An electrocardiogram examination operation method and system for simulation teaching

Abstract

The present invention relates to the field of teaching technologies, and specifically discloses an electrocardiogram examination operation method and system for simulation teaching, including: obtaining a data change record during the data synchronization process, where the data change record includes data definition language (DDL) sample set information, an operation timestamp, and operation content; obtaining a plurality of statement sample information according to the data definition language (DDL) sample set information, and obtaining the semantic similarity of each statement sample information; obtaining an operation identifier corresponding to each statement sample information according to the operation content. By obtaining operation identifiers, calculating operation similarity values, operation data difference indexes, etc., the present invention comprehensively and deeply solves the problem of insufficient operation analysis in the prior art. It can accurately extract key operation feature information, clarify the actual meaning and association of data operations in combination with context identifiers, and avoid misjudgments caused by inaccurate information extraction and lack of context consideration.

Description

Technical Field

[0001] The present invention relates to the field of teaching technology, and in particular to an electrocardiogram examination operation method and system for simulation teaching. Background Art

[0002] In the field of medical education, electrocardiogram examination, as a key clinical diagnosis skill, the quality of its teaching is crucial for cultivating professional medical talents. Traditional electrocardiogram examination teaching mainly adopts the methods of theoretical explanation, teacher demonstration and students' actual operation practice.

[0003] The traditional electrocardiogram examination teaching method mainly relies on on-site teacher demonstration and students' actual operation practice. The evaluation method for students' actual operation practice is too single. Most only focus on the diagnostic results, but ignore important factors such as operation standardization and signal quality. At the same time, the evaluation method with fixed weights cannot be adjusted according to the individual differences of students and the dynamic changes in the learning process, and cannot comprehensively and accurately measure the learning achievements of students. This makes it difficult for teachers to discover the specific problems of students in the operation process and unable to provide targeted improvement suggestions for students, which is not conducive to the improvement of students' operation skills. Summary of the Invention

[0004] The purpose of the present invention is to provide an electrocardiogram examination operation method and system for simulation teaching to solve the technical problems raised in the above background art.

[0005] To achieve the above purpose, the present invention provides the following technical solutions:

[0006] An electrocardiogram examination operation method for simulation teaching, comprising:

[0007] Obtaining the operation information of students in the electrocardiogram simulation teaching scenario, and constructing a digital twin model of the students' electrocardiogram simulation operation according to the operation information;

[0008] Obtaining the simulated operation data of the physical teaching equipment according to the digital twin model, and generating dynamic simulated pathological parameters according to the simulated operation data, where the dynamic simulated pathological parameters include electrode positioning deviation sequences, pacing signal dynamic response characteristics, and diagnostic conclusion virtual mapping relationships;

[0009] Obtaining a multi-source dynamic simulation total evaluation set according to the dynamic simulated pathological parameters, and performing priority classification on the multi-source dynamic simulation total evaluation set through spatio-temporal correlation analysis to generate a multi-source data priority set;

[0010] Obtaining multi-modal evaluation data according to the multi-source data priority set;

[0011] Obtaining an operation comprehensive score according to the multi-modal evaluation data;

[0012] Generate personalized simulated teaching scheduling information based on the comprehensive score of the operations, and push dynamically adapted electrocardiogram simulated teaching resources based on the simulated teaching scheduling information.

[0013] Preferably, the step of constructing a digital twin model of the electrocardiogram simulation operation of the trainee according to the operation information includes:

[0014] Obtain the three-dimensional space trajectory of the electrocardiogram electrode patch, the contact pressure of the electrocardiogram electrode patch, and the operation timing data of the electrocardiogram electrode patch according to the operation information;

[0015] Based on the physical-virtual bidirectional mapping algorithm, map the three-dimensional space trajectory into a virtual electrode positioning topological map;

[0016] Map the contact pressure into a virtual signal strength parameter,

[0017] Map the operation timing data into a timestamp sequence of the digital twin;

[0018] Fuse the virtual electrode positioning topological map, the virtual signal strength parameter, and the timestamp sequence into a digital twin model.

[0019] Preferably, the step of generating dynamic simulation pathological parameters according to the simulation operation data includes:

[0020] Obtain a simulated pacing signal and a simulated electrode positioning set according to the simulation operation data, wherein the simulated electrode positioning set includes a plurality of simulated electrode positioning data;

[0021] Obtain a pacing phase offset according to the simulated pacing signal;

[0022] Obtain the dynamic response characteristics of the pacing signal according to the pacing phase offset, wherein the dynamic response characteristics of the pacing signal include the signal frequency change degree, the signal amplitude fluctuation degree, and the waveform distortion degree;

[0023] Generate simulated diagnosis information according to the signal frequency change degree, the signal amplitude fluctuation degree, and the waveform distortion degree;

[0024] Compare the simulated diagnosis information with the cases in the standard pathological library one by one, and generate a mapping relationship matrix including the misdiagnosis probability through semantic matching and feature similarity calculation;

[0025] Obtain the standard electrode positioning data, and obtain the simulated electrode positioning deviation corresponding to each simulated electrode positioning data according to the standard electrode positioning data;

[0026] Obtain the time sequence arrangement according to the operation timing data of the electrocardiogram electrode patch;

[0027] Sort all the simulated electrode positioning deviations in chronological order to obtain an electrode positioning deviation sequence;

[0028] Perform comprehensive fusion based on the electrode positioning deviation sequence, the dynamic response characteristics of the pacing signal, and the mapping relationship matrix to generate dynamic simulated pathological parameters.

[0029] Preferably, the step of obtaining a multi-source dynamic simulation total evaluation set according to the dynamic simulated pathological parameters includes:

[0030] Generate an electrode positioning error heat map through a spatial interpolation algorithm according to the electrode positioning deviation sequence;

[0031] Obtain electrode positioning error distribution information according to the electrode positioning error heat map;

[0032] Obtain the error distribution density corresponding to each time sequence according to the electrode positioning error distribution information;

[0033] Obtain heart rhythm characteristic data according to the dynamic response characteristics of the pacing signal;

[0034] Generate an arrhythmia complexity spectrum through a spectrum analysis algorithm according to the heart rhythm characteristic data;

[0035] Obtain the heart rhythm urgency corresponding to each time sequence according to the arrhythmia complexity spectrum;

[0036] Obtain the time sequence simulation evaluation value corresponding to each time sequence according to the error distribution density and the heart rhythm urgency;

[0037] Obtain a multi-source dynamic simulation total evaluation set according to the time sequence simulation evaluation values corresponding to all time sequences.

[0038] Preferably, the step of obtaining multi-modal evaluation data according to the multi-source data priority set includes:

[0039] Build a spatio-temporal evolution feature extraction network for the multi-source data priority set, and use a multi-layer convolutional neural network to extract the local morphological features of the multi-source data priority set to obtain multi-source electrocardiogram waveform feature information;

[0040] Obtain the time sequence information of the multi-source electrocardiogram waveform feature information, and model the time sequence information of the multi-source electrocardiogram waveform feature information to obtain the change trend of the heart rate and the time dependence relationship between different waveforms;

[0041] Obtain abnormal electrocardiogram waveform nodes according to the time dependence relationship;

[0042] Compare the abnormal electrocardiogram waveform nodes with the same type of features in the historical pathological data, and calculate the similarity of the abnormal electrocardiogram waveform nodes;

[0043] Determine whether the similarity of the abnormal electrocardiogram waveform nodes exceeds a preset threshold;

[0044] If it exceeds, generate multi-modal evaluation data based on the abnormal electrocardiogram waveform nodes.

[0045] Preferably, the step of obtaining the operation comprehensive score according to the multi-modal evaluation data includes:

[0046] Extract key operation features from the multi-modal evaluation data through the adaptive feature selector of the digital twin, where the key operation features include the virtual electrode fitting degree index, the signal fidelity parameter, and the diagnostic logic consistency score;

[0047] Based on the dynamic weight optimizer of the digital twin, adjust the weight allocation of the key operation features in real time according to the historical operation data of the trainee;

[0048] Map the weighted key operation features to a standardized operation comprehensive score through the fuzzy comprehensive evaluation algorithm.

[0049] The present invention also provides an electrocardiogram examination operation system for simulation teaching, including:

[0050] The first acquisition module is used to acquire the operation information of the trainee in the electrocardiogram simulation teaching scenario and construct a digital twin model of the trainee's electrocardiogram simulation operation according to the operation information;

[0051] The second acquisition module is used to acquire the simulated operation data of the physical teaching equipment according to the digital twin model and generate dynamic simulated pathological parameters according to the simulated operation data. The dynamic simulated pathological parameters include the electrode positioning deviation sequence, the dynamic response characteristics of the pacing signal, and the virtual mapping relationship of the diagnostic conclusion;

[0052] The third acquisition module is used to acquire a multi-source dynamic simulation total evaluation set according to the dynamic simulated pathological parameters and perform priority classification on the multi-source dynamic simulation total evaluation set through spatio-temporal correlation analysis to generate a multi-source data priority set;

[0053] The fourth acquisition module is used to acquire multi-modal evaluation data according to the multi-source data priority set;

[0054] The fifth acquisition module is used to obtain the operation comprehensive score according to the multi-modal evaluation data;

[0055] The push module is used to generate personalized simulation teaching scheduling information according to the operation comprehensive score and push dynamically adapted electrocardiogram simulation teaching resources based on the simulation teaching scheduling information.

[0056] Preferably, the first acquisition module includes:

[0057] A first acquisition unit, configured to acquire electrocardiogram electrode patch three-dimensional space trajectory, electrocardiogram electrode patch contact pressure, and electrocardiogram electrode patch operation timing data according to the operation information;

[0058] A first mapping unit, configured to map the three-dimensional space trajectory into a virtual electrode positioning topology map based on a physical-virtual bidirectional mapping algorithm;

[0059] A second mapping unit, configured to map the contact pressure into a virtual signal strength parameter,

[0060] A third mapping unit, configured to map the operation timing data into a timestamp sequence of the digital twin;

[0061] A first fusion unit, configured to fuse the virtual electrode positioning topology map, the virtual signal strength parameter, and the timestamp sequence into a digital twin model.

[0062] Preferably, the second acquisition module includes:

[0063] A second acquisition unit, configured to acquire a simulated pacing signal and a simulated electrode positioning set according to the simulation operation data, where the simulated electrode positioning set includes a plurality of simulated electrode positioning data;

[0064] A third acquisition unit, configured to acquire a pacing phase offset according to the simulated pacing signal;

[0065] A fourth acquisition unit, configured to acquire a pacing signal dynamic response feature according to the pacing phase offset, where the pacing signal dynamic response feature includes a signal frequency change degree, a signal amplitude fluctuation degree, and a waveform distortion degree;

[0066] A first generation unit, configured to generate simulated diagnostic information according to the signal frequency change degree, the signal amplitude fluctuation degree, and the waveform distortion degree;

[0067] A second generation unit, configured to compare the simulated diagnostic information with cases in a standard pathology library one by one, and generate a mapping relationship matrix including a misdiagnosis probability through semantic matching and feature similarity calculation;

[0068] A fifth acquisition unit, configured to acquire standard electrode positioning data, and acquire a simulated electrode positioning deviation corresponding to each simulated electrode positioning data according to the standard electrode positioning data;

[0069] A sixth acquisition unit, configured to acquire a time sequence arrangement according to the electrocardiogram electrode patch operation timing data;

[0070] A sorting unit, configured to sort all simulated electrode positioning deviations according to the time sequence arrangement to obtain an electrode positioning deviation sequence;

[0071] A second fusion unit, configured to perform comprehensive fusion according to the electrode positioning deviation sequence, the dynamic response characteristics of the pacing signal, and the mapping relationship matrix, and generate dynamic simulated pathological parameters.

[0072] Preferably, the third acquisition module includes:

[0073] A third generation unit, configured to generate an electrode positioning error heat map according to the electrode positioning deviation sequence through a spatial interpolation algorithm;

[0074] A seventh acquisition unit, configured to acquire electrode positioning error distribution information according to the electrode positioning error heat map;

[0075] An eighth acquisition unit, configured to acquire the error distribution density corresponding to each time sequence according to the electrode positioning error distribution information;

[0076] A ninth acquisition unit, configured to acquire heart rhythm characteristic data according to the dynamic response characteristics of the pacing signal;

[0077] A tenth acquisition unit, configured to generate an arrhythmia complexity spectrum according to the heart rhythm characteristic data through a spectrum analysis algorithm;

[0078] An eleventh acquisition unit, configured to acquire the heart rhythm urgency corresponding to each time sequence according to the arrhythmia complexity spectrum;

[0079] A twelfth acquisition unit, configured to acquire the time-sequence simulation evaluation value corresponding to each time sequence according to the error distribution density and the heart rhythm urgency;

[0080] A thirteenth acquisition unit, configured to acquire a multi-source dynamic simulation total evaluation set according to the time-sequence simulation evaluation values corresponding to all time sequences.

[0081] The beneficial effects of this application are as follows: By comprehensively collecting the operation information of trainees, the present invention constructs a digital twin model, which accurately reflects the operation details, solves the problems of incomplete collection and inaccurate analysis of traditional teaching information, and facilitates teachers to carry out personalized teaching. Generate dynamic simulated pathological parameters, closely associate the operation with the pathology, help trainees understand the internal relationship between the operation and the disease diagnosis, overcome the drawback of the disconnection between the two in traditional teaching, evaluate the pathological parameters and classify the priorities of multi-source data, reasonably allocate computing resources, improve the evaluation efficiency and pertinence, and ensure the timely processing of key data. Construct an evaluation model to obtain multi-modal evaluation data, comprehensively evaluate the operation level of trainees from multiple dimensions, make up for the defect of the single traditional evaluation method, obtain operation classification data and comprehensive scores, quantify the performance of trainees, and provide an objective basis for personalized teaching. Generate teaching scheduling information according to the comprehensive score and push the adapted teaching resources, solve the problems of unreasonable teaching resource allocation and untimely pushing, meet the personalized learning needs of trainees, improve the utilization efficiency of teaching resources, and ultimately effectively improve the teaching quality of the simulated teaching electrocardiogram examination operation, and help trainees better master the relevant knowledge and skills of electrocardiogram examination. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 It is a schematic flowchart of the method according to an embodiment of the present application.

[0083] Figure 2 It is a schematic structural diagram of the system according to an embodiment of the present application.

[0084] The realization of the purpose, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0085] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0086] As Figure 1 shown, the present application provides a method for electrocardiogram examination operation in simulated teaching, which is applied to the circuit component feature database and includes:

[0087] S1. Obtain the operation information of trainees in the electrocardiogram simulation teaching scenario, and construct a digital twin model of the trainees' electrocardiogram simulation operation according to the operation information;

[0088] S2. Obtain the simulated operation data of the physical teaching equipment according to the digital twin model, and generate dynamic simulated pathological parameters according to the simulated operation data, where the dynamic simulated pathological parameters include electrode positioning deviation sequences, pacing signal dynamic response characteristics, and diagnostic conclusion virtual mapping relationships;

[0089] S3. Obtain a multi-source dynamic simulation total evaluation set according to the dynamic simulation pathological parameters, and perform priority classification on the multi-source dynamic simulation total evaluation set through spatio-temporal correlation analysis to generate a multi-source data priority set;

[0090] S4. Obtain multi-modal evaluation data according to the multi-source data priority set;

[0091] S5. Obtain an operation comprehensive score according to the multi-modal evaluation data;

[0092] S6. Generate personalized simulation teaching scheduling information according to the operation comprehensive score, and push dynamically adapted electrocardiogram simulation teaching resources based on the simulation teaching scheduling information.

[0093] As described in the above steps S1 - S6, the present invention collects the three-dimensional spatial trajectory, touch pressure, and operation timing data of the electrode patch during the operation of the trainee, and uses the physical-virtual bidirectional mapping algorithm to transform and fuse to construct a digital twin model. This model can clearly present the operation problems of the trainee, such as inaccurate electrode patch pasting position, uneven force, etc., to help teachers carry out targeted teaching; then, obtain simulation operation data according to the digital twin model, extract the simulated pacing signal and the simulated electrode positioning set. Obtain the pacing phase offset from the simulated pacing signal, further analyze the dynamic response characteristics of the pacing signal, and generate simulated diagnosis information. Calculate the misdiagnosis probability by comparing with the standard pathology library, obtain the deviation of the standard electrode positioning data and sort it, and comprehensively generate dynamic simulation pathological parameters, solving the problem of the disconnection between operation and pathology, and helping trainees understand the impact of the operation on the diagnosis result; then, use the spatial interpolation algorithm to generate a heat map according to the electrode positioning deviation sequence, combine it with the pacing signal response characteristics, generate an arrhythmia complexity spectrum through spectral analysis, comprehensively calculate the simulated evaluation value of each time sequence, generate a multi-source dynamic simulation evaluation set and classify it, solving the problems of single traditional teaching evaluation and unreasonable resource allocation, giving priority to processing key data and avoiding misdiagnosis; then, build a spatio-temporal evolution feature extraction network, use a multi-layer convolutional neural network to extract local morphological features, analyze the time sequence information, find abnormal waveform nodes and compare them with historical pathological data to generate multi-modal evaluation data, comprehensively evaluate the operation level of trainees, and facilitate teachers to guide trainees to learn pathological knowledge and diagnosis key points; then, extract key operation features through the relevant technology of the digital twin, adjust the weight according to the historical data of the trainee, and use the fuzzy comprehensive evaluation algorithm to obtain a standardized operation comprehensive score, providing a basis for personalized teaching; then, match teaching resources according to the operation comprehensive score, generate simulation teaching scheduling information, meet the personalized learning needs of trainees, and improve the resource utilization efficiency; finally, monitor the resource load status in real time, select the optimal transmission path to push resources, and dynamically adjust the pushed content and method according to the feedback of the trainee, realizing the closed-loop feedback control of the teaching scenario.

[0094] In one embodiment, the step of constructing a digital twin model for simulating the electrocardiogram operation of a trainee according to the operation information includes:

[0095] S101. Obtain the three-dimensional spatial trajectory of the electrocardiogram electrode patch, the contact pressure of the electrocardiogram electrode patch, and the operation timing data of the electrocardiogram electrode patch according to the operation information;

[0096] S102. Based on the physical-virtual bidirectional mapping algorithm, map the three-dimensional spatial trajectory into a virtual electrode positioning topological graph;

[0097] S103. Map the contact pressure into a virtual signal strength parameter,

[0098] S104. Map the operation timing data into a timestamp sequence of the digital twin;

[0099] S105. Integrate the virtual electrode positioning topological graph, the virtual signal strength parameter, and the timestamp sequence into a digital twin model.

[0100] As described in the above steps S101 - S105, the present invention comprehensively collects information when a trainee operates an electrocardiogram electrode patch by using a high - precision three - dimensional motion sensor, a pressure sensor, and a time recording device, and obtains the three - dimensional space trajectory of the electrode patch, the contact pressure, and the operation timing data. These data accurately reflect the trainee's operation habits and skill levels. Problems such as chaotic operation sequences and uneven contact pressures of the trainee can be precisely captured, providing a key basis for teachers to carry out personalized teaching; then, based on the physical - virtual bidirectional mapping algorithm, the three - dimensional space trajectory data is converted into a virtual electrode positioning topology map. By comparing the virtual topology map of the standard electrode patch paste position and the topology map generated by the trainee's operation, the deviation in the electrode patch paste position can be quickly discovered, and teachers can accordingly guide the trainee to adjust specifically to improve teaching efficiency; after that, according to the corresponding relationship between the contact pressure and the signal strength established in the early stage, the contact pressure data is converted into virtual signal strength parameters. This enables the trainee to intuitively understand the impact of the operation force on the electrocardiogram signal quality and adds a new dimension to teaching evaluation. If the contact pressure is too small when the trainee pastes the electrode patch, the virtual signal strength parameter will show an abnormality, and the teacher can timely remind the trainee to adjust, and the trainee can also standardize the operation by observing the parameter changes; then, the operation timing data is sorted in chronological order and time stamps are assigned to form a time - stamp sequence of the digital twin. With the help of this sequence, teachers can understand in detail the correctness of the trainee's operation steps and the rationality of the time spent on each step. If the trainee's operation shows a situation where the electrode patch is pasted first and then the position is adjusted, resulting in a long time, the time - stamp sequence can clearly present it, and teachers can help the trainee standardize the operation sequence and improve the operation rhythm and standardization based on this; finally, technologies such as data association and feature extraction are used to integrate the virtual electrode positioning topology map, the virtual signal strength parameters, and the time - stamp sequence to construct a complete digital twin model and visualize it. This model comprehensively reflects the trainee's operation situation, solves the problems of scattered traditional teaching data, difficult comprehensive analysis, and inability to carry out personalized teaching. Teachers can evaluate the trainee's performance in aspects such as electrode patch position, contact pressure, and operation timing through the model, and formulate personalized teaching plans for the problems existing in the trainee. For example, for trainees with electrode patch position deviation, uneven contact pressure, and chaotic operation timing, a teaching plan including positioning training, guidance on adjusting the strength, and practice on standardizing the timing is formulated.

[0101] In one embodiment, the step of generating dynamic simulation pathological parameters according to the simulation operation data includes:

[0102] S201. Obtain a simulated pacing signal and a simulated electrode positioning set according to the simulation operation data, wherein the simulated electrode positioning set includes a plurality of simulated electrode positioning data;

[0103] S202. Obtain a pacing phase offset according to the simulated pacing signal;

[0104] S203. Obtain the dynamic response characteristics of the pacing signal according to the pacing phase offset, wherein the dynamic response characteristics of the pacing signal include the signal frequency change degree, the signal amplitude fluctuation degree, and the waveform distortion degree;

[0105] S204. Generate simulated diagnostic information according to the signal frequency change degree, the signal amplitude fluctuation degree, and the waveform distortion degree;

[0106] S205. Compare the simulated diagnostic information with the cases in the standard pathology database one by one, and generate a mapping relationship matrix including the misdiagnosis probability through semantic matching and feature similarity calculation;

[0107] S206. Obtain the standard electrode positioning data, and obtain the corresponding simulated electrode positioning deviation of each simulated electrode positioning data according to the standard electrode positioning data;

[0108] S207. Obtain the time sequence arrangement according to the electrocardiogram electrode patch operation timing data;

[0109] S208. Sort all the simulated electrode positioning deviations according to the time sequence arrangement to obtain an electrode positioning deviation sequence;

[0110] S209. Perform comprehensive fusion according to the electrode positioning deviation sequence, the dynamic response characteristics of the pacing signal, and the mapping relationship matrix to generate dynamic simulated pathological parameters.

[0111] As described in the above steps S201 - S209, during the simulation teaching of the present invention, the sensor network of the simulation teaching device collects data, and a specific data parsing program separates the simulated pacing signal and the simulated electrode positioning set. The simulated electrode positioning set records the position information of each electrode on the simulated human body. If the electrode position is incorrect during the operation of the trainee, the simulated electrode positioning set and the simulated pacing signal can record relevant data, and the teacher can use this to guide the trainee to understand the importance of electrode positioning for pacing signal detection. Subsequently, a signal processing algorithm is used to filter and denoise the simulated pacing signal, and then the current phase is compared with the normal reference phase through a phase detection algorithm to obtain the pacing phase offset. Taking the simulated premature ventricular contraction as an example, the teacher explains the cardiac electrophysiological changes, electrocardiogram manifestations and diagnostic significance in combination with the offset, improving the trainee's diagnostic ability for arrhythmia and solving the problem of inaccurate analysis of abnormal cardiac pacing signals in traditional teaching. Then, with the help of mathematical models and signal analysis methods, the signal frequency change degree, amplitude fluctuation degree and waveform distortion degree are calculated based on the pacing phase offset. When atrioventricular block is simulated, these characteristics can show various changes in the electrocardiogram, helping the trainee to understand the internal relationship between cardiac electrophysiology and electrocardiogram waveform, and solving the problem of single analysis of electrocardiogram waveform changes in traditional teaching. After that, a correlation model between signal characteristics and disease diagnosis is established, converting complex signal characteristics into simulated diagnosis information to cultivate the trainee's clinical diagnosis thinking. For example, when specific distortion characteristics appear in the simulated electrocardiogram, simulated diagnosis information is generated, and the teacher guides the trainee to analyze the reasons. Then, the simulated diagnosis information is compared with the standard pathology database, and through semantic matching and feature similarity calculation, a mapping relationship matrix including the misdiagnosis probability is formed to help the trainee understand the diagnostic accuracy and cultivate the self - assessment ability. For example, after the trainee diagnoses "sinus tachycardia" and the signal feature differences are found through comparison, the teacher can point out the problems to deepen the trainee's understanding of the diagnostic criteria. Then, the standard electrode positioning data is retrieved from the simulation teaching system database and compared with the simulated electrode positioning data to calculate the simulated electrode positioning deviation, enabling the trainee to realize the importance of electrode positioning accuracy. At the same time, the operation timing data is extracted, and a time - ordered record is formed through a timestamp sorting algorithm to be used to discover operation process problems and cultivate the trainee's standardized operation habits. And based on this arrangement of the simulated electrode positioning deviation data, a deviation sequence is formed to reflect the dynamic change of the deviation, providing a basis for targeted teaching. Finally, data fusion technology is used to integrate the electrode positioning deviation sequence, the dynamic response characteristics of the pacing signal and the mapping relationship matrix to generate comprehensive and dynamic simulated pathological parameters. For example, when there are multiple problems in the trainee's operation, the generated parameters are displayed on the visualization interface, and the teacher guides the trainee to analyze the problems to improve the teaching effect, solving the problem of simple and one - sided generation of pathological parameters in traditional teaching.

[0112] In one embodiment, the step of obtaining the multi - source dynamic simulation total evaluation set according to the dynamic simulated pathological parameters includes:

[0113] S301. Generate an electrode positioning error heat map through a spatial interpolation algorithm based on the electrode positioning deviation sequence;

[0114] S302. Obtain electrode positioning error distribution information based on the electrode positioning error heat map;

[0115] S303. Obtain the error distribution density corresponding to each time sequence according to the electrode positioning error distribution information;

[0116] S304. Obtain heart rhythm characteristic data based on the dynamic response characteristics of the pacing signal;

[0117] S305. Generate an arrhythmia complexity spectrum through a spectral analysis algorithm based on the heart rhythm characteristic data;

[0118] S306. Obtain the heart rhythm urgency corresponding to each time sequence according to the arrhythmia complexity spectrum;

[0119] S307. Obtain the time sequence simulation evaluation value corresponding to each time sequence according to the error distribution density and the heart rhythm urgency;

[0120] S308. Obtain a multi-source dynamic simulation total evaluation set according to the time sequence simulation evaluation values corresponding to all time sequences.

[0121] As described in the above steps S301 - S308, in the simulation teaching of the present invention, when the electrode positioning deviation sequence is obtained, a spatial interpolation algorithm is used to generate a heat map of electrode positioning error. This algorithm will make reasonable speculation and filling in the surrounding space based on the known electrode positioning deviation data points. In this way, the electrode positioning deviation data is converted into an intuitive heat map. In the heat map, different colors represent different degrees of error, and the darker the color, the greater the error. In this way, students and teachers can clearly see the spatial distribution of electrode positioning errors and quickly locate the error concentration areas. For example, after a simulation operation, the generated heat map shows that the colors of several lead regions in the chest are darker, which clearly indicates that the electrode positioning errors in this region are larger, providing a clear direction for subsequent teaching guidance and operation improvement, and solving the problem in traditional teaching that the analysis of electrode positioning errors is not intuitive and it is difficult to quickly find the problem areas. Then, after obtaining the heat map of electrode positioning error, data analysis will be carried out on it. Specifically, first determine the error range corresponding to different colors in the heat map, and then through image recognition and data analysis techniques, statistics such as the area, position of different color regions and their mutual relationships are obtained. At the same time, whether there is a certain law in the error distribution will also be analyzed. Through this step, more specific and detailed error distribution information can be extracted from the heat map to help students and teachers deeply understand the characteristics of electrode positioning errors. For example, after analysis, it is found that a student's electrode positioning errors are mainly concentrated in the right chest leads, and the error values in these regions are generally large, and are related to the student's operation habits of the right hand during the operation. This provides a basis for formulating targeted improvement measures in the future and solves the problem in traditional teaching that the characteristics of electrode positioning errors are not accurately grasped. Then, after obtaining the electrode positioning error distribution information and combining with the time record of the operation, the system will divide this information according to the time sequence. For each specific time period, calculate the ratio of the area of the error region to the area of the entire monitoring region during this period to obtain the error distribution density corresponding to each time sequence. This is like observing the density of the water flow in a river at different times by calculating the proportion of the water flow occupying a specific area. In this way, the change in the density of electrode positioning errors at different times during the operation process can be reflected. For example, during the simulation operation, it is found that the error distribution density significantly increases from the 2nd to the 4th minute after the operation starts. Further observation reveals that this is because the student pasted multiple electrodes simultaneously during this period and the operation was relatively hasty, resulting in an increase in positioning errors. This helps students and teachers dynamically understand the problems during the operation process, discover the unstable periods of the operation, and provides a basis for optimizing the operation process, solving the problem in traditional teaching that the change of electrode positioning errors cannot be dynamically tracked. In the simulation teaching, the system will analyze and process the dynamic response characteristics of the pacing signal.Starting from aspects such as the signal frequency change degree, signal amplitude fluctuation degree, and waveform distortion degree, extract features related to heart rhythm. For example, pay attention to the heart rate changes in the signal frequency change degree, including the average heart rate and the fluctuation range of the heart rate; analyze the maximum value, minimum value of the amplitude, and the fluctuation frequency in the signal amplitude fluctuation degree; study whether abnormal wave peaks, wave troughs, etc. appear in the waveform distortion degree. By comprehensively extracting these features, obtain the feature data reflecting the heart rhythm situation. This step helps trainees deeply understand the relationship between cardiac electrophysiological activities and electrocardiogram manifestations, providing a key data basis for subsequent analysis of arrhythmias and other problems. For example, when the simulated pacing signal shows a large signal frequency change degree and obvious amplitude fluctuations, the extracted heart rhythm feature data shows that the average heart rate is too fast and the fluctuation is unstable. Teachers can combine these data to explain to trainees that this situation may be related to arrhythmias, guide trainees to further analyze the electrocardiogram waveform, and improve trainees' understanding of the diagnosis of heart diseases, solving the problem in traditional teaching of insufficient understanding of pacing signals and difficulty in effectively associating them with heart rhythm features. Next, the spectral analysis algorithm will be used to process the heart rhythm feature data. This algorithm will transform the heart rhythm feature data in the time domain to the frequency domain, and analyze the proportion and distribution of different frequency components in the heart rhythm signal. Through specific calculation methods, generate a map reflecting the complexity of arrhythmias, that is, the arrhythmia complexity spectrum. In this spectrum, different frequency components and amplitudes represent different arrhythmia features. This is like decomposing a complex piece of music into different notes and melodies, and judging the complexity of the music by analyzing their combination and changes. By generating the arrhythmia complexity spectrum, transform the heart rhythm feature data into an intuitive map form, helping trainees more clearly understand the types and severity of arrhythmias. For example, in the simulated operation, after the obtained heart rhythm feature data is processed by the spectral analysis algorithm, the generated arrhythmia complexity spectrum shows a high amplitude in a specific frequency range. Teachers can explain according to this spectrum that this may represent a specific type of arrhythmia, such as ventricular tachycardia, and introduce in detail the characteristic manifestations of this arrhythmia on the spectrum, improving trainees' diagnostic and analytical abilities for complex arrhythmias, solving the problem in traditional teaching of insufficiently intuitive and comprehensive analysis of arrhythmias. After that, according to the pre-set rules and standards, the system will evaluate the features of the arrhythmia complexity spectrum at each time sequence. If at a certain time sequence, the amplitudes of some key frequency components in the arrhythmia complexity spectrum exceed the set threshold, or a specific frequency combination pattern appears, the heart rhythm urgency at that time sequence will be set to a higher level. At the same time, factors such as the trend of heart rhythm changes, such as whether the arrhythmia complexity rapidly increases in a short period of time, will also be considered. This step is like attaching different labels to the severity of arrhythmias, enabling trainees and teachers to understand the changing urgency of the cardiac electrophysiological condition in real time.For example, during the simulated electrocardiogram (ECG) monitoring process, the complexity spectrum of arrhythmia at a certain time sequence shows that the amplitude of a specific frequency suddenly increases. According to the rules, the urgency degree of the heart rhythm at this time sequence is determined to be relatively high. Teachers can use this to emphasize to trainees that when encountering such a situation in actual clinical practice, corresponding measures need to be taken in a timely manner, such as further observing the patient's symptoms, preparing first-aid equipment, etc., to cultivate the trainees' clinical emergency handling capabilities, solving the problem that it is difficult to evaluate the urgency degree of arrhythmia in real time in traditional teaching. Then, according to a certain weight assignment principle, the error distribution density and the heart rhythm urgency degree are comprehensively calculated. For example, different weights are set for the error distribution density and the heart rhythm urgency degree according to their importance. Multiply the error distribution density of each time sequence by its corresponding weight, and then add the heart rhythm urgency degree multiplied by the corresponding weight, and the time sequence simulation evaluation value of this time sequence is obtained. This is like in an exam, different subjects are included in the total score according to different weights. In this way, the factors of electrode positioning error and heart rhythm situation in two different aspects are combined to form a comprehensive evaluation index. For example, at a certain time sequence, the electrode positioning error distribution density is relatively high, and at the same time the heart rhythm urgency degree is also relatively high. After calculation according to the set weights, the obtained time sequence simulation evaluation value is relatively low. Teachers can, based on this evaluation value, comprehensively point out the problems existing in the operation to trainees, including deficiencies in electrode positioning and heart rhythm judgment, providing a more objective and accurate basis for teaching evaluation, solving the problem that the traditional teaching evaluation method is single and cannot comprehensively consider multiple factors. Finally, the time sequence simulation evaluation values corresponding to each time sequence are summarized and sorted, and arranged in chronological order to form a complete data set, which is the multi-source dynamic simulation total evaluation set. At the same time, further data analysis can be performed on this evaluation set, such as calculating statistical quantities such as the average value and standard deviation, to more deeply understand the operation performance of trainees. This multi-source dynamic simulation total evaluation set is like a detailed operation record file, recording the comprehensive evaluation information of trainees at different moments during the entire ECG examination operation process, reflecting the dynamic changes during the operation process. For example, after completing a simulated ECG examination operation, teachers find through analyzing this evaluation set that due to a relatively large electrode positioning error in the early stage of the operation, the time sequence simulation evaluation value of trainees is relatively low; while in the later stage of the operation, although the electrode positioning has improved, due to a deviation in the judgment of the heart rhythm, the heart rhythm urgency degree is relatively high, and the evaluation value is still not ideal. Based on this, teachers can formulate a targeted teaching plan to strengthen the training of trainees in electrode positioning and heart rhythm diagnosis, helping trainees improve their operation skills, solving the problem that there is a lack of comprehensive and dynamic evaluation of trainees' operation process in traditional teaching.

[0122] In one embodiment, the step of obtaining multi-modal evaluation data according to the multi-source data priority set includes:

[0123] S401. Build a spatio-temporal evolution feature extraction network for the multi-source data priority set, and use a multi-layer convolutional neural network to extract the local morphological features of the multi-source data priority set to obtain multi-source electrocardiogram waveform feature information;

[0124] S402. Obtain the temporal information of the multi-source electrocardiogram waveform feature information, and model the temporal information of the multi-source electrocardiogram waveform feature information to obtain the change trend of the heart rate and the time-dependent relationship between different waveforms;

[0125] S403. Obtain abnormal electrocardiogram waveform nodes according to the time-dependent relationship;

[0126] S404. Compare the abnormal electrocardiogram waveform nodes with the same type of features in the historical pathological data, and calculate the similarity of the abnormal electrocardiogram waveform nodes;

[0127] S405. Determine whether the similarity of the abnormal electrocardiogram waveform nodes exceeds a preset threshold;

[0128] If it exceeds, generate multi-modal evaluation data according to the abnormal electrocardiogram waveform nodes.

[0129] As described in the above steps S401 - S405, the present invention constructs a spatio - temporal evolution feature extraction network for a multi - source data priority set and uses a multi - layer convolutional neural network to process the multi - source data priority set. An intelligent data "analyzer" is constructed, and this network can deeply explore data in both time and space dimensions. The multi - layer convolutional neural network is like a delicate "feature detector", which slides and convolves on the data through different convolutional kernels, and extracts local morphological features layer by layer from the multi - source data priority set. After such processing, complex electrocardiogram data is transformed into more representative multi - source electrocardiogram waveform feature information. For example, when analyzing an electrocardiogram, it is difficult for traditional methods to detect the subtle jerks of the P wave on some leads. However, through this network and the convolutional neural network, these hidden local morphological features can be accurately extracted, providing a key basis for subsequent analysis, and solving the problem that traditional electrocardiogram analysis is difficult to comprehensively and accurately extract local features of complex waveforms. Then, after obtaining the multi - source electrocardiogram waveform feature information, the timing information therein is extracted, that is, the time - point information corresponding to each waveform feature. Then, methods such as time - series analysis models are used to model these timing information. This process is like drawing a "dynamic scroll" of cardiac electrical activity. Through the model output, the change trend of the heart rate at different time points can be clearly seen, as well as the time sequence and interdependence relationship between different waveforms such as the P wave, QRS complex, and T wave. For example, in simulated electrocardiogram monitoring, it is found through modeling that the patient's heart rate gradually increases, and at the same time, the time interval between the QRS complex and the T wave gradually shortens, which implies that there may be an abnormality in the cardiac electrical conduction system, helping doctors or trainees to more comprehensively understand the laws of cardiac electrophysiological activities from a dynamic perspective and solving the problem that traditional analysis only focuses on static waveforms and ignores time correlations. Then, based on the time - dependence relationship obtained above, some judgment rules and thresholds are set. For example, when the time when a certain waveform appears deviates greatly from the normal time - dependence relationship, or the waveform morphology changes abnormally at a specific time point, this waveform node is marked as an abnormal electrocardiogram waveform node. This step is like setting special "marking points" in a complex electrocardiogram "maze", which can accurately find out those abnormal nodes that may imply heart diseases from numerous waveforms. For example, in an electrocardiogram, under normal circumstances, the time interval between the QRS complex and the T wave is relatively stable. However, if it is found that the T wave appears significantly earlier and its morphology changes at a certain time point, according to the set rules, this T - wave node will be determined to be abnormal, providing a key clue for subsequent diagnosis and solving the problem of difficult accurate identification of abnormal waveforms in complex data. After finding the abnormal electrocardiogram waveform nodes, they are compared with similar features in historical pathological data. Samples similar to the current abnormal node features are selected from the historical pathological data, and then methods such as shape - matching algorithms and feature - vector similarity calculations are used to measure the similarity between the current abnormal node and the historical samples.This is like taking the newly discovered "puzzle pieces" and comparing them with the past "puzzle templates" to see how much can be matched. For example, the currently discovered abnormalities are ST-segment elevation and T-wave inversion. By comparison, it is found that they are similar to the typical electrocardiogram features of myocardial infarction. After calculation, the similarity reaches a certain value, which provides a direction for diagnosis and solves the problem that doctors lack experience and have difficulty judging the disease type when facing new abnormal waveforms. Next, a similarity threshold is set in advance based on a large number of clinical studies and experiences. The calculated similarity of the abnormal electrocardiogram waveform nodes is compared with this preset threshold. If the similarity exceeds the threshold, it indicates that the current abnormal situation is very likely related to the known disease characteristics. At this time, various characteristics of the abnormal waveform nodes, such as waveform morphology, time relationship, similarity, etc., are comprehensively considered to generate multi-modal evaluation data. These data cover multi-faceted information such as preliminary diagnosis, possible disease types, and disease severity. For example, the preset threshold is 70%, and the similarity between the current abnormal node and the myocardial infarction sample is 85%, exceeding the threshold. The system will generate multi-modal evaluation data, indicating that the patient may have myocardial infarction and is in the early stage. Doctors can formulate subsequent examination and treatment plans based on this, avoiding the subjectivity of diagnosis and making the diagnosis more standardized and objective.

[0130] In one embodiment, the step of obtaining an operation comprehensive score according to the multi-modal evaluation data includes:

[0131] S501. Extract key operation features from the multi-modal evaluation data through the adaptive feature selector of the digital twin, where the key operation features include the virtual electrode fitting index, the signal fidelity parameter, and the diagnostic logic consistency score;

[0132] S502. Based on the dynamic weight optimizer of the digital twin, adjust the weight distribution of the key operation features in real time according to the historical operation data of the trainee;

[0133] S503. Map the weighted key operation features to a standardized operation comprehensive score through the fuzzy comprehensive evaluation algorithm.

[0134] As described in the above steps S501 - S503, the present invention utilizes the adaptive feature selector of the digital twin to extract key operation features from multi-modal evaluation data. The multi-modal evaluation data contains various complex information during the operation of the trainee. The adaptive feature selector is like an intelligent filter. It automatically analyzes the patterns and correlations in the data based on machine learning algorithms and preset rules. For example, when analyzing electrode-related data, it comprehensively considers multiple aspects of information such as the position of the electrode in the virtual environment and the tightness of contact with the skin, and then calculates the virtual electrode fitting index to reflect the electrode fitting situation. For the signal fidelity parameter, it determines the value by comprehensively considering factors such as the noise level and waveform integrity of the electrocardiogram signal. The diagnostic logic consistency score is obtained by carefully comparing the diagnostic ideas of the trainee and the standard diagnostic process. For example, during a simulated operation by a trainee, the paste positions of some electrodes are not very accurate. The adaptive feature selector accurately identifies through the analysis of relevant data that the virtual electrode fitting index is relatively low, and at the same time, the signal fidelity parameter is also affected because the loose electrode contact causes slight interference and distortion of the signal. In addition, in the diagnostic session, although the trainee's diagnostic result is correct, some key steps are skipped, which also makes the diagnostic logic consistency score not ideal. Through the extraction of these key operation features, the teacher can clearly understand the specific problems in the trainee's operation, providing a clear direction for subsequent targeted teaching, effectively solving the problem that it is difficult to extract key information due to the complexity of data in traditional teaching evaluation. Then, based on the dynamic weight optimizer of the digital twin, the weight distribution of key operation features is adjusted in real time according to the trainee's historical operation data, which realizes the personalization of the evaluation. The dynamic weight optimizer continuously collects and deeply analyzes the trainee's historical operation data, which covers various aspects of information such as the previous comprehensive operation score, the specific performance of each key operation feature, and the trainee's progress during the learning process. For example, when trainee B first started learning, the electrode paste was not proficient, and both the virtual electrode fitting index and the signal fidelity parameter were relatively low. However, as learning progressed, these two indicators improved significantly, but there were more errors in the diagnostic logic. The dynamic weight optimizer, based on this historical data, uses data analysis algorithms and models to determine that for trainee B, the current diagnostic logic consistency score is more important for evaluating their operation level, so it appropriately increases the weight of this item. In this way, in the calculation of the subsequent comprehensive operation score, the performance in the diagnostic logic has a greater impact on the total score, more accurately reflecting the actual operation level of trainee B.This way of adjusting weights in real time according to the individual differences and learning dynamics of students avoids the drawbacks of the traditional fixed-weight evaluation method, fully considers the characteristics of each student, stimulates the learning enthusiasm of students, enables teachers to formulate more targeted teaching strategies according to the actual situation of students, and effectively solves the problem that traditional evaluation cannot adapt to the personalized needs of students. Finally, through the fuzzy comprehensive evaluation algorithm, the weighted key operation features are mapped into a standardized comprehensive operation score, which makes the evaluation results more objective, comprehensive and comparable. Before performing the fuzzy comprehensive evaluation, the weighted calculation is first performed on the key operation features after the weights are adjusted by the dynamic weight optimizer. For example, assume that the weight of the virtual electrode fitting index is 0.3, the weight of the signal fidelity parameter is 0.3, and the weight of the diagnostic logic consistency score is 0.4. If a student's virtual electrode fitting index score is 80 points, the signal fidelity parameter score is 75 points, and the diagnostic logic consistency score is 70 points, then the weighted scores are 24 points, 22.5 points and 28 points respectively. Next, the fuzzy comprehensive evaluation algorithm performs a comprehensive calculation on these weighted scores based on the preset fuzzy relation matrix and composition operator. It is like reasonably integrating the performances in different aspects and finally obtaining a standardized comprehensive operation score. This score is on a unified standard scale, which is convenient for comparing and analyzing the operation levels of different students. For example, when evaluating a group of students, student C has a relatively high virtual electrode fitting index of 90 points, but a relatively low diagnostic logic consistency score of 65 points and a signal fidelity parameter of 70 points; student D has a virtual electrode fitting index of 75 points, a relatively high diagnostic logic consistency score of 85 points, and a signal fidelity parameter of 70 points. After weighted calculation and processing by the fuzzy comprehensive evaluation algorithm, the comprehensive operation score of student C is 74 points, and that of student D is 76 points. Through this standardized score, not only can it be intuitively seen that the overall operation level of student D is slightly higher than that of student C, but also based on the scores of each key operation feature, the advantages and disadvantages of each student can be deeply analyzed to provide a reference for teaching, successfully solving the problems of one-sided evaluation of a single indicator and difficulty in comprehensively measuring multiple indicators in traditional evaluation.

[0135] As Figure 2 shown, the present invention also provides an electrocardiogram examination operating system for simulation teaching, including:

[0136] A first acquisition module 1, configured to acquire the operation information of a student in the electrocardiogram simulation teaching scenario, and construct a digital twin model of the student's electrocardiogram simulation operation according to the operation information;

[0137] The second acquisition module 2 is configured to obtain the simulated operation data of the physical teaching equipment according to the digital twin model, and generate dynamic simulated pathological parameters according to the simulated operation data. The dynamic simulated pathological parameters include an electrode positioning deviation sequence, a dynamic response characteristic of a pacing signal, and a virtual mapping relationship of a diagnosis conclusion;

[0138] The third acquisition module 3 is configured to obtain a multi-source dynamic simulation total evaluation set according to the dynamic simulated pathological parameters, and perform priority classification on the multi-source dynamic simulation total evaluation set through spatio-temporal correlation analysis to generate a multi-source data priority set;

[0139] The fourth acquisition module 4 is configured to obtain multi-modal evaluation data according to the multi-source data priority set;

[0140] The fifth acquisition module 5 is configured to obtain an operation comprehensive score according to the multi-modal evaluation data;

[0141] The push module 6 is configured to generate personalized simulated teaching scheduling information according to the operation comprehensive score, and push dynamically adapted electrocardiogram simulated teaching resources based on the simulated teaching scheduling information.

[0142] In one embodiment, the first acquisition module 1 includes:

[0143] The first acquisition unit is configured to obtain the three-dimensional spatial trajectory of the electrocardiogram electrode patch, the contact pressure of the electrocardiogram electrode patch, and the operation timing data of the electrocardiogram electrode patch according to the operation information;

[0144] The first mapping unit is configured to map the three-dimensional spatial trajectory into a virtual electrode positioning topological map based on a physical-virtual bidirectional mapping algorithm;

[0145] The second mapping unit is configured to map the contact pressure into a virtual signal strength parameter,

[0146] The third mapping unit is configured to map the operation timing data into a timestamp sequence of the digital twin;

[0147] The first fusion unit is configured to fuse the virtual electrode positioning topological map, the virtual signal strength parameter, and the timestamp sequence into a digital twin model.

[0148] In one embodiment, the second acquisition module 2 includes:

[0149] The second acquisition unit is configured to obtain a simulated pacing signal and a simulated electrode positioning set according to the simulated operation data, where the simulated electrode positioning set includes a plurality of simulated electrode positioning data;

[0150] The third acquisition unit is configured to obtain a pacing phase offset according to the simulated pacing signal;

[0151] A fourth acquisition unit, configured to acquire a dynamic response feature of a pacing signal according to the pacing phase offset, where the dynamic response feature of the pacing signal includes a signal frequency change degree, a signal amplitude fluctuation degree, and a waveform distortion degree;

[0152] A first generation unit, configured to generate simulated diagnosis information according to the signal frequency change degree, the signal amplitude fluctuation degree, and the waveform distortion degree;

[0153] A second generation unit, configured to compare the simulated diagnosis information with cases in a standard pathology library one by one, and generate a mapping relationship matrix including a misdiagnosis probability through semantic matching and feature similarity calculation;

[0154] A fifth acquisition unit, configured to acquire standard electrode positioning data, and acquire a simulated electrode positioning deviation corresponding to each simulated electrode positioning data according to the standard electrode positioning data;

[0155] A sixth acquisition unit, configured to acquire a time sequence arrangement according to the electrocardiogram electrode patch operation timing data;

[0156] A sorting unit, configured to sort all simulated electrode positioning deviations according to the time sequence arrangement to obtain an electrode positioning deviation sequence;

[0157] A second fusion unit, configured to perform comprehensive fusion according to the electrode positioning deviation sequence, the dynamic response feature of the pacing signal, and the mapping relationship matrix to generate dynamic simulated pathological parameters

[0158] In one embodiment, the third acquisition module 3 includes:

[0159] A third generation unit, configured to generate an electrode positioning error heat map according to the electrode positioning deviation sequence through a spatial interpolation algorithm;

[0160] A seventh acquisition unit, configured to acquire electrode positioning error distribution information according to the electrode positioning error heat map;

[0161] An eighth acquisition unit, configured to acquire an error distribution density corresponding to each time sequence according to the electrode positioning error distribution information;

[0162] A ninth acquisition unit, configured to acquire heart rhythm feature data according to the dynamic response feature of the pacing signal;

[0163] A tenth acquisition unit, configured to generate an arrhythmia complexity spectrum through a spectrum analysis algorithm according to the heart rhythm feature data;

[0164] An eleventh acquisition unit, configured to acquire a heart rhythm urgency corresponding to each time sequence according to the arrhythmia complexity spectrum;

[0165] A twelfth acquisition unit, configured to obtain a timing simulation evaluation value corresponding to each timing according to the error distribution density and the heart rhythm urgency;

[0166] A thirteenth acquisition unit, configured to obtain a multi-source dynamic simulation total evaluation set according to the timing simulation evaluation values corresponding to all timings.

[0167] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, value library, or other medium provided in the present application and used in the embodiments can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0168] It should be noted that in this article, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, device, article or method. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, device, article or method including that element.

[0169] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent results or equivalent process transformations made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, are equally included in the patent protection scope of the present invention.

Claims

1. An electrocardiogram examination operation method for simulation teaching, characterized in that, Including: Obtaining the operation information of trainees in the electrocardiogram simulation teaching scenario; Obtaining the three-dimensional spatial trajectory of the electrocardiogram electrode patch, the contact pressure of the electrocardiogram electrode patch, and the operation timing data of the electrocardiogram electrode patch according to the operation information; Based on the physical-virtual bidirectional mapping algorithm, mapping the three-dimensional spatial trajectory into a virtual electrode positioning topology map; Mapping the contact pressure into a virtual signal strength parameter, Mapping the operation timing data into a timestamp sequence of the digital twin; Fusing the virtual electrode positioning topology map, the virtual signal strength parameter, and the timestamp sequence into a digital twin model; Obtaining the simulated operation data of the physical teaching equipment according to the digital twin model, and generating dynamic simulated pathological parameters according to the simulated operation data, where the dynamic simulated pathological parameters include an electrode positioning deviation sequence, a pacing signal dynamic response characteristic, and a diagnostic conclusion virtual mapping relationship; Obtaining a multi-source dynamic simulation total evaluation set according to the dynamic simulated pathological parameters, and performing priority classification on the multi-source dynamic simulation total evaluation set through spatio-temporal correlation analysis to generate a multi-source data priority set; Obtaining multi-modal evaluation data according to the multi-source data priority set; Obtaining an operation comprehensive score according to the multi-modal evaluation data; Generating personalized simulation teaching scheduling information according to the operation comprehensive score, and pushing dynamically adapted electrocardiogram simulation teaching resources based on the simulation teaching scheduling information.

2. The electrocardiogram examination operation method for simulation teaching according to claim 1, wherein The step of generating dynamic simulated pathological parameters according to the simulated operation data includes: Obtaining a simulated pacing signal and a simulated electrode positioning set according to the simulated operation data, where the simulated electrode positioning set includes a plurality of simulated electrode positioning data; Obtaining a pacing phase offset according to the simulated pacing signal; Obtaining a pacing signal dynamic response characteristic according to the pacing phase offset, where the pacing signal dynamic response characteristic includes a signal frequency change degree, a signal amplitude fluctuation degree, and a waveform distortion degree; Generating simulated diagnostic information according to the signal frequency change degree, the signal amplitude fluctuation degree, and the waveform distortion degree; Comparing the simulated diagnostic information with the cases in the standard pathological database one by one, and generating a mapping relationship matrix including a misdiagnosis probability through semantic matching and feature similarity calculation; Obtaining standard electrode positioning data, and obtaining the simulated electrode positioning deviation corresponding to each simulated electrode positioning data according to the standard electrode positioning data; Obtaining a time sequence arrangement according to the electrocardiogram electrode patch operation timing data; Sorting all the simulated electrode positioning deviations according to the time sequence arrangement to obtain an electrode positioning deviation sequence; Performing comprehensive fusion according to the electrode positioning deviation sequence, the pacing signal dynamic response characteristic, and the mapping relationship matrix to generate dynamic simulated pathological parameters.

3. The electrocardiogram examination operation method for simulation teaching according to claim 1, characterized in that, The step of obtaining a multi-source dynamic simulation total evaluation set according to the dynamic simulated pathological parameters includes: Generating an electrode positioning error heat map according to the electrode positioning deviation sequence through a spatial interpolation algorithm; Obtaining electrode positioning error distribution information according to the electrode positioning error heat map; Obtaining the error distribution density corresponding to each timing according to the electrode positioning error distribution information; Obtaining cardiac rhythm characteristic data according to the pacing signal dynamic response characteristic; Generate an arrhythmia complexity spectrum through a spectral analysis algorithm based on the cardiac rhythm characteristic data; Obtain the cardiac rhythm urgency corresponding to each time series according to the arrhythmia complexity spectrum; Obtain the time series simulation evaluation value corresponding to each time series according to the error distribution density and the cardiac rhythm urgency; Obtain a multi-source dynamic simulation total evaluation set according to the time series simulation evaluation values corresponding to all time series.

4. The electrocardiogram examination operation method for simulation teaching according to claim 1, wherein, The step of obtaining multi-modal evaluation data according to the multi-source data priority set includes: Build a spatio-temporal evolution feature extraction network for the multi-source data priority set, and use a multi-layer convolutional neural network to extract the local morphological features of the multi-source data priority set to obtain multi-source electrocardiogram waveform feature information; Obtain the time series information of the multi-source electrocardiogram waveform feature information, and model the time series information of the multi-source electrocardiogram waveform feature information to obtain the change trend of the heart rate and the time-dependent relationship between different waveforms; Obtain abnormal electrocardiogram waveform nodes according to the time-dependent relationship; Compare the abnormal electrocardiogram waveform nodes with the same type of features in the historical pathological data, and calculate the similarity of the abnormal electrocardiogram waveform nodes; Judge whether the similarity of the abnormal electrocardiogram waveform nodes exceeds a preset threshold; If it exceeds, generate multi-modal evaluation data according to the abnormal electrocardiogram waveform nodes.

5. A method for electrocardiogram examination operation in simulation teaching according to claim 1, characterized in that The step of obtaining an operation comprehensive score according to the multi-modal evaluation data includes: Extract key operation features from the multi-modal evaluation data through the adaptive feature selector of the digital twin, where the key operation features include the virtual electrode fitting index, the signal fidelity parameter, and the diagnostic logic consistency score; Based on the dynamic weight optimizer of the digital twin, adjust the weight distribution of the key operation features in real time according to the historical operation data of the trainee; Map the weighted key operation features to a standardized operation comprehensive score through a fuzzy comprehensive evaluation algorithm.

6. An electrocardiogram examination operating system for simulation teaching, characterized in that, Include: A first acquisition module for acquiring the operation information of the trainee in the electrocardiogram simulation teaching scenario; Obtain the three-dimensional space trajectory of the electrocardiogram electrode patch, the contact pressure of the electrocardiogram electrode patch, and the operation time series data of the electrocardiogram electrode patch according to the operation information; Based on the physical-virtual bidirectional mapping algorithm, map the three-dimensional space trajectory to a virtual electrode positioning topological map; Map the contact pressure to a virtual signal strength parameter, Map the operation time series data to the timestamp sequence of the digital twin; Fuse the virtual electrode positioning topological map, the virtual signal strength parameter, and the timestamp sequence into a digital twin model; A second acquisition module for obtaining the simulation operation data of the physical teaching equipment according to the digital twin model, and generating dynamic simulation pathological parameters according to the simulation operation data, where the dynamic simulation pathological parameters include an electrode positioning deviation sequence, a pacing signal dynamic response feature, and a diagnostic conclusion virtual mapping relationship; A third acquisition module for obtaining a multi-source dynamic simulation total evaluation set according to the dynamic simulation pathological parameters, and performing priority classification on the multi-source dynamic simulation total evaluation set through spatio-temporal correlation analysis to generate a multi-source data priority set; A fourth acquisition module for obtaining multi-modal evaluation data according to the multi-source data priority set; A fifth acquisition module, configured to obtain an operation comprehensive score according to the multimodal evaluation data; A push module, configured to generate personalized simulated teaching scheduling information according to the operation comprehensive score, and push dynamically adapted electrocardiogram simulated teaching resources based on the simulated teaching scheduling information.

7. An electrocardiogram examination operating system for simulation teaching according to claim 6, characterized in that, The second acquisition module includes: A second acquisition unit, configured to obtain a simulated pacing signal and a simulated electrode positioning set according to the simulated operation data, wherein the simulated electrode positioning set includes a plurality of simulated electrode positioning data; A third acquisition unit, configured to obtain a pacing phase offset according to the simulated pacing signal; A fourth acquisition unit, configured to obtain a pacing signal dynamic response feature according to the pacing phase offset, wherein the pacing signal dynamic response feature includes a signal frequency change degree, a signal amplitude fluctuation degree, and a waveform distortion degree; A first generation unit, configured to generate simulated diagnosis information according to the signal frequency change degree, the signal amplitude fluctuation degree, and the waveform distortion degree; A second generation unit, configured to compare the simulated diagnosis information with cases in a standard pathology library one by one, and generate a mapping relationship matrix including a misdiagnosis probability through semantic matching and feature similarity calculation; A fifth acquisition unit, configured to obtain standard electrode positioning data, and obtain a simulated electrode positioning deviation corresponding to each simulated electrode positioning data according to the standard electrode positioning data; A sixth acquisition unit, configured to obtain a time sequence arrangement according to the electrocardiogram electrode patch operation timing data; A sorting unit, configured to sort all simulated electrode positioning deviations according to the time sequence arrangement to obtain an electrode positioning deviation sequence; A second fusion unit, configured to perform comprehensive fusion according to the electrode positioning deviation sequence, the pacing signal dynamic response feature, and the mapping relationship matrix to generate dynamic simulated pathological parameters.

8. An electrocardiogram examination operating system for simulation teaching according to claim 6, characterized in that, The third acquisition module includes: A third generation unit, configured to generate an electrode positioning error heat map according to the electrode positioning deviation sequence through a spatial interpolation algorithm; A seventh acquisition unit, configured to obtain electrode positioning error distribution information according to the electrode positioning error heat map; An eighth acquisition unit, configured to obtain an error distribution density corresponding to each time sequence according to the electrode positioning error distribution information; A ninth acquisition unit, configured to obtain heart rhythm feature data according to the pacing signal dynamic response feature; A tenth acquisition unit, configured to generate an arrhythmia complexity spectrum according to the heart rhythm feature data through a spectrum analysis algorithm; An eleventh acquisition unit, configured to obtain a heart rhythm urgency corresponding to each time sequence according to the arrhythmia complexity spectrum; A twelfth acquisition unit, configured to obtain a time sequence simulation evaluation value corresponding to each time sequence according to the error distribution density and the heart rhythm urgency; A thirteenth acquisition unit, configured to obtain a multi-source dynamic simulation total evaluation set according to the time sequence simulation evaluation values corresponding to all time sequences.

Images