrocardiogram data set based on the deviated-normal electrocardiogram data; Cardiology department remote electrocardiogram data monitoring medical history information is generated and updated based on the abnormal single electrocardiogram data set, the calibrated normal single electrocardiogram data set, and the deviated normal electrocardiogram data set.

7. The method according to any one of claims 1 to 6, characterized in that The method further comprises: Obtain and initialize the initial digital twin model of the heart; If the remote ECG weight coefficient information exceeds a preset remote ECG weight coefficient threshold, selecting ECG abnormality core identification information that matches the remote ECG preliminary diagnosis information from the ECG abnormality identification information of each lead channel; Generate cardiac abnormality visualization annotation information based on the electrocardiogram abnormality core identification information; The visual annotation information of the heart abnormality is marked on the initial digital twin model of the heart, the initial digital twin model of the heart is updated, and a digital twin model with annotated heart abnormality is generated.

8. A remote ECG data monitoring and analysis device for cardiology, characterized in that: The device comprises: An ECG data acquisition module is used to acquire ECG data and perform lead channel matching on the ECG data to obtain lead channel marked ECG data; an ECG anomaly recognition module, configured to input the lead channel marked ECG data of each lead channel into each lead ECG anomaly recognition model corresponding to the lead channel in the lead ECG anomaly recognition model group, and obtain ECG anomaly recognition information of the lead channel where the ECG data is abnormal; A remote preliminary diagnosis module, configured to generate remote ECG preliminary diagnosis information and remote ECG weight coefficient information corresponding to the remote ECG preliminary diagnosis information based on the ECG abnormality identification information of each lead channel; A remote parameter adjustment module is used to adjust the data transmission parameters of the lead channel marked electrocardiogram data and the acquisition accuracy parameters of the electrocardiogram data according to the remote electrocardiogram weight coefficient information. The data transmission parameters are used to regulate the data quality and / or transmission efficiency of the lead channel marked electrocardiogram data.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images