Electrocardiosignal visualization enhancement method and system, electronic equipment and medium

Abstract

The invention discloses an electrocardiosignal visualization enhancement method and system, electronic equipment and a medium. The method comprises the steps that first electrocardiosignals of multiple leads in a preset time period are obtained; extracting a frequency characteristic, a multi-lead consistency characteristic, a dynamic characteristic and an enhanced signal characteristic of the first electrocardiosignal of the plurality of leads; mapping the frequency feature into a first pseudo-color map, mapping the multi-lead consistency feature into a second pseudo-color map, and mapping the dynamic feature and the enhanced signal feature into a third pseudo-color map; and constructing a three-dimensional trajectory diagram according to the first pseudo-color diagram, the second pseudo-color diagram and the third pseudo-color diagram. According to the method and the device, the electrocardiogram signal interpretation precision and efficiency can be improved, so that the electrocardiogram signal visualization effect is improved.

Description

Technical Field

[0001] The present application relates to the technical field of electrocardiogram signal processing, and in particular to an electrocardiogram signal visualization enhancement method, system, electronic equipment and medium. Background Art

[0002] The electrocardiogram (ECG) is a non-invasive and easily accessible tool for recording the electrical activity of the heart. Each cardiac cycle can be identified by its characteristic periodic waveform, and changes in the ECG signal are usually closely related to changes in cardiac dynamics. Traditionally, ECG examinations are usually performed by simple visual analysis of one-dimensional (1D) time series waveforms. Studies have shown that visualization technology is significantly effective in helping clinicians interpret ECGs.

[0003] However, there are some difficulties and challenges in traditional ECG signal visualization methods. Since ECGs usually contain multiple lead data, each lead provides information about the heart's electrical activity from a different perspective. The display of a single color or grayscale image limits the observation of different signal strengths and differences between leads, resulting in feature details that may not be easily perceived due to differences between leads or signal complexity. Therefore, traditional ECG signal visualization methods have relatively low accuracy and efficiency in interpreting ECG signals, reducing the visualization effect of ECG signals. Summary of the invention

[0004] The present application aims to propose an electrocardiogram (ECG) signal visualization enhancement method, system, electronic device and medium, which can improve the accuracy and efficiency of interpreting ECG signals, thereby improving the visualization effect of ECG signals.

[0005] In a first aspect, an embodiment of the present application provides a method for visualizing and enhancing an electrocardiogram signal, the method comprising: Acquire first electrocardiogram signals of multiple leads within a preset time period; Extracting frequency features, multi-lead consistency features, dynamic features and enhanced signal features of the first ECG signals of the multiple leads, wherein the frequency feature is the frequency distribution of the first ECG signal of each lead in the frequency domain, the multi-lead consistency feature is a comprehensive measurement value between the synchronization of the first ECG signals of two different leads in the time domain and the phase difference of the first ECG signals of two different leads in the frequency domain, the dynamic feature is the first-order derivative and the second-order derivative of the first ECG signal of each lead, and the enhanced signal feature is the ECG signal after weighting the first ECG signal of each lead through the attention mechanism; Mapping the frequency feature into a first pseudo color map, mapping the multi-lead consistency feature into a second pseudo color map, and mapping the dynamic feature and the enhanced signal feature into a third pseudo color map; A three-dimensional trajectory map is constructed according to the first pseudo-color map, the second pseudo-color map, and the third pseudo-color map.

[0006] Compared with the prior art, the first aspect of the present application has the following beneficial effects: The method obtains the first ECG signals of multiple leads within a preset time period; extracts the frequency characteristics, multi-lead consistency characteristics, dynamic characteristics and enhanced signal characteristics of the first ECG signals of multiple leads, wherein the frequency characteristics are the frequency distribution of the first ECG signal of each lead in the frequency domain, the multi-lead consistency characteristics are the comprehensive measurement value between the synchronization of the first ECG signals of two different leads in the time domain and the phase difference of the first ECG signals of two different leads in the frequency domain, the dynamic characteristics are the first-order derivative and the second-order derivative of the first ECG signal of each lead, and the enhanced signal characteristics are the ECG signals after the first ECG signals of each lead are weighted by the attention mechanism; maps the frequency characteristics into a first pseudo-color image, maps the multi-lead consistency characteristics into a second pseudo-color image, and maps the dynamic characteristics and the enhanced signal characteristics into a third pseudo-color image; and constructs a three-dimensional trajectory image according to the first pseudo-color image, the second pseudo-color image and the third pseudo-color image. In this way, by extracting various types of features from ECG signals of multiple leads, mapping the various types of features into multiple color images, and finally constructing a three-dimensional trajectory map based on the multiple color images, the accuracy and efficiency of interpreting ECG signals are improved, and the advantages of pseudo-color in enhancing signal distinction and feature visibility are fully utilized, thereby improving the visualization effect of ECG signals.

[0007] In some embodiments, extracting frequency features, multi-lead consistency features, dynamic features, and enhanced signal features of the first electrocardiogram signals of the multiple leads includes: Calculating the synchronization of the first electrocardiogram signals of two different leads in the time domain to obtain a first result; Calculating the phase difference between the first electrocardiogram signals of the two different leads in the frequency domain to obtain a second result; According to the first result and the second result, a comprehensive metric value between the first electrocardiogram signals of the two different leads is calculated, and the comprehensive metric value is used as a multi-lead consistency feature.

[0008] In some implementations, calculating the comprehensive metric value between the first electrocardiogram signals of the two different leads according to the first result and the second result includes: ; in, Represents a comprehensive measurement value, represents the weight coefficient, represents the total length of the lag time window for the cross-correlation, represents the first result obtained by calculating the synchronization of the first ECG signals of two different leads in the time domain through the cross-correlation function, represents the lag time window length, represents the number of frequency components of Fourier transform, It represents the second result obtained by calculating the phase difference of the first ECG signals of two different leads in the frequency domain. Indicates frequency components.

[0009] In some embodiments, extracting frequency features, multi-lead consistency features, dynamic features, and enhanced signal features of the first electrocardiogram signals of the multiple leads includes: Calculating the first-order derivative of the first electrocardiogram signal of each lead; Calculating the second-order derivative of the first electrocardiogram signal of each lead; The first-order derivative and the second-order derivative of the first electrocardiogram signal of each lead are used as dynamic features.

[0010] In some embodiments, extracting frequency features, multi-lead consistency features, dynamic features, and enhanced signal features of the first electrocardiogram signals of the multiple leads includes: Segmenting the first electrocardiogram signal of each lead within the preset time period into second electrocardiogram signals of a plurality of time segments; The second ECG signals of the multiple time segments are input into a multi-head attention mechanism for attention weighting to obtain enhanced signal features corresponding to the first ECG signal of each lead.

[0011] In some embodiments, the inputting the second ECG signals of the multiple time segments into a multi-head attention mechanism for attention weighting to obtain enhanced signal features corresponding to the first ECG signals of each lead includes: ; in, Indicates enhanced signal characteristics, Indicates the sampling time point The first ECG signal at Indicates the number of time slices, represents the Dirac function, Indicates the time point corresponding to the maximum attention weight output by the multi-head attention mechanism, Indicates Time Segment The value of the maximum attention weight in , Indicates The time point that the multi-head attention mechanism pays most attention to in the time segment The corresponding second ECG signal.

[0012] In some embodiments, mapping the dynamic feature and the enhanced signal feature into a third pseudo color map comprises: Summing the first-order derivative of the first electrocardiogram signal of each lead, the second-order derivative of the first electrocardiogram signal of each lead, and the enhanced signal feature to obtain a combined feature; The combined features are mapped into a third pseudo-color map.

[0013] In a second aspect, an embodiment of the present application further provides an electrocardiogram signal visualization enhancement system, the system comprising: A data acquisition unit, used to acquire first electrocardiogram signals of multiple leads within a preset time period; A feature extraction unit, used to extract frequency features, multi-lead consistency features, dynamic features and enhanced signal features of the first ECG signals of the multiple leads, wherein the frequency feature is the frequency distribution of the first ECG signal of each lead in the frequency domain, the multi-lead consistency feature is a comprehensive measure of the synchronization of the first ECG signals of two different leads in the time domain and the phase difference of the first ECG signals of two different leads in the frequency domain, the dynamic feature is the first-order derivative and the second-order derivative of the first ECG signal of each lead, and the enhanced signal feature is the ECG signal after the first ECG signal of each lead is weighted by the attention mechanism; A feature mapping unit, used for mapping the frequency feature into a first pseudo color map, mapping the multi-lead consistency feature into a second pseudo color map, and mapping the dynamic feature and the enhanced signal feature into a third pseudo color map; The trajectory map construction unit is used to construct a three-dimensional trajectory map according to the first pseudo-color map, the second pseudo-color map and the third pseudo-color map.

[0014] In a third aspect, an embodiment of the present application also provides an electronic device, comprising at least one control processor and a memory for communicating with the at least one control processor; the memory stores instructions that can be executed by the at least one control processor, and the instructions are executed by the at least one control processor so that the at least one control processor can execute a method for visualizing and enhancing an electrocardiogram signal as described above.

[0015] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the electrocardiogram signal visualization enhancement method as described above.

[0016] It can be understood that the beneficial effects of the second to fourth aspects compared with the related art are the same as the beneficial effects of the first aspect compared with the related art. Please refer to the relevant description in the first aspect, and no further details will be given here. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which: Figure 1 is a flowchart of an embodiment of the method for visualizing and enhancing electrocardiogram signals provided by the present application; Figure 2 It is a flowchart of the best embodiment of the method for visualizing and enhancing electrocardiogram signals provided by the present application; Figure 3 is a structural schematic diagram of an embodiment of an electrocardiogram signal visualization enhancement system provided by the present application; Figure 4 It is a schematic diagram of the structure of an embodiment of the electronic device provided by the present application. DETAILED DESCRIPTION

[0018] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be understood as limiting the present application.

[0019] In the description of this application, if there is a description of first, second, etc., it is only for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0020] In the description of the present application, it should be understood that the descriptions involving orientation, such as the orientation or positional relationship indicated as up, down, etc., are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0021] In the description of this application, it should be noted that, unless otherwise clearly defined, terms such as setting, installing, connecting, etc. should be understood in a broad sense, and technicians in the relevant technical field can reasonably determine the specific meaning of the above terms in this application based on the specific content of the technical solution.

[0022] The electrocardiogram (ECG) is a non-invasive and easily accessible tool for recording the electrical activity of the heart. Each cardiac cycle can be identified by its characteristic periodic waveform, and changes in the ECG signal are usually closely related to changes in cardiac dynamics. Traditionally, ECG examinations are usually performed by simple visual analysis of one-dimensional (1D) time series waveforms. Studies have shown that visualization technology is significantly effective in helping clinicians interpret ECGs.

[0023] However, there are some difficulties and challenges in traditional ECG signal visualization methods. Since ECGs usually contain multiple lead data, each lead provides information about the heart's electrical activity from a different perspective. The display of a single color or grayscale image limits the observation of different signal strengths and differences between leads, resulting in feature details that may not be easily perceived due to differences between leads or signal complexity. Therefore, traditional ECG signal visualization methods have relatively low accuracy and efficiency in interpreting ECG signals, reducing the visualization effect of ECG signals.

[0024] In order to solve the problem that the traditional ECG signal visualization method has low accuracy and efficiency in interpreting ECG signals, thereby reducing the visualization effect of ECG signals, the present application proposes an ECG signal visualization enhancement method, system, electronic device and medium.

[0025] Reference Figure 1 The present application embodiment provides a method for visualizing and enhancing an electrocardiogram signal, and the method comprises the following steps: Step S100, acquiring first electrocardiogram signals of multiple leads within a preset time period; Step S200, extracting frequency features, multi-lead consistency features, dynamic features, and enhanced signal features of the first ECG signals of multiple leads, wherein the frequency feature is the frequency distribution of the first ECG signal of each lead in the frequency domain, the multi-lead consistency feature is a comprehensive measurement value between the synchronization of the first ECG signals of two different leads in the time domain and the phase difference of the first ECG signals of two different leads in the frequency domain, the dynamic feature is the first-order derivative and the second-order derivative of the first ECG signal of each lead, and the enhanced signal feature is the ECG signal after the first ECG signal of each lead is weighted by the attention mechanism; Step S300, mapping the frequency feature into a first pseudo color map, mapping the multi-lead consistency feature into a second pseudo color map, and mapping the dynamic feature and the enhanced signal feature into a third pseudo color map; Step S400: construct a three-dimensional trajectory map according to the first pseudo-color map, the second pseudo-color map and the third pseudo-color map.

[0026] In this embodiment, by acquiring the first ECG signals of multiple leads within a preset time period; extracting the frequency characteristics, multi-lead consistency characteristics, dynamic characteristics and enhanced signal characteristics of the first ECG signals of multiple leads, wherein the frequency characteristics are the frequency distribution of the first ECG signal of each lead in the frequency domain, the multi-lead consistency characteristics are the comprehensive measurement value between the synchronization of the first ECG signals of two different leads in the time domain and the phase difference of the first ECG signals of two different leads in the frequency domain, the dynamic characteristics are the first-order derivatives and second-order derivatives of the first ECG signals of each lead, and the enhanced signal characteristics are the weighted ECG signals of the first ECG signals of each lead through the attention mechanism; mapping the frequency characteristics into a first pseudo-color image, mapping the multi-lead consistency characteristics into a second pseudo-color image, and mapping the dynamic characteristics and the enhanced signal characteristics into a third pseudo-color image; constructing a three-dimensional trajectory image according to the first pseudo-color image, the second pseudo-color image and the third pseudo-color image. In this way, by extracting various types of features from ECG signals of multiple leads, mapping the various types of features into multiple color images, and finally constructing a three-dimensional trajectory map based on the multiple color images, the accuracy and efficiency of interpreting ECG signals are improved, and the advantages of pseudo-color in enhancing signal distinction and feature visibility are fully utilized, thereby improving the visualization effect of ECG signals.

[0027] The above-mentioned acquisition of the first ECG signals of multiple leads within the preset time period may be performed by using a suitable acquisition device to perform sampling at a preset Hertz (HZ) for a period of time (the preset time period may be a manually set time period), and sampling the first ECG signals of multiple leads.

[0028] The above-mentioned suitable acquisition device may be a device used for sampling electrocardiogram signals in the prior art, and this embodiment does not impose any specific limitation thereto.

[0029] The above-mentioned mapping of frequency characteristics into a first pseudo-color image, mapping of multi-lead consistency characteristics into a second pseudo-color image, and mapping of dynamic characteristics and enhanced signal characteristics into a third pseudo-color image can be achieved by using a pseudo-color mapping function to map the frequency characteristics into a first pseudo-color image, mapping the multi-lead consistency characteristics into a second pseudo-color image, and mapping the dynamic characteristics and enhanced signal characteristics into a third pseudo-color image.

[0030] The three-dimensional trajectory map is constructed according to the first pseudo-color map, the second pseudo-color map and the third pseudo-color map, and the first pseudo-color map, the second pseudo-color map and the third pseudo-color map are merged into one map, that is, the three-dimensional trajectory map is constructed. For example, the existing software Python can be used to realize the construction of the three-dimensional trajectory map.

[0031] In some embodiments, extracting frequency features, multi-lead consistency features, dynamic features, and enhanced signal features of the first electrocardiogram signals of multiple leads includes: Calculating the synchronization of the first electrocardiogram signals of two different leads in the time domain to obtain a first result; Calculate the phase difference between the first electrocardiogram signals of two different leads in the frequency domain to obtain a second result; According to the first result and the second result, a comprehensive metric value between the first electrocardiogram signals of two different leads is calculated, and the comprehensive metric value is used as a multi-lead consistency feature.

[0032] In this embodiment, the first result is obtained by calculating the synchronization of the first ECG signals of two different leads in the time domain; the second result is obtained by calculating the phase difference of the first ECG signals of two different leads in the frequency domain; based on the first result and the second result, the comprehensive measurement value between the first ECG signals of two different leads is calculated, and the comprehensive measurement value is used as the multi-lead consistency feature. In this way, by adjusting the influence of synchronization and phase difference on the comprehensive measurement value, the comprehensive measurement value reflects synchronization and phase difference at the same time, synchronization provides similarity measurement in the time domain, and phase difference provides difference in the frequency domain, and the correlation between ECG signals of different leads is analyzed, which better improves the understanding and analysis ability of complex ECG signals.

[0033] In some implementations, calculating a comprehensive metric value between first ECG signals of two different leads according to the first result and the second result includes: ; in, Represents a comprehensive measurement value, represents the weight coefficient, represents the total length of the lag time window for the cross-correlation, represents the first result obtained by calculating the synchronization of the first ECG signals of two different leads in the time domain through the cross-correlation function, represents the lag time window length, represents the number of frequency components of Fourier transform, It represents the second result obtained by calculating the phase difference of the first ECG signals of two different leads in the frequency domain. Indicates frequency components.

[0034] In some embodiments, extracting frequency features, multi-lead consistency features, dynamic features, and enhanced signal features of the first electrocardiogram signals of multiple leads includes: Calculate the first derivative of the first ECG signal of each lead; Calculate the second-order derivative of the first ECG signal of each lead; The first-order derivative and the second-order derivative of the first ECG signal of each lead are taken as dynamic features.

[0035] In this embodiment, the first-order derivative of the first ECG signal of each lead is calculated; the second-order derivative of the first ECG signal of each lead is calculated; and the first-order derivative and the second-order derivative of the first ECG signal of each lead are used as dynamic features. In this way, the local rate of change of the ECG signal is captured by the first-order derivative to detect the location of sudden or rapid changes in the signal, and the dynamic trend of the signal is reflected by the second-order derivative to help identify the overall acceleration or deceleration of the signal, thereby further improving the ability to understand and analyze complex ECG signals.

[0036] In some embodiments, extracting frequency features, multi-lead consistency features, dynamic features, and enhanced signal features of the first electrocardiogram signals of multiple leads includes: Segmenting the first electrocardiogram signal of each lead within a preset time period into second electrocardiogram signals of a plurality of time segments; The second ECG signals of multiple time segments are input into the multi-head attention mechanism for attention weighting to obtain the enhanced signal features corresponding to the first ECG signal of each lead.

[0037] In this embodiment, the first ECG signal of each lead within a preset time period is divided into second ECG signals of multiple time segments; the second ECG signals of multiple time segments are input into the multi-head attention mechanism for attention weighting, and the enhanced signal features corresponding to the first ECG signal of each lead are obtained. In this way, using the multi-head attention mechanism, different weights are assigned to ECG signals at different time points, highlighting the ECG signals corresponding to important time segments, and further improving the ability to understand and analyze complex ECG signals.

[0038] In some embodiments, the second ECG signals of multiple time segments are input into a multi-head attention mechanism for attention weighting to obtain enhanced signal features corresponding to the first ECG signal of each lead, including: ; in, Indicates enhanced signal characteristics, Indicates the sampling time point The first ECG signal at Indicates the number of time slices, represents the Dirac function, Indicates the time point corresponding to the maximum attention weight output by the multi-head attention mechanism, Indicates Time Segment The value of the maximum attention weight in , Indicates The time point that the multi-head attention mechanism pays most attention to in the time segment The corresponding second ECG signal.

[0039] In some embodiments, mapping the dynamic features and the enhanced signal features into a third pseudo color map comprises: The first-order derivative of the first electrocardiogram signal of each lead, the second-order derivative of the first electrocardiogram signal of each lead, and the enhanced signal feature are summed to obtain a combined feature; The combined features are mapped into a third pseudo color map.

[0040] In this embodiment, the first-order derivative of the first ECG signal of each lead, the second-order derivative of the first ECG signal of each lead, and the enhanced signal feature are summed to obtain a combined feature, and the combined feature is mapped to a third pseudo-color image. In this way, the enhanced signal feature, the rate of change (i.e., the first-order derivative) and the dynamic trend (i.e., the second-order derivative) can be balanced. By mapping the combined feature to the third pseudo-color image, it is possible to intuitively see which time segments are given higher attention through the image, thereby enhancing the visual expression of the ECG signal features.

[0041] To facilitate understanding by those skilled in the art, a set of best embodiments is provided below: The electrocardiogram (ECG) is a non-invasive and easily accessible tool for recording the electrical activity of the heart. Each cardiac cycle can be identified by its characteristic periodic waveform, and changes in the ECG signal are usually closely related to changes in cardiac dynamics. Traditionally, ECG examinations are usually performed through simple visual analysis of one-dimensional (1D) time series waveforms. Studies have shown that visualization technology has a significant effect in helping clinicians interpret ECGs, not only improving the detection rate of heart attacks, but also helping non-professionals identify when ECG signals differ from the normal baseline so that medical help can be sought in a timely manner.

[0042] However, there are some difficulties and challenges in the traditional ECG signal visualization method. First, ECG signals are easily interfered by various physiological and environmental noises, such as electromyographic interference, baseline drift, and electromagnetic noise, which increases the difficulty of accurate interpretation. Secondly, the ECG usually contains multiple lead data, each lead provides cardiac electrical activity information from a different perspective, and the display method of a single color or grayscale image limits the observation of different signal intensities and differences between leads, resulting in feature details that may not be easily perceived due to differences between leads or signal complexity. In addition, as the amount of ECG data increases, manual visual inspection is inefficient and difficult to meet the needs of large-scale data analysis.

[0043] In order to solve the above problems, this embodiment proposes a method for visualizing and enhancing ECG signals based on multi-step pseudo-color mapping and 3D trajectory diagram (i.e., three-dimensional trajectory diagram). Multi-step pseudo-color mapping technology maps features at different levels into colors, making key waveform features more intuitive and visible, helping to distinguish and identify specific ECG activities. The 3D trajectory diagram provides a more comprehensive perspective by displaying time, frequency distribution, dynamic enhancement signals, and lead consistency information in three-dimensional space, so that the differences and correlations between different leads can be clearly presented.

[0044] This embodiment provides an innovative method for visualizing and enhancing ECG signals, which has important technical and practical value. This method not only improves the accuracy and efficiency of ECG signal interpretation, but can also be widely used in scientific research, clinical monitoring, and smart health equipment, providing an effective tool for in-depth analysis and real-time monitoring of ECG data.

[0045] The goal of this embodiment is to provide a new method for visualizing and enhancing ECG signals by comprehensively applying multi-step pseudo-color mapping and 3D trajectory diagrams. This method can fully utilize the advantages of pseudo-color in enhancing signal differentiation and feature visibility, as well as the ability of 3D trajectory diagrams in displaying multi-dimensional data and highlighting the dynamic changes of waveforms, and significantly improve the visualization effect in multi-dimensional analysis and feature recognition of complex ECG signals. Specifically, it includes the following contents: 1. Define the problem.

[0046] Data preprocessing: Preprocess the ECG signal, including denoising, filtering, and normalization, to remove physiological and environmental noise and improve signal clarity and accuracy.

[0047] Multi-parameter feature extraction: Extract multiple parameters from ECG signals: frequency features, synchronization and phase difference between different leads (i.e., multi-lead consistency features), local change rate and overall trend change of ECG signals (i.e., dynamic features), and signal attention weight enhancement features (i.e., enhanced signal features).

[0048] Multi-step pseudo color mapping: multi-dimensional features (i.e., multi-parameter features) are mapped to the visual image through a pseudo color function.

[0049] 3D trajectory map construction: Pseudo-color information and features are integrated into the 3D trajectory map to show the dynamic changes of the signal in time, frequency and signal enhancement.

[0050] Performance evaluation: Design appropriate technical indicators, such as color contrast, image clarity, and dynamic effects, and collect feedback through user experience surveys to evaluate the practicality and application value of visualization methods.

[0051] 2. Objectives and constraints.

[0052] (1) Objective:

[0053] The goal of this embodiment is to provide an innovative method for visualizing and enhancing ECG signal features, based on multi-step pseudo-color mapping and 3D trajectory map technology, to achieve efficient and intuitive display of ECG signals. Taking advantage of the technology, the ECG signal is frequency analyzed, multi-head attention signal enhancement is used, and synchronization with other leads is performed through multi-step pseudo-color mapping technology; at the same time, a 3D trajectory map is used to display time, dynamically enhance signals, and lead consistency for multi-layer coloring, providing a multi-dimensional signal perspective. This method not only improves the ability to understand and analyze complex ECG signals, but also can adapt to different signal forms and noise conditions, and improve the accuracy and robustness of visualization. Experiments have proved that this embodiment has broad application prospects in clinical ECG analysis, intelligent health monitoring, and large-scale data research.

[0054] (2) Constraints.

[0055] Hardware constraints: Due to the computational complexity of 3D trajectory maps and pseudo-color rendering, the computing resource limitations of the device must be considered to ensure stable operation on different hardware platforms, especially on portable and low-power devices.

[0056] Data privacy: During the visualization process, the patient's ECG data is fully protected in accordance with relevant laws, regulations and ethical standards to avoid leakage of sensitive information.

[0057] Practicality: Ensure that the method can be widely used in actual medical scenarios, be concise in design, easy to use, and suitable for real-time and efficient ECG signal analysis and display in clinical environments and smart health devices.

[0058] Compatibility: Considering the multi-lead nature of the ECG, ensure the compatibility of the method between different lead data, and be able to simultaneously process and display ECG signals from multiple leads. For example, in wearable device applications, the visualization of data will focus on a small number of leads (such as a single lead or three leads), and ensure the effective display of data in 3D trajectory diagrams and pseudo-color rendering.

[0059] Adjustability: Allows users to customize visualization parameters and analysis angles according to specific needs.

[0060] 3. Implementation of specific technical solutions.

[0061] This embodiment combines the method of pseudo color mapping and 3D trajectory map, referring to Figure 2 The specific technical solution of this embodiment includes the following steps: Step 1: Signal acquisition and preprocessing.

[0062] (1) Signal acquisition: Acquire 12-lead electrocardiogram (ECG) signal data over a period of time and select appropriate acquisition equipment to ensure the integrity and accuracy of the data, including correct placement of electrodes and equipment calibration to ensure the integrity and accuracy of the data.

[0063] Specifically, select high-quality acquisition equipment to obtain 12-lead electrocardiogram (ECG) signal data. The equipment used may include conventional ECG monitoring equipment or hospital-grade ECG recorders. It is necessary to ensure that the equipment is fully calibrated before acquisition to reduce system errors.

[0064] (2) Data preprocessing: De-noising, filtering and normalization are performed on the collected ECG signals to eliminate physiological and environmental noise, improve signal clarity and lay the foundation for subsequent feature extraction and visualization analysis.

[0065] Specifically, a high-pass filter can be used to remove low-frequency noise and baseline drift, and a low-pass filter can be used to remove high-frequency noise. Normalization refers to adjusting the signal to a uniform scale range, and the Z-Score normalization method can be used for normalization.

[0066] Step 2: Multi-parameter feature extraction.

[0067] (1) Frequency distribution: The time domain information is converted to the frequency domain through fast Fourier transform to obtain the frequency distribution (i.e., frequency characteristics) of the ECG signal (i.e., the first ECG signal).

[0068] Specifically, for the sampling point ECG signal value at , use Fast Fourier Transform (FFT) to analyze the frequency characteristics, the calculation formula is: ; in, represents the spectrum of the signal, The index representing the frequency, Indicates the signal length, usually the number of sampling points of ECG data. and , corresponding to different frequency components. Each lead corresponds to a time series, and FFT is performed separately to obtain the spectrum information of each lead.

[0069] (2) Multi-lead consistency: Calculate the synchronization and phase relationship between the ECG signals of different leads (i.e., the first ECG signal) and analyze the time correlation of different leads.

[0070] Specifically, a formula combining synchronization and phase difference is constructed, wherein synchronization provides a similarity measure in the time domain, and phase difference provides a difference measure in the frequency domain.

[0071] Synchronicity can be evaluated by calculating the cross-correlation function of the ECG signals of different leads, using the following formula: ; in, Represents the ECG signal of two leads and The cross-correlation function of and Represents the ECG signals of two different leads. Indicates lead The ECG signal value, offset in time The sampling point value after that indicates Signal relative to The signal is delayed by a time operation.

[0072] In frequency index The phase difference of ECG signals of different leads in the frequency domain is for: ; and Respectively represent leads and leads The frequency domain signal is indexed at frequency The phase of .

[0073] In order to balance the synchronization and phase difference, this embodiment sets a weight parameter , and obtain an overall measure (i.e., a comprehensive measure value).

[0074] ; in, Indicates lead and The comprehensive measure of represents the cross-correlation function of the two leads in the time domain, which measures the synchronization. represents the phase difference between the two leads in the frequency domain, The length of the lag time window representing the cross-correlation is usually the same as the signal length, represents the number of frequency components of Fourier transform, Represents the weight coefficient, which is used to adjust the impact of synchronization and phase difference on the final measurement.

[0075] (3) Dynamic trend: The first-order derivative and second-order derivative of the ECG signal (i.e., the first ECG signal) are calculated to capture the dynamic characteristics. The dynamic trend can reflect the overall change characteristics of the signal over time.

[0076] Specifically, the first-order derivative can capture the local rate of change of the ECG signal and is used to detect locations where the signal changes suddenly or rapidly. Usually, there will be obvious high-amplitude changes when the signal changes rapidly (such as at the peak of the R wave), which helps us identify locations where the ECG signal changes rapidly, such as the QRS complex.

[0077] For ECG signals , the calculation formula of its rate of change is: ; in, Indicates that at the sampling point The signal change rate at and Respectively represent the ECG signal values ​​at adjacent sampling points.

[0078] The second-order derivative can reflect the dynamic trend of the signal and help identify the overall acceleration or deceleration of the signal, such as the rise or fall of the ST segment. The specific calculation formula is: ; in, Indicates that at the sampling point The second-order derivative of the signal at is used to reflect the trend change and acceleration of the signal over time, helping to capture larger trend turning points.

[0079] (4) Signal enhancement: A multi-head attention mechanism is added to assign different weights to signals at different time points, highlighting important ECG signal segments (such as waveforms when the heart is abnormal). The extracted attention weights can be used as important features to reflect the key areas of the model's attention.

[0080] Specifically, a multi-head attention mechanism is used to assign different weights to signals at different time points to highlight important ECG signal segments.

[0081] For each lead, the ECG signal sequence within a period of time is divided into multiple time segments as the input of the model. This segmentation method is conducive to the multi-head attention mechanism to focus on different time segments. Assume that the sequence is divided into segments, each segment contains sampling points.

[0082] ; in, Indicates The segmented time segments are modeled using a multi-head attention mechanism. The goal of the multi-head attention mechanism is to learn the relationship between each time segment and other time segments.

[0083] a. Construction of query, key and value vectors.

[0084] For each time segment , we map it to a query Q, key K, and value V vector. This is achieved through a linear transformation: ; ; ; in, , and Represents the trainable weight matrices used to map inputs to query, key, and value vectors, respectively.

[0085] b. Calculation of attention weights.

[0086] For a given query vector and key vector , calculate its similarity through dot product and convert it into weight through Softmax function: ; in, Represents a time segment For time segments The attention weight, Represents the dimension of the key vector, used for scaling. The larger the weight, the more attention the model pays to the segment.

[0087] c. Weighted output Use the calculated attention weights Pair value vector Perform weighted summation to get the output of each fragment: ; in, Indicates The value vector of each time segment represents the information carried by the segment. The result of the weighted summation is the output of the time segment that the model considers important. This step focuses on important signal segments through the attention mechanism.

[0088] d.Multiple attention.

[0089] In order to enable the model to capture different temporal features from multiple angles, this embodiment can use a multi-head attention mechanism. Multi-head attention models different parts of the signal by parallelizing multiple attention heads. Each attention head has an independent query, key, and value matrix: ; in, Indicates The output of the attention head is: Represents a learnable weight matrix for linear transformation, summarizing the results of multiple heads, usually initialized with random values ​​and continuously optimized during training.

[0090] The attention output of each head is: ; in, , and Represent independent weight matrices for queries, keys, and values, respectively.

[0091] For each fragment , the focused time segment can be determined by the maximum attention weight: ; In the original ECG signal The attention weight is superimposed on the basis of , the specific calculation formula is: ; in, Indicates enhanced signal characteristics, Indicates the sampling time point The first ECG signal at Indicates the number of time slices, represents the Dirac function, which is used to determine The key time point is ,but ,otherwise . Indicates the time point corresponding to the maximum attention weight output by the multi-head attention mechanism, Indicates Time Segment The value of the maximum attention weight in , the corresponding time point is , Indicates The time point that the multi-head attention mechanism pays most attention to in the time segment The corresponding second ECG signal.

[0092] Step 3: Multi-step pseudo-color mapping.

[0093] (1) Frequency layer coloring: The calculated ECG signal frequency is mapped into a color gradient from green to purple from low frequency to high frequency and placed in the first layer.

[0094] Specifically, the frequency characteristics of the extracted ECG signal are mapped to a pseudo color (ie, the first pseudo color map). The pseudo color mapping formula used is: ; in, Indicates signal strength The mapped color, Represents a pseudo color mapping function, such as a heatmap or rainbow.

[0095] (2) Consistency layer coloring: Use transparency to indicate synchronization. High synchronization areas have lower transparency and are highlighted.

[0096] Specifically, the calculated synchronization and phase difference between different leads are combined to define the multi-lead consistency feature, and one lead is taken as the main object, and the consistency curve between the lead and other leads is drawn. The mapping formula used for coloring is also a pseudo color mapping function.

[0097] (3) Dynamic enhancement coloring: Combining dynamic trends and signal enhancement, it emphasizes areas with high signal change rates and reflects dynamic trends.

[0098] Specifically, the dynamic trend (i.e., dynamic feature) extracted from the feature is combined with the signal enhancement (i.e., enhanced signal feature) to define a new dynamic enhanced signal (i.e., combined feature). and dynamic trends With the enhanced signal Combined, the combined feature calculation formula is: ; in, and Represents the weight coefficient, which is used to balance the enhanced signal, rate of change, and dynamic trend. In this way, we can intuitively see which time segments are given higher attention through the post-colorization image.

[0099] Then, the curve is drawn and colored, and the mapping formula used for coloring is also a pseudo-color mapping function.

[0100] Step 4: Construct three-dimensional trajectory map.

[0101] (1) Multi-dimensional feature fusion: The pseudo-color information after multi-step coloring is integrated into the three-dimensional trajectory, so that the trajectory line can show multiple features at the same time.

[0102] Specifically, the three colored curve graphs in step 3 are merged, and the three curve graphs are merged by software in the prior art to construct a three-dimensional trajectory graph.

[0103] (2) User-defined settings: Allow users to customize visualization parameters and analysis angles according to specific needs, such as adjusting the color scale of pseudo-color mapping, selecting different leads or three-dimensional view angles, etc., to meet different analysis needs, and use evaluation indicators to detect the performance of the method.

[0104] Specifically, users are allowed to customize visualization parameters and analysis angles according to specific needs. Users can adjust the color scale of pseudo-color mapping, select different lead displays, adjust the angle of the three-dimensional view, etc. to meet different analysis needs.

[0105] In the above method, it can be evaluated based on three evaluation indicators: readability, feature separation and computational efficiency.

[0106] Readability can be indirectly reflected by the signal-to-noise ratio (SNR) of the graph. The SNR formula is: ; in, represents the power of the signal, Represents the power of the noise.

[0107] Feature Separation measures whether the visualization can effectively distinguish different ECG features (such as P wave, QRS wave and T wave) and abnormal waveforms. The distance or overlap between features can be calculated by automated algorithms. Specifically: ; in, Representation characteristics The average value of Representation characteristics The standard deviation of .

[0108] Computational Efficiency measures the time and resource consumption of generating visualization graphics. The evaluation indicators used include generation time (in seconds) and memory usage (in MB).

[0109] ; Here the output quality ( ) can be measured by the aforementioned readability and feature separation, Indicates resource consumption.

[0110] The above method combines pseudo color and 3D trajectory diagram to visualize and enhance various ECG signal features. Errors and missed detections are corrected through continuous optimization steps to improve detection accuracy.

[0111] The application scenarios of the technical solution of this embodiment include: The technical solution of this embodiment is suitable for advanced analysis and visualization of ECG signals, especially in scientific research experiments, data mining and engineering analysis. Through this method, researchers and technical developers can accurately extract features and display three-dimensionally complex ECG signals, and enhance the understanding and analysis capabilities of specific waveforms. This technology is not only suitable for laboratory research, but can also be integrated into intelligent health monitoring and signal processing systems as a core tool for data visualization and feature analysis. In addition, this method is also suitable for the development of machine learning models, helping to extract and analyze key features in large-scale ECG data, and providing new possibilities for the research and application of ECG signals.

[0112] The innovation of this embodiment is to combine pseudo-color coding with 3D trajectory diagram technology to fully display the frequency distribution, dynamic signal enhancement and inter-lead correlation information of ECG signals, making the waveform feature display more prominent. The use of dynamic color gradient and animation effects enhances the visual representation of ECG signal features, helping researchers and technicians to quickly identify key features in large amounts of data. The introduction of automatic detection and annotation algorithms provides strong support for in-depth research and technical applications of ECG signals.

[0113] Reference Figure 3 The embodiment of the present application also provides an electrocardiogram signal visualization enhancement system, which includes a data acquisition unit 100, a feature extraction unit 200, a feature mapping unit 300 and a trajectory diagram construction unit 400, wherein: The data acquisition unit 100 is used to acquire first electrocardiogram signals of multiple leads within a preset time period; The feature extraction unit 200 is used to extract frequency features, multi-lead consistency features, dynamic features and enhanced signal features of the first ECG signals of multiple leads, wherein the frequency feature is the frequency distribution of the first ECG signal of each lead in the frequency domain, the multi-lead consistency feature is the comprehensive measurement value between the synchronization of the first ECG signals of two different leads in the time domain and the phase difference of the first ECG signals of two different leads in the frequency domain, the dynamic feature is the first-order derivative and the second-order derivative of the first ECG signal of each lead, and the enhanced signal feature is the ECG signal after the first ECG signal of each lead is weighted by the attention mechanism; A feature mapping unit 300, configured to map the frequency feature into a first pseudo color map, map the multi-lead consistency feature into a second pseudo color map, and map the dynamic feature and the enhanced signal feature into a third pseudo color map; The trajectory map construction unit 400 is used to construct a three-dimensional trajectory map according to the first pseudo-color map, the second pseudo-color map and the third pseudo-color map.

[0114] It should be noted that, since the ECG signal visualization enhancement system in this embodiment and the ECG signal visualization enhancement method described above are based on the same inventive concept, the corresponding contents in the method embodiment are also applicable to the system embodiment and will not be described in detail here.

[0115] Reference Figure 4 The embodiment of the present application further provides an electronic device, the electronic device comprising: at least one memory; at least one processor; at least one program; The programs are stored in the memory, and the processor executes at least one program to implement the above-mentioned ECG signal visualization enhancement method implemented in the present disclosure.

[0116] The electronic device may be any intelligent terminal including a mobile phone, a tablet computer, a personal digital assistant (PDA), a car computer, etc.

[0117] The electronic device according to the embodiment of the present application is described in detail below.

[0118] The processor 1600 may be implemented by a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided by the embodiments of the present disclosure; The memory 1700 may be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1700 may store an operating system and other application programs. When the technical solution provided in the embodiments of this specification is implemented by software or firmware, the relevant program code is stored in the memory 1700, and the processor 1600 calls and executes the electrocardiogram signal visualization enhancement method of the embodiment of the present disclosure.

[0119] Input / output interface 1800, used to implement information input and output; Communication interface 1900, used to realize communication interaction between the device and other devices, which can be realized through wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.); Bus 2000 , which transmits information between various components of the device (e.g., processor 1600 , memory 1700 , input / output interface 1800 , and communication interface 1900 ); The processor 1600 , the memory 1700 , the input / output interface 1800 , and the communication interface 1900 are connected to each other in communication within the device via the bus 2000 .

[0120] The embodiment of the present disclosure further provides a storage medium, which is a computer-readable storage medium. The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the above-mentioned electrocardiogram signal visualization enhancement method.

[0121] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0122] The embodiments described in the embodiments of the present disclosure are intended to more clearly illustrate the technical solutions of the embodiments of the present disclosure and do not constitute a limitation on the technical solutions provided by the embodiments of the present disclosure. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of the present disclosure are also applicable to similar technical problems.

[0123] Those skilled in the art will appreciate that the technical solutions shown in the figures do not limit the embodiments of the present disclosure and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0124] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0125] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0126] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0127] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0128] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0129] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0130] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0131] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable an electronic device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store programs. The above is a detailed description of the embodiments of the present application in conjunction with the accompanying drawings, but the present application is not limited to the above embodiments. Within the scope of knowledge possessed by ordinary technicians in the relevant technical field, various changes can be made without departing from the purpose of the present application.

Claims

1. A method for visualizing and enhancing an electrocardiogram signal, characterized in that: The method comprises: Acquire first electrocardiogram signals of multiple leads within a preset time period; Extracting frequency features, multi-lead consistency features, dynamic features and enhanced signal features of the first ECG signals of the multiple leads, wherein the frequency feature is the frequency distribution of the first ECG signal of each lead in the frequency domain, the multi-lead consistency feature is a comprehensive measurement value between the synchronization of the first ECG signals of two different leads in the time domain and the phase difference of the first ECG signals of two different leads in the frequency domain, the dynamic feature is the first-order derivative and the second-order derivative of the first ECG signal of each lead, and the enhanced signal feature is the ECG signal after weighting the first ECG signal of each lead through the attention mechanism; Mapping the frequency feature into a first pseudo color map, mapping the multi-lead consistency feature into a second pseudo color map, and mapping the dynamic feature and the enhanced signal feature into a third pseudo color map; A three-dimensional trajectory map is constructed according to the first pseudo-color map, the second pseudo-color map, and the third pseudo-color map.

2. The method for visualizing and enhancing an electrocardiogram signal according to claim 1, characterized in that: The extracting of frequency features, multi-lead consistency features, dynamic features, and enhanced signal features of the first electrocardiogram signals of the multiple leads includes: Calculating the synchronization of the first electrocardiogram signals of two different leads in the time domain to obtain a first result; Calculating the phase difference between the first electrocardiogram signals of the two different leads in the frequency domain to obtain a second result; According to the first result and the second result, a comprehensive metric value between the first electrocardiogram signals of the two different leads is calculated, and the comprehensive metric value is used as a multi-lead consistency feature.

3. The method for visualizing and enhancing electrocardiogram signals according to claim 2, characterized in that: The step of calculating the comprehensive metric value between the first electrocardiogram signals of the two different leads according to the first result and the second result includes: ； in, Represents a comprehensive measurement value, represents the weight coefficient, represents the total length of the lag time window for the cross-correlation, represents the first result obtained by calculating the synchronization of the first ECG signals of two different leads in the time domain through the cross-correlation function, represents the lag time window length, represents the number of frequency components of Fourier transform, It represents the second result obtained by calculating the phase difference of the first ECG signals of two different leads in the frequency domain. Indicates frequency components.

4. The method for visualizing and enhancing electrocardiogram signals according to claim 1, characterized in that: The extracting of frequency features, multi-lead consistency features, dynamic features, and enhanced signal features of the first electrocardiogram signals of the multiple leads includes: Calculating the first-order derivative of the first electrocardiogram signal of each lead; Calculating the second-order derivative of the first electrocardiogram signal of each lead; The first-order derivative and the second-order derivative of the first electrocardiogram signal of each lead are used as dynamic features.

5. The method for visualizing and enhancing electrocardiogram signals according to claim 1, characterized in that: The extracting of frequency features, multi-lead consistency features, dynamic features, and enhanced signal features of the first electrocardiogram signals of the multiple leads includes: Segmenting the first electrocardiogram signal of each lead within the preset time period into second electrocardiogram signals of a plurality of time segments; The second ECG signals of the multiple time segments are input into a multi-head attention mechanism for attention weighting to obtain enhanced signal features corresponding to the first ECG signal of each lead.

6. The method for visualizing and enhancing electrocardiogram signals according to claim 5, characterized in that: The inputting the second ECG signals of the multiple time segments into the multi-head attention mechanism for attention weighting to obtain enhanced signal features corresponding to the first ECG signals of each lead includes: ； in, Indicates enhanced signal characteristics, Indicates the sampling time point The first ECG signal at Indicates the number of time slices, represents the Dirac function, Indicates the time point corresponding to the maximum attention weight output by the multi-head attention mechanism, Indicates Time Segment The value of the maximum attention weight in , Indicates The time point that the multi-head attention mechanism pays most attention to in the time segment The corresponding second ECG signal.

7. The method for visualizing and enhancing electrocardiogram signals according to claim 1, characterized in that: Mapping the dynamic feature and the enhanced signal feature into a third pseudo color map includes: Summing the first-order derivative of the first electrocardiogram signal of each lead, the second-order derivative of the first electrocardiogram signal of each lead, and the enhanced signal feature to obtain a combined feature; The combined features are mapped into a third pseudo-color map.

8. An electrocardiogram signal visualization enhancement system, characterized in that: The system comprises: A data acquisition unit, used to acquire first electrocardiogram signals of multiple leads within a preset time period; A feature extraction unit, used to extract frequency features, multi-lead consistency features, dynamic features and enhanced signal features of the first ECG signals of the multiple leads, wherein the frequency feature is the frequency distribution of the first ECG signal of each lead in the frequency domain, the multi-lead consistency feature is a comprehensive measure of the synchronization of the first ECG signals of two different leads in the time domain and the phase difference of the first ECG signals of two different leads in the frequency domain, the dynamic feature is the first-order derivative and the second-order derivative of the first ECG signal of each lead, and the enhanced signal feature is the ECG signal after the first ECG signal of each lead is weighted by the attention mechanism; A feature mapping unit, used for mapping the frequency feature into a first pseudo color map, mapping the multi-lead consistency feature into a second pseudo color map, and mapping the dynamic feature and the enhanced signal feature into a third pseudo color map; The trajectory map construction unit is used to construct a three-dimensional trajectory map according to the first pseudo-color map, the second pseudo-color map and the third pseudo-color map.

9. An electronic device, characterized in that: It includes at least one control processor and a memory for communicating with the at least one control processor; the memory stores instructions that can be executed by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to execute the electrocardiogram signal visualization enhancement method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the electrocardiogram signal visualization enhancement method according to any one of claims 1 to 7.

Images