A magnetocardiogram analysis method, device, equipment, and medium based on spatio-temporal information

Abstract

The present application relates to the technical field of biomedical signal analysis, and provides a magnetocardiogram analysis method and device, equipment, and medium based on spatio-temporal information. The method includes: obtaining an original magnetocardiogram data set, and drawing a time wave group diagram and a spatial wave group diagram according to the original magnetocardiogram data set; respectively segmenting time waveform data and spatial waveform data of at least one band in the time wave group diagram and the spatial wave group diagram; determining a plurality of time waveform features according to the time waveform data, and determining a plurality of spatial waveform features according to the spatial waveform data; calculating the waveform deviation degree between the original magnetocardiogram and a preset reference diagram according to the plurality of time waveform features and / or the plurality of spatial waveform features, and obtaining a magnetocardiogram analysis result according to the waveform deviation degree. The magnetocardiogram analysis method of the present application can respectively extract time features and spatial features according to the time wave group diagram and the spatial wave group diagram, comprehensively analyze time information and spatial information, and obtain a more accurate magnetocardiogram analysis result.

Description

Technical Field

[0001] The present application relates to the technical field of biomedical signal analysis, and in particular, to a magnetocardiogram analysis method, device, equipment, and medium based on spatio-temporal information. Background Technique

[0002] Magnetocardiogram is a non-invasive cardiac function detection technology that reflects the electrophysiological activities of the heart by measuring the magnetic fields generated by cardiac activities. A magnetocardiograph has dozens of channels, equivalent to dozens of chest leads, which can record more changes in the depolarization and repolarization intensities, rates, waveforms, etc. of the heart, and can more reflect the depolarization and repolarization change characteristics of each part of the cardiac myocardium.

[0003] Regarding these changes in intensity, rate, waveform, etc., most of the existing magnetocardiogram analysis methods are based on the one-dimensional butterfly diagram of magnetocardiogram data. The horizontal axis of the one-dimensional butterfly diagram is time, and the vertical axis is magnetic field intensity. Usually, the changes in time and amplitude are extracted from the one-dimensional butterfly diagram of the magnetocardiogram data set, such as the QT interval, QRS interval, magnetic field intensity ratio of R wave to T wave, etc. Although this method can identify some information contained in the magnetocardiogram, it has certain limitations and cannot fully exploit the spatio-temporal information of the magnetocardiogram in the time and space dimensions. Therefore, the analysis ability is limited, affecting the accuracy of the analysis results. Summary of the Invention

[0004] In view of this, the purpose of the present application is to provide a magnetocardiogram analysis method, device, equipment, and medium based on spatio-temporal information, which respectively extracts time features and space features according to the time wave group diagram and the space wave group diagram, comprehensively analyzes the time information and space information, and obtains more accurate magnetocardiogram analysis results.

[0005] The present application mainly includes the following aspects:

[0006] In the first aspect, an embodiment of the present application provides a magnetocardiogram analysis method based on spatio-temporal information, and the method includes:

[0007] Obtain the original magnetocardiogram data set, and draw a time wave group diagram and a space wave group diagram according to the original magnetocardiogram data set;

[0008] Segment at least one band of time waveform data in the time wave group diagram, and segment at least one band of space waveform data in the space wave group diagram;

[0009] Determine a plurality of time waveform features according to the time waveform data, and determine a plurality of space waveform features according to the space waveform data;

[0010] Calculate the waveform deviation degree between the original magnetocardiogram and the reference diagram according to the multiple time waveform features and / or the multiple spatial waveform features, and obtain the analysis result of the original magnetocardiogram according to the waveform deviation degree.

[0011] In a second aspect, an embodiment of the present application further provides a magnetocardiogram analysis device based on spatio-temporal information. The magnetocardiogram analysis device based on spatio-temporal information includes:

[0012] A wave group diagram drawing module, configured to obtain an original magnetocardiogram data set, and draw a time wave group diagram and a spatial wave group diagram according to the original magnetocardiogram data set;

[0013] A wave band segmentation module, configured to segment time waveform data of at least one wave band in the time wave group diagram, and segment spatial waveform data of at least one wave band in the spatial wave group diagram;

[0014] A feature extraction module, configured to determine multiple time waveform features according to the time waveform data, and determine multiple spatial waveform features according to the spatial waveform data;

[0015] A magnetocardiogram analysis module, configured to calculate the waveform deviation degree between the original magnetocardiogram and the reference diagram according to the multiple time waveform features and / or the multiple spatial waveform features, and obtain the analysis result of the original magnetocardiogram according to the waveform deviation degree.

[0016] In a third aspect, an embodiment of the present application further provides an electronic device, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device runs, the processor communicates with the memory through the bus. When the machine-readable instructions are executed by the processor, the steps of the magnetocardiogram analysis method based on spatio-temporal information described above are executed.

[0017] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is run by a processor, the steps of the magnetocardiogram analysis method based on spatio-temporal information described above are executed.

[0018] The magnetocardiogram analysis method, device, equipment, and medium based on spatio-temporal information provided by the embodiments of the present application use the collected original magnetocardiogram data to draw the time wave curves and spatial wave curves of each channel, and obtain the time wave group diagram and the spatial wave group diagram; then, in the time wave group diagram and the spatial wave group diagram of the entire heartbeat cycle, several representative wave bands are segmented, and time waveform features and spatial waveform features are respectively extracted from the time data and spatial data of these wave bands; the extracted time waveform features and spatial waveform features reflect the spatio-temporal information contained in the time wave curves and spatial wave curves. By comprehensively analyzing these features, the final magnetocardiogram analysis result can be obtained.

[0019] Compared with the magnetocardiogram analysis methods in the prior art, the present application extracts corresponding waveform feature parameters based on time wave group diagrams and spatial wave group diagrams, which can effectively characterize the cardiac electrophysiological changes behind different wave bands and spatial positions, realizes the accurate extraction of the time and spatial information of magnetocardiogram signals, can more comprehensively characterize the information contained in the magnetocardiogram, and thus improves the magnetocardiogram analysis ability and the accuracy of the analysis results.

[0020] To make the above objects, features, and advantages of the present application more obvious and understandable, the following specifically gives preferred embodiments and detailed descriptions in conjunction with the accompanying drawings. Description of the Drawings

[0021] To more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation of the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0022] Figure 1 Shows a flowchart of a magnetocardiogram analysis method based on spatio-temporal information provided by an embodiment of the present application;

[0023] Figure 2 Shows a schematic diagram of amplitude ratio calculation provided by an embodiment of the present application;

[0024] Figure 3 Shows a schematic diagram of wave peak length ratio calculation provided by an embodiment of the present application;

[0025] Figure 4 Shows a schematic diagram of positive wave shape fraction calculation provided by an embodiment of the present application;

[0026] Figure 5 Shows a schematic diagram of negative wave shape fraction calculation provided by an embodiment of the present application;

[0027] Figure 6 Shows a schematic diagram of the number of channels with opposite signs of wave peaks calculation provided by an embodiment of the present application;

[0028] Figure 7 Shows a schematic diagram of wave peak zero line diagonal calculation provided by an embodiment of the present application;

[0029] Figure 8 Shows a schematic diagram of wave peak zero line rotation angle calculation provided by an embodiment of the present application;

[0030] Figure 9The flowchart of another magnetocardiogram analysis method based on spatio-temporal information provided by the embodiments of the present application is shown;

[0031] Figure 10 The structural schematic diagram of a magnetocardiogram analysis device based on spatio-temporal information provided by the embodiments of the present application is shown;

[0032] Figure 11 The structural schematic diagram of an electronic device provided by the embodiments of the present application is shown.

[0033] In the figure:

[0034] 1000 Magnetocardiogram analysis device based on spatio-temporal information, 1010 Wave group diagram drawing module, 1020 Band segmentation module, 1030 Feature extraction module, 1040 Magnetocardiogram analysis module, 1100 Electronic device, 1110 Processor, 1120 Memory, 1130 Bus. Specific embodiments

[0035] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Apparently, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and illustrated herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the present application claimed, but merely represents selected embodiments of the present application. Based on the embodiments of the present application, every other embodiment obtained by those skilled in the art without creative efforts belongs to the scope of protection of the present application.

[0036] First, the applicable application scenarios of the present application are introduced. The present application belongs to the field of image analysis. By analyzing magnetocardiogram images, the electrical activity of the heart can be evaluated and can be applied to the fields of medicine and bioengineering, etc.

[0037] It has been found through research that most of the existing magnetocardiogram analysis methods are based on the one-dimensional butterfly diagram of magnetocardiogram data. The horizontal axis of the one-dimensional butterfly diagram is time, and the vertical axis is magnetic field strength. Usually, the changes in time and amplitude are extracted from the one-dimensional butterfly diagram of the magnetocardiogram data set. Through the changes in time and amplitude, that is, the changes in magnetic field strength, the corresponding analysis results are obtained. This method cannot fully exploit the spatio-temporal information of the magnetocardiogram in the time and space dimensions. Therefore, the analysis ability is limited, affecting the accuracy of the analysis results.

[0038] Based on this, the embodiments of the present application provide a magnetocardiogram analysis method based on spatio-temporal information to fully exploit the spatio-temporal information of the magnetocardiogram in the time and space dimensions and improve the accuracy of the magnetocardiogram analysis results.

[0039] Please refer to Figure 1 , Figure 1 which is a flowchart of a magnetocardiogram analysis method based on spatio-temporal information provided by an embodiment of the present application. As Figure 1 shown in

[0040] S101. Obtain the original magnetocardiogram data set, and draw a time wave group diagram and a space wave group diagram according to the original magnetocardiogram data set.

[0041] In this step, read the original magnetocardiogram data set collected by a multi-channel magnetocardiograph, and draw a time wave curve of each channel during the entire cardiac cycle according to the data collected by each channel. The time wave curves of each channel are combined to obtain a time wave group diagram; similarly, a space wave curve of each channel during the entire cardiac cycle can be drawn, and the space wave curves of each channel are combined to obtain a space wave group diagram.

[0042] S102. Segment time waveform data of at least one band in the time wave group diagram, and segment space waveform data of at least one band in the space wave group diagram.

[0043] In this step, since the time wave group diagram obtained in the previous step is the waveform of the entire cardiac cycle, several representative bands can be segmented from it based on the characteristics of each band in the heartbeat cycle, and then the time waveform data and space waveform data are analyzed. For example, the QRS band and the T band can be segmented in the time wave group diagram to obtain the time waveform data of the QRS band and the T band; similarly, the QRS band and the T band are segmented in the space wave group diagram to obtain the space waveform data of the QRS band and the T band.

[0044] S103. Determine a plurality of time waveform features according to the time waveform data, and determine a plurality of space waveform features according to the space waveform data.

[0045] In this step, a plurality of representative time waveform features and space waveform features are extracted from the time waveform data and the space waveform data respectively. Among them, the time waveform features can reflect the characteristics of the time waveform in terms of amplitude, curve shape, etc., and the space waveform features can reflect the characteristics of the space waveform in terms of peak orientation and peak position, etc.

[0046] S104. Calculate the waveform deviation degree between the original magnetocardiogram and the reference diagram according to a plurality of time waveform features and / or a plurality of space waveform features, and obtain the analysis result of the original magnetocardiogram according to the waveform deviation degree.

[0047] In this step, the waveform deviation degree between the original magnetocardiogram and a preset reference diagram is calculated. If the deviation degree is low, it can be considered that the original magnetocardiogram is relatively similar to the reference diagram. Therefore, the analysis result of the reference diagram can be used to analyze the original magnetocardiogram to obtain the analysis result of the original magnetocardiogram. Herein, the reference diagram is a magnetocardiogram with an analysis result set in advance. For example, the magnetocardiogram of a coronary heart disease patient is used as the reference diagram. If the waveform deviation degree between the original magnetocardiogram and this reference diagram is low, it can be considered that the cardiac electrical activity possibly reflected by the original magnetocardiogram may be abnormal, and there may be a possibility of coronary heart disease.

[0048] The magnetocardiogram analysis method based on spatio-temporal information provided by the embodiments of the present application uses the collected original magnetocardiogram data to draw the time wave curves and space wave curves of each channel, and obtains a time wave group diagram and a space wave group diagram; then, in the time wave group diagram and space wave group diagram of the entire heartbeat cycle, several representative bands are segmented, and time waveform features and space waveform features are respectively extracted from the time data and space data of these bands; the extracted time waveform features and space waveform features reflect the spatio-temporal information contained in the time wave curves and space wave curves. By comprehensively analyzing these features, the final magnetocardiogram analysis result can be obtained.

[0049] Compared with the magnetocardiogram analysis methods in the prior art, the present application extracts corresponding waveform feature parameters based on the time wave group diagram and space wave group diagram, can effectively depict the cardiac electrophysiological changes behind different bands and spatial positions, realizes the accurate extraction of the time and space information of the magnetocardiogram signal, can more comprehensively depict the information contained in the magnetocardiogram, and therefore improves the magnetocardiogram analysis ability and the accuracy of the analysis result.

[0050] Further, the band includes at least one of the following: QRS band, ST band, and T band, wherein the QRS band includes the R band.

[0051] Specifically, a complete heartbeat cycle includes multiple bands, and each band represents different heartbeat activities. For example, the P wave represents the start of atrial electrical activity, the TP band represents the cardiac electrical activity interval, and different bands have different spatio-temporal numerical distribution characteristics. Since the spatio-temporal information of the QRS band and T band is relatively complete and significant, in this embodiment, the time data and space data of the QRS band, ST band, and T band are segmented within the time wave group diagram and space wave group diagram of the entire heartbeat cycle. Among them, the QRS band further includes the Q band, R band, and S band. In this embodiment, the time data and space data of the more representative R band are also separately extracted from the QRS band.

[0052] In addition, it can be understood that other bands also have spatio-temporal information. Therefore, the embodiments of the present application can also segment the time data and space data of other bands for subsequent calculations. For example, the Q band, R band, and S band can also be segmented, and the corresponding time waveform features and space waveform features can be extracted therefrom. During subsequent operations, the structure and parameters of the formula are adjusted accordingly according to different bands.

[0053] Furthermore, the time waveform features are extracted from the time wave, specifically determined according to the amplitude, curve shape, and alignment degree of each channel of the time wave of each band. The time waveform features may specifically include the following features:

[0054] Amplitude ratio, determined according to the maximum value of the average signal of all channels of each band. Specifically, since magnetocardiograms collect data from multiple channels, there are waveform diagrams of multiple channels for each band. Specifically, the time wave group diagram of each band contains multiple curves, and each curve corresponds to one channel. In this embodiment, for the R band and T band respectively, the curve of the average signal of all channels can be determined according to the waveform curves of each channel, and the maximum value of the average signal, that is, the peak of the curve, can be found in this curve. Then, the ratio between these two peaks is used as the amplitude ratio qrsttrt of the QRS band and T band.

[0055] Please refer to Figure 2 , Figure 2 which is a schematic diagram for calculating the amplitude ratio provided by the embodiments of the present application. As Figure 2 shown, the lower part is the time wave group diagram of 36 channels, and the upper part is the upper and lower bounds of the time wave group diagram and the average signal wave. The average signal is the root mean square of the 36-channel values. In Figure 2 , the maximum value of the average signal of the R band is at time 214, and the maximum value of the average signal of the T band is at time 411. Therefore, the amplitude ratio is the ratio of the R wave average signal value at time 241 to the T wave average signal value at time 411.

[0056] In addition, for the amplitude ratio, the maximum value of the average signal is used for calculation in this embodiment, and the maximum or minimum channel of each band can also be used for calculation. For example, for the R band and T band, the curves with the largest peaks in all their channels are determined respectively, and the ratio of these two peaks is used as the amplitude ratio qrsttrt of the QRS band and T band.

[0057] Similarly, during the calculation of other features, if the maximum value of the average signal is used, it can also be replaced by the maximum or minimum channel of each band for calculation, which is similar to the replacement method of the amplitude ratio and will not be elaborated here.

[0058] The peak length ratio is determined based on the maximum channel of the peak in the waveband and the duration of the waveband. In this embodiment, in the waveforms of each channel in the T waveband, the channel with the maximum peak value can be determined, and the ratio between its amplitude and the duration of the T waveband is used as the peak length ratio ttflt of the T waveband.

[0059] Please refer to Figure 3 , Figure 3 which is a schematic diagram for calculating the peak length ratio provided by the embodiment of the present application. As Figure 3 shown, the following is a time wave group diagram of 36 channels. It can be seen from the figure that the period from time 365 to 446 corresponds to the T waveband, and its duration is 81; the channel with the maximum peak in the T waveband can be found in the figure, and then the ratio between its amplitude and 81 is calculated to obtain the peak length ratio.

[0060] In addition, for the peak length ratio, in this embodiment, the ratio is calculated using the amplitude of the maximum channel, and the average signal or the minimum channel can also be used for calculation. For example, in the T waveband, the curve of the average signal is determined according to the waveforms of each channel, and the maximum value of the average signal, that is, the amplitude of the average signal, is found in this curve. The ratio between it and the duration of the T waveband is used as the peak length ratio ttflt of the T waveband.

[0061] Similarly, in the calculation process of other features, if the maximum channel is used, it can also be replaced by the average signal or the minimum channel for calculation, which is similar to the replacement method of the peak length ratio and will not be elaborated here.

[0062] The positive wave shape fraction is determined based on the curve of the maximum channel of the peak in the waveband. In this embodiment, in the waveforms of each channel in the T waveband, the channel with the maximum peak value can be determined, and the positive wave shape fraction ttstpp of the T waveband is calculated according to the area under the waveform curve of this channel and the area of the triangle formed by the peak of this channel.

[0063] Please refer to Figure 4 , Figure 4 which is a schematic diagram for calculating the positive wave shape fraction provided by the embodiment of the present application. As Figure 4 shown, the following is a time wave group diagram of 36 channels. It can be seen from the figure that the part from time 365 to 446 corresponds to the T waveband. The channel with the maximum peak in the T waveband can be found in the figure. The area of the region enclosed by the waveform curve of this channel, the horizontal axis (i.e., the baseline), and the two vertical lines at 365 and 446 is S1, as shown in the purple vertical stripe part in Figure 4 ; taking the peak as one vertex and the points at positions 365 and 446 on the horizontal axis as the other two vertices, the area of the formed triangle region is S2, as shown in the red diagonal stripe part in Figure 4 ; the ratio of S1 to S2 is the positive wave shape fraction of the T waveband.

[0064] The negative wave shape fraction is determined according to the minimum peak channel curve in the wave band. In this embodiment, similar to the aforementioned positive wave shape fraction, in the waveforms of each channel in the T band, the channel with the minimum peak value can be determined, and based on the area under the waveform curve of this channel and the area of the triangle formed by the peak value of this channel, the negative wave shape fraction ttstpn of the T band is calculated.

[0065] Please refer to Figure 5 , Figure 5 which is a schematic diagram for calculating the negative wave shape fraction provided by the embodiment of the present application. As Figure 5 shown, below is a time wave group diagram of 36 channels. It can be seen from the figure that the part corresponding to the time from 365 to 446 is the T band. The channel with the minimum peak in the T band can be found in the figure. The area of the region enclosed by the waveform curve of this channel, the horizontal axis (i.e., the baseline), and the two vertical lines at the 365 and 446 parts is S1, as shown in detail in Figure 5 the purple vertical stripe part; taking the peak value as one vertex and the points at the 365 and 446 positions on the horizontal axis as the other two vertices, the area of the formed triangle region is S2, as shown in detail in Figure 5 the red diagonal stripe part; the ratio of S1 to S2 is the negative wave shape fraction of the T band.

[0066] The elevation fraction is determined according to the amplitudes of each channel in the wave band. In this embodiment, in the ST band, the absolute value of the average amplitude of each channel can be determined, and the maximum value among these absolute values is used as the elevation fraction strs of the ST band.

[0067] In addition, for the elevation fraction, in this embodiment, the average amplitude of each channel is used to calculate the average amplitude value, and the average signal can also be used for calculation. For example, in the ST band, the absolute value of the average signal of each channel is determined, and the maximum value among them is used as the elevation fraction strs of the ST band.

[0068] As Figure 2 shown, it can be seen from the figure that the part corresponding to the time from 251 to 350 is the ST band. Denote the average amplitude values of the 36 channels in this band as fi (i = 1,..., 36), then strs = max(|fi|).

[0069] The wave smoothness is determined according to the second-order difference values and extreme difference values of each channel in the wave band. Specifically, the wave smoothness reflects the volatility of the wave band. In this embodiment, the second-order difference values of each channel in the ST band can be determined respectively, and the maximum second-order difference value among them can be determined; the maximum amplitude and the minimum amplitude of each channel in the ST band are determined respectively, and the difference between the two is used to obtain the amplitude extreme difference of each channel, and the maximum extreme difference value among them is determined; the ratio between the maximum second-order difference value and the maximum extreme difference value is used as the ST wave smoothness tsm.

[0070] Specifically, in the ST band of the time wave group diagram, the mean value of the absolute value of the second-order difference of each channel is denoted as dfi (i = 1, …, 36), and the maximum value among the 36 dfi is denoted as maxdfi; the maximum value of the amplitude range (the maximum amplitude minus the minimum amplitude) of each channel is denoted as dif; then the ST wave smoothness tsm is the ratio of maxdfi to dif.

[0071] The peak alignment is determined according to the positions of the maximum amplitudes of each channel in the band. Specifically, the peak alignment reflects the alignment of the peaks in 36 channels. In this embodiment, the maximum positive amplitude and the maximum negative amplitude with the largest absolute value among all channels in the T band can be determined first, and then several channels are selected from the 36 channels according to these two amplitudes for the calculation of the peak alignment. In the specific calculation process, the peak positions of each selected channel are determined respectively, and then the range is calculated for these peak positions as the peak alignment tmdf of the T band.

[0072] Specifically, in the T band, that is Figure 2 in the time part from 365 to 446, among the maximum amplitudes of 36 channels, the maximum positive amplitude maxf and the maximum negative amplitude minf with the largest absolute value are determined respectively, and one-third of maxf and one-third of minf are used as thresholds to select channels; the maximum amplitude of each of the 36 channels is denoted as fi (i = 1, …, 36), and based on the threshold-selected channels, the positive amplitudes reaching one-third of maxf and the negative amplitudes reaching one-third of minf are selected from all the maximum amplitudes fi; the corresponding time ti of each selected maximum amplitude fi is determined respectively, and finally the range is obtained by subtracting the minimum ti value from the maximum ti value, which is also the peak alignment tmdf of the T band.

[0073] Among them, in this embodiment, one-third of maxf and one-third of minf are selected as the thresholds to select channels, and other values can also be used, such as one-fourth of maxf and one-fourth of minf, or a specific value.

[0074] Furthermore, the spatial waveform features are extracted from the spatial wave, and are specifically determined according to the peak orientation, peak position and curve shape of each band. The spatial waveform features may specifically include the following features:

[0075] The number of channels with opposite-sign peaks is determined according to the peak orientation and position in different bands. Specifically, the number of channels with opposite-sign peaks reflects the number of channels with opposite-sign peak positions, that is, opposite orientations, in different bands. In this embodiment, such as Figure 6As shown, the upward and downward directions of the R-wave peak and the T-wave peak in each channel are respectively counted, and there are two possibilities for the direction, which are upward and downward, and two possibilities for the position, which are positive and negative. The number of channels where the directions and positions of the R-wave peak and the T-wave peak are opposite is counted as the number rtsgnum of channels with opposite R-wave and T-wave peaks.

[0076] The diagonal property of the peak zero line is determined according to the peak zero line of the wave band, where the peak zero line is determined according to the amplitudes of each channel in a preset number of frames before and after the peak of the wave band. Specifically, the maximum amplitudes of each channel in several frames (for example, 5 frames) before and after the peak are counted to form a new matrix. Then, in this matrix, according to the positive and negative of the peak, a broken line separating positive and negative values is drawn as the peak zero line of this wave band. Finally, the distances between each endpoint of the peak zero line and the diagonal endpoints are calculated respectively, and the sum of their squares is used as the diagonal property of the peak zero line.

[0077] Furthermore, the diagonal property of the peak zero line includes the diagonal property zeroR of the R-wave peak zero line and the diagonal property zeroT of the T-wave peak zero line. Their calculation methods are the same, and only the peaks of the corresponding wave bands need to be taken for calculation.

[0078] Please refer to Figure 7 , Figure 7 which is a schematic diagram for calculating the diagonal property of the peak zero line provided by the embodiment of the present application. As Figure 7 shown, taking the R wave band as an example, the figure contains a matrix with 6 rows and 6 columns, which contains the maximum amplitudes of 36 channels in the R wave band. According to the positive and negative of the peak, a broken line-shaped peak zero line is drawn, and the matrix is divided into two parts by the peak zero line. A diagonal line from the upper left corner to the lower right corner is drawn in this matrix. For the upper left endpoint (1, 6) of the peak zero line, calculate its distance from the upper left endpoint (0, 6) of the diagonal line; for the lower right endpoint (4, 0) of the peak zero line, calculate its distance from the lower right endpoint (6, 0) of the diagonal line, and then calculate the sum of the squares of the two distances to obtain the diagonal line zeroR of the R-wave peak zero line = 1×1 + 2×2 = 5.

[0079] The diagonal property of the area zero line is determined according to the area under the curve of each channel in the wave band. Specifically, similar to the diagonal line of the peak zero line, the areas enclosed by the curves and the baseline of each channel in this wave band are counted to form a new matrix. In this matrix, the area below the baseline is negative, and the area above the baseline is positive. A broken line separating positive and negative values can be drawn according to the positive and negative of the area as the area zero line of this wave band. Finally, the distances between each endpoint of the area zero line and the diagonal endpoints are calculated respectively, and the sum of their squares is used as the diagonal property of the area zero line.

[0080] For example, for the QRS band, the area under the curve of each channel of the QRS band can be statistically calculated to form a new matrix. A broken line separating positive and negative values is plotted, and a diagonal line from the upper left corner to the lower right corner is plotted in this matrix. Then, the distances between the two endpoints of this broken line and the endpoints of the diagonal line are calculated, and the sum of the squares of the distances is used as the QRS area zero-line diagonal property zeroqrs.

[0081] The wave peak zero-line rotation angle is determined according to the angle difference between the wave peak zero-lines of two bands. Specifically, for the aforementioned R wave peak zero-line and T wave peak zero-line, connect the two endpoints of the R wave peak zero-line, connect the two endpoints of the T wave peak zero-line, and use the angle difference between the two lines as the R wave peak - T wave peak zero-line rotation angle zeroRTrot.

[0082] Please refer to Figure 8 , Figure 8 which is a schematic diagram for calculating the wave peak zero-line rotation angle provided by an embodiment of this application. As Figure 8 shown, taking the R band as an example, the broken line in the figure is the R wave peak zero-line. Connect the two endpoints of the broken line to obtain the inclined line segment in the figure. In the same way, the two endpoints of the T wave peak zero-line can be connected to also obtain a line segment. Calculate the angle difference between these two line segments to obtain zeroRTrot.

[0083] Furthermore, the waveform deviation degree includes the time waveform deviation degree. In step S104, according to multiple time waveform features and / or multiple space waveform features, calculate the waveform deviation degree between the original magnetocardiogram and the preset reference diagram, including the following steps:

[0084] Step a1, calculate the superimposed waveform score between the original magnetocardiogram and the reference diagram according to multiple time waveform features, and determine the superimposed waveform score as the time waveform deviation degree;

[0085] The calculation formula for the superimposed waveform score is:

[0086] ovws = a11×qrsttrt + max(a21×strs, a22×tsm) + max([a31×ttflt, a32×ttstpp, a33×ttstpn, a34×tmdf]);

[0087] where ovws is the superimposed waveform score; a11, a21, a22, a31, a32, a33, a34 are preset coefficients; qrsttrt is the QRS / T average signal amplitude ratio; strs is the ST elevation score; tsm is the ST wave smoothness; ttflt is the wave peak length ratio of the T band; ttstpp is the positive wave shape score of the T band; ttstpn is the negative wave shape score of the T band; tmdf is the T wave peak alignment.

[0088] In this step, according to the time waveform features obtained in the previous step, the superposition waveform score of the original magnetocardiogram and the preset reference figure is calculated, and further the time waveform deviation degree between the original magnetocardiogram and the preset reference figure is obtained.

[0089] Among them, the first item of the calculation formula of the superposition waveform score reflects the time waveform deviation degree between the R wave band and the T wave; the second item reflects the time waveform deviation degree of the ST wave; the third item reflects the time waveform deviation degree of the T wave; the preset coefficients a11, a21, a22, a31, a32, a33, a34 are weight coefficients and can be set according to experience.

[0090] Furthermore, the waveform deviation degree includes the spatial waveform deviation degree. In step S104, according to multiple time waveform features and / or multiple spatial waveform features, the waveform deviation degree between the original magnetocardiogram and the preset reference figure is calculated, including the following steps:

[0091] Step c1, calculating the channel waveform score of the original magnetocardiogram and the reference figure according to the multiple spatial waveform features, and determining the channel waveform score as the spatial waveform deviation degree;

[0092] The calculation formula of the channel waveform score is:

[0093]

[0094] Among them, cnws is the channel waveform score; b11, b12, b21, b22, b31, R0, R1, T0 are preset coefficients; zeroR is the diagonal property of the R wave peak zero line; zeroqrs is the diagonal property of the QRS area zero line; rtsgnum is the number of channels with different signs between the R wave peak and the T wave peak; zeroRTrot is the rotation angle of the R wave peak and the T wave peak zero line; zeroT is the diagonal property of the T wave peak zero line.

[0095] In this step, according to the spatial waveform features obtained in the previous step, the channel waveform score of the original magnetocardiogram and the preset reference figure is calculated, and further the spatial waveform deviation degree between the original magnetocardiogram and the preset reference figure is obtained.

[0096] Among them, the first item of the calculation formula of the channel waveform score reflects the spatial waveform deviation degree of the R wave; the second item reflects the spatial waveform deviation degree between the R wave and the T wave; the third item reflects the spatial waveform deviation degree of the T wave; the preset coefficients b11, b12, b21, b22, b31 are weight parameters and can be set according to experience; the preset coefficients R0, R1, T0 are basic coefficients and are determined according to the parameter value distribution of the data set.

[0097] Further, the waveform deviation degree includes the spatial waveform deviation degree. In step S104, according to multiple time waveform features and / or multiple spatial waveform features, calculate the waveform deviation degree between the original magnetocardiogram and the preset reference diagram, including the following steps:

[0098] Step d1, calculate the superimposed waveform score between the original magnetocardiogram and the reference diagram according to multiple time waveform features.

[0099] Among them, the description of step d1 can refer to the description of step a1 and can achieve the same technical effect, so it will not be elaborated here.

[0100] Step d2, calculate the channel waveform score between the original magnetocardiogram and the reference diagram according to multiple spatial waveform features.

[0101] Among them, the description of step d2 can refer to the description of step c1 and can achieve the same technical effect, so it will not be elaborated here.

[0102] Step d3, calculate the comprehensive waveform score between the original magnetocardiogram and the reference diagram according to the superimposed waveform score and the channel waveform score, and determine the comprehensive waveform score as the comprehensive waveform deviation degree;

[0103] The calculation formula of the comprehensive waveform score is:

[0104] cpws = b1 × ovws + b2 × cnws;

[0105] Among them, cpws is the comprehensive waveform score; b1 and b2 are preset coefficients; ovws is the superimposed waveform score; cnws is the channel waveform score.

[0106] In this step, comprehensively analyze the time waveform deviation degree obtained by calculating the superimposed waveform score and the spatial waveform deviation degree obtained by calculating the channel waveform score in the foregoing steps, and calculate the comprehensive waveform score to obtain the final comprehensive waveform deviation degree. The higher the deviation degree, the higher the similarity between the original magnetocardiogram and the reference image; conversely, it is considered that the similarity between the original magnetocardiogram and the reference image is lower. With such a design, the heart characteristics reflected by the reference image can be used to analyze the original magnetocardiogram. Specifically, the weighted summation method can be used for calculation. In the calculation formula of the comprehensive waveform score, the preset coefficients b1 and b2 are weight coefficients and can be determined by the linear regression method according to the pre-constructed reference image set.

[0107] In addition, if only some time waveform features and spatial waveform features are extracted in the foregoing steps, the calculation formulas of the time waveform deviation degree, the spatial waveform deviation degree, and the final waveform deviation degree can be adjusted accordingly. For example, if only time waveform features are extracted and spatial waveform features are not extracted, b1 can be adjusted to 1 and b2 can be adjusted to 0.

[0108] Please refer to Figure 9 , Figure 9 which shows a schematic flowchart of a magnetocardiogram analysis method based on moment information according to another embodiment of the present application. As Figure 9 shown, the method includes the following steps:

[0109] S901. Based on the magnetocardiogram data set collected by a multi-channel magnetocardiograph, draw a time wave group diagram and a space wave group diagram for the entire cycle;

[0110] S902. Based on the time wave group diagram and the space wave group diagram, obtain the waveform data of the QRS band and the T band;

[0111] S903. Based on the waveform data of the QRS band and the T band, extract a series of time feature parameters;

[0112] S904. Based on the waveform data of the QRS band and the T band, extract a series of space feature parameters;

[0113] S905. According to a series of time feature parameters and space feature parameters, judge the deviation degree of the comprehensive waveform score between the magnetocardiogram image and the reference image set.

[0114] In this embodiment, first, use the magnetocardiogram data set collected by a multi-channel magnetocardiograph to draw a time wave group diagram and a space wave group diagram for the entire heartbeat cycle. Among them, the time wave group diagram includes the time waveform curves of each channel, and the space wave group diagram includes the space waveform curves of each channel; then, in the time wave group diagram and the space wave group diagram of the entire heartbeat cycle, divide the waveform data of the QRS band and the T band respectively, and according to the waveform data of these two segments, extract several representative time waveform features (i.e., time feature parameters) and space waveform features (i.e., space feature parameters) respectively; finally, use the extracted time waveform features and space waveform features to analyze the waveform deviation degree between the magnetocardiogram and the reference diagram from two directions of time and space, and obtain the final waveform deviation degree judgment result, that is, the comprehensive waveform score deviation degree.

[0115] In addition, before step 1, first construct a data set of 1000 samples and add labels to it according to the analysis results of the reference images. Respectively obtain the magnetocardiogram data sets of each sample, draw a time wave group diagram and a space wave group diagram, and obtain the waveform data of the QRS band and the T band; then use the analysis method in the foregoing embodiment to extract time waveform features and space waveform features, and delimit the value range of the weight parameters when calculating each deviation degree according to the distribution of the time waveform features and space waveform features of all samples. Thereafter, the waveform deviation degree between the magnetocardiogram to be analyzed and the reference image can be calculated according to the time waveform features and space waveform features of the magnetocardiogram.

[0116] The feature parameter extraction method and system based on magnetocardiogram spatio-temporal information in this embodiment can accurately describe the overall change characteristics of magnetocardiogram signals, improve the magnetocardiogram analysis ability, thus improving the sensitivity and specificity of magnetocardiogram detection, and making important contributions to the extraction of feature parameters in heart function judgment.

[0117] Please refer to Figure 10 , Figure 10 which is a schematic structural diagram of a magnetocardiogram analysis device based on spatio-temporal information provided by an embodiment of this application. As Figure 10 shown in

[0118] The wave group diagram drawing module 1010 is configured to obtain an original magnetocardiogram data set and draw a time wave group diagram and a space wave group diagram according to the original magnetocardiogram data set;

[0119] The band segmentation module 1020 is configured to segment time waveform data of at least one band in the time wave group diagram and segment space waveform data of at least one band in the space wave group diagram;

[0120] The feature extraction module 1030 is configured to determine a plurality of time waveform features according to the time waveform data and determine a plurality of space waveform features according to the space waveform data;

[0121] The magnetocardiogram analysis module 1040 is configured to calculate the waveform deviation degree between the original magnetocardiogram and the reference diagram according to a plurality of time waveform features and / or a plurality of space waveform features, and obtain the analysis result of the original magnetocardiogram according to the waveform deviation degree.

[0122] Furthermore, a plurality of time waveform features are determined according to the amplitude, curve shape, and alignment degree of each channel of the time wave of each band; the plurality of time waveform features include at least one or more of amplitude ratio, wave peak length ratio, positive wave shape fraction, negative wave shape fraction, elevation fraction, wave smoothness, and wave peak alignment;

[0123] Among them, the amplitude ratio is determined according to the maximum value of the average signal of all channels of each band; the wave peak length ratio is determined according to the wave peak maximum value channel in the band and the duration of the band; the positive wave shape fraction is determined according to the curve of the wave peak maximum value channel in the band; the negative wave shape fraction is determined according to the curve of the wave peak minimum value channel in the band; the elevation fraction is determined according to the amplitude of each channel in the band; the wave smoothness is determined according to the second-order difference value and the extreme value of each channel in the band; the wave peak alignment is determined according to the maximum amplitude position of each channel in the band.

[0124] Further, multiple spatial waveform features are determined according to the peak orientation, peak position, and curve shape of the spatial waves in each band; the multiple spatial waveform features include at least one or more of the number of channels with opposite-sign peaks, peak zero-line diagonal property, area zero-line diagonal property, and peak zero-line rotation angle;

[0125] Among them, the number of channels with opposite-sign peaks is determined according to the peak orientation and peak position of each peak in different bands; the peak zero-line diagonal property and the peak zero-line rotation angle are determined according to the peak zero-line of the band, including the R-wave peak zero-line diagonal property and the T-wave peak zero-line diagonal property; the peak zero-line is determined according to the amplitudes of each channel in a preset number of frames before and after the peak of the band; the area zero-line diagonal property is determined according to the area under the curve of each channel in the band; the peak zero-line rotation angle is determined according to the angle difference between the peak zero-line of the T band and the peak zero-line of the R band.

[0126] Further, the waveform deviation degree includes the time waveform deviation degree. When the magnetocardiogram analysis module 1040 is used to calculate the waveform deviation degree between the original magnetocardiogram and the reference diagram according to multiple time waveform features and / or multiple spatial waveform features, the magnetocardiogram analysis module 1040 is specifically used for:

[0127] Calculating the superimposed waveform score between the original magnetocardiogram and the reference diagram according to multiple time waveform features, and determining the superimposed waveform score as the time waveform deviation degree;

[0128] The calculation formula for the superimposed waveform score is:

[0129] ovws = a11×qrsttrt + max(a21×strs, a22×tsm) + max([a31×ttflt, a32×ttstpp, a33×ttstpn, a34×tmdf]);

[0130] Among them, ovws is the superimposed waveform score; a11, a21, a22, a31, a32, a33, a34 are preset coefficients; qrsttrt is the QRS / T average signal amplitude ratio; strs is the ST elevation score; tsm is the ST wave smoothness; ttflt is the peak length ratio of the T band; ttstpp is the positive wave shape score of the T band; ttstpn is the negative wave shape score of the T band; tmdf is the T-wave peak alignment.

[0131] Further, the waveform deviation degree includes the spatial waveform deviation degree. When the magnetocardiogram analysis module 1040 is used to calculate the waveform deviation degree between the original magnetocardiogram and the reference diagram according to multiple time waveform features and / or multiple spatial waveform features, the magnetocardiogram analysis module 1040 is specifically used for:

[0132] Calculate the channel waveform score of the original magnetocardiogram and the reference diagram based on multiple spatial waveform features, and determine the channel waveform score as the degree of spatial waveform deviation;

[0133] The calculation formula for the channel waveform score is:

[0134]

[0135] where cnws is the channel waveform score; b11, b12, b21, b22, b31, R0, R1, T0 are preset coefficients; zeroR is the diagonal property of the R-wave peak zero line; zeroqrs is the diagonal property of the QRS area zero line; risgnum is the number of channels with different signs between the R-wave peak and the T-wave peak; zeroRTrot is the rotation angle of the R-wave peak T-wave peak zero line; zeroT is the diagonal property of the T-wave peak zero line.

[0136] Furthermore, the waveform deviation degree includes the comprehensive waveform deviation degree. When the magnetocardiogram analysis module 1040 is used to calculate the waveform deviation degree of the original magnetocardiogram and the reference diagram according to multiple time waveform features and / or multiple spatial waveform features, the magnetocardiogram analysis module 1040 is specifically used for:

[0137] Calculate the superimposed waveform score of the original magnetocardiogram and the reference diagram according to multiple time waveform features;

[0138] Calculate the channel waveform score of the original magnetocardiogram and the reference diagram according to multiple spatial waveform features;

[0139] Calculate the comprehensive waveform score of the original magnetocardiogram and the reference diagram according to the superimposed waveform score and the channel waveform score, and determine the comprehensive waveform score as the comprehensive waveform deviation degree;

[0140] The calculation formula for the comprehensive waveform score is:

[0141] cpws = b1×ovws + b2×cnws;

[0142] where cpws is the comprehensive waveform score; b1, b2 are preset coefficients; ovws is the superimposed waveform score; cnws is the channel waveform score.

[0143] Furthermore, the wave band includes at least one of the following: QRS wave band, ST wave band, and T wave band, where the QRS wave band includes the R wave band.

[0144] It should be noted that for other corresponding descriptions of each functional module involved in the magnetocardiogram analysis device 1000 based on spatio-temporal information provided in this application embodiment, reference can be made to the corresponding descriptions in the above method, which will not be elaborated here.

[0145] Please refer to Figure 11 , Figure 11The following is a schematic structural diagram of an electronic device provided by an embodiment of the present application. As Figure 11 shown in the figure, the electronic device 1100 includes a processor 1110, a memory 1120, and a bus 1130.

[0146] The memory 1120 stores machine-readable instructions executable by the processor 1110. When the electronic device 1100 runs, the processor 1110 communicates with the memory 1120 through the bus 1130. When the machine-readable instructions are executed by the processor 1110, the steps of the magnetocardiogram analysis method based on spatio-temporal information provided in the above embodiment can be executed. For the specific implementation manner, reference can be made to the method embodiment, which will not be elaborated here.

[0147] An embodiment of the present application also provides a computer-readable storage medium. A computer program is stored on the computer-readable storage medium. When the computer program is run by a processor, the steps of the magnetocardiogram analysis method based on spatio-temporal information provided in the above embodiment can be executed. For the specific implementation manner, reference can be made to the method embodiment, which will not be elaborated here.

[0148] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the foregoing method embodiments, which will not be elaborated here.

[0149] In several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For another example, 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 couplings or direct couplings or communication connections shown or discussed with each other can be through some communication interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0150] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they can be located in one place, or they can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0151] In addition, in each embodiment of the present application, the functional units can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.

[0152] When the above-mentioned functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solution of this application, in essence, or the part that contributes to the prior art or part of this technical solution can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs.

[0153] Finally, it should be noted that the above-mentioned embodiments are only specific implementation manners of this application, used to illustrate the technical solutions of this application, rather than limiting it. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: any person skilled in the art within the technical scope disclosed in this application can still modify the technical solutions described in the foregoing embodiments or easily think of changes, or make equivalent replacements for some of the technical features; and these modifications, changes, or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

1. A magnetocardiogram analysis method based on spatio-temporal information, characterized in that, The method includes: Obtaining an original magnetocardiogram dataset, and drawing a time wave group diagram and a spatial wave group diagram according to the original magnetocardiogram dataset; Segmenting time waveform data of at least one band in the time wave group diagram, and segmenting spatial waveform data of at least one band in the spatial wave group diagram; Determining a plurality of time waveform features according to the time waveform data, and determining a plurality of spatial waveform features according to the spatial waveform data; the plurality of time waveform features are determined according to the amplitude, curve shape, and alignment degree of each channel of the time wave of each band; the plurality of time waveform features include amplitude ratio, peak length ratio, positive wave shape fraction, negative wave shape fraction, elevation fraction, wave smoothness, and peak alignment; wherein, the amplitude ratio is determined according to the maximum value of the average signal of all channels of each band; the peak length ratio is determined according to the peak maximum value channel in the band and the duration of the band; the positive wave shape fraction is determined according to the curve of the peak maximum value channel in the band; the negative wave shape fraction is determined according to the curve of the peak minimum value channel in the band; the elevation fraction is determined according to the amplitude of each channel in the band; the wave smoothness is determined according to the second-order difference value and the extreme value of each channel in the band; the peak alignment is determined according to the maximum amplitude position of each channel in the band; the plurality of spatial waveform features are determined according to the peak orientation, peak position, and curve shape of the spatial wave of each band; the plurality of spatial waveform features include the number of channels with opposite peak signs, peak zero-line diagonal property, area zero-line diagonal property, and peak zero-line rotation angle; wherein, the number of channels with opposite peak signs is determined according to the peak orientation and peak position of each peak in different bands; the peak zero-line diagonal property is determined according to the peak zero-line of the band, including R-wave peak zero-line diagonal property and T-wave peak zero-line diagonal property; the peak zero-line is determined according to the amplitudes of each channel in a preset number of frames before and after the peak of the band, counting the maximum amplitudes of each channel in the preset number of frames before and after the peak to form a new matrix, and then in this matrix, drawing a broken line separating positive and negative values according to the positive and negative of the peak as the peak zero-line of the band; the area zero-line diagonal property is determined according to the area under the curve of each channel in the band, counting the areas enclosed by the curve and the baseline of each channel in the band to form a new matrix, in which the area below the baseline is negative and the area above the baseline is positive, and drawing a broken line separating positive and negative values according to the positive and negative of the area as the area zero-line of the band; the peak zero-line rotation angle is determined according to the angle difference between the peak zero-line of the T band and the peak zero-line of the R band; Calculating the waveform deviation degree between the original magnetocardiogram and the reference diagram according to the plurality of time waveform features and the plurality of spatial waveform features, and obtaining the analysis result of the original magnetocardiogram according to the waveform deviation degree; the waveform deviation degree includes time waveform deviation degree and spatial waveform deviation degree; The calculating the waveform deviation degree between the original magnetocardiogram and the reference diagram according to the plurality of time waveform features and the plurality of spatial waveform features includes: Calculate the superimposed waveform score of the original magnetocardiogram and the reference diagram based on the multiple time waveform features, and determine the superimposed waveform score as the degree of deviation of the time waveform; The calculation formula of the superimposed waveform score is: ovws = a11×qrsttrt + max(a21×strs, a22×tsm) + max([a31×ttflt, a32×ttstpp, a33×ttstpn, a34×tmdf]); where ovws is the superimposed waveform score; a11, a21, a22, a31, a32, a33, a34 are preset coefficients; qrsttrt is the QRS / T average signal amplitude ratio; strs is the ST elevation score; tsm is the ST wave smoothness; ttflt is the peak length ratio of the T band; ttstpp is the positive wave shape score of the T band; ttstpn is the negative wave shape score of the T band; tmdf is the T wave peak alignment; Calculate the channel waveform score of the original magnetocardiogram and the reference diagram based on the multiple spatial waveform features, and determine the channel waveform score as the degree of deviation of the spatial waveform; The calculation formula of the channel waveform score is: where cnws is the channel waveform score; b11, b12, b21, b22, b31, R0, R1, T0 are preset coefficients; zeroR is the R wave peak zero line diagonal property; zeroqrs is the QRS area zero line diagonal property; rtsgnum is the number of channels with different signs between the R wave peak and the T wave peak; zeroRTrot is the R wave peak T wave peak zero line rotation angle; zeroT is the T wave peak zero line diagonal property.

2. The magnetocardiogram analysis method based on spatio-temporal information according to claim 1, wherein The degree of waveform deviation includes the comprehensive waveform deviation degree. Calculating the degree of waveform deviation of the original magnetocardiogram and the reference diagram based on the multiple time waveform features and the multiple spatial waveform features includes: Calculate the comprehensive waveform score of the original magnetocardiogram and the reference diagram according to the superimposed waveform score and the channel waveform score, and determine the comprehensive waveform score as the comprehensive waveform deviation degree; The calculation formula of the comprehensive waveform score is: cpws = b1×ovws + b2×cnws; where cpws is the comprehensive waveform score; b1, b2 are preset coefficients; ovws is the superimposed waveform score; cnws is the channel waveform score.

3. A magnetocardiogram analysis device based on spatio-temporal information, characterized in that, The magnetocardiogram analysis device based on spatio-temporal information includes: A wave group diagram drawing module, configured to obtain an original magnetocardiogram data set, and draw a time wave group diagram and a spatial wave group diagram according to the original magnetocardiogram data set; A band segmentation module, configured to segment time waveform data of at least one band in the time wave group diagram, and segment spatial waveform data of at least one band in the spatial wave group diagram; A feature extraction module, which is used to determine a plurality of time waveform features according to the time waveform data and determine a plurality of spatial waveform features according to the spatial waveform data; the plurality of time waveform features are determined according to the amplitude, curve shape, and alignment degree of each channel of the time wave in each band; the plurality of time waveform features include amplitude ratio, peak length ratio, positive wave shape fraction, negative wave shape fraction, elevation fraction, wave smoothness, and peak alignment; wherein, the amplitude ratio is determined according to the maximum value of the average signal of all channels in each band; the peak length ratio is determined according to the peak maximum value channel in the band and the duration of the band; the positive wave shape fraction is determined according to the curve of the peak maximum value channel in the band; the negative wave shape fraction is determined according to the curve of the peak minimum value channel in the band; the elevation fraction is determined according to the amplitude of each channel in the band; the wave smoothness is determined according to the second-order difference value and the extreme value of each channel in the band; the peak alignment is determined according to the maximum amplitude position of each channel in the band; the plurality of spatial waveform features are determined according to the peak orientation, peak position, and curve shape of the spatial wave in each band; the plurality of spatial waveform features include the number of channels with opposite peak signs, peak zero-line diagonal property, area zero-line diagonal property, and peak zero-line rotation angle; wherein, the number of channels with opposite peak signs is determined according to the peak orientation and peak position of each peak in different bands; the peak zero-line diagonal property is determined according to the peak zero-line of the band, including R-wave peak zero-line diagonal property and T-wave peak zero-line diagonal property; the peak zero-line is determined according to the amplitudes of each channel in a preset number of frames before and after the peak of the band, the maximum amplitudes of each channel in the preset number of frames before and after the peak are counted to form a new matrix, and then in this matrix, according to the positive and negative of the peak, a broken line separating the positive and negative values is drawn as the peak zero-line of the band; the area zero-line diagonal property is determined according to the area under the curve of each channel in the band, the areas enclosed by the curves and the baseline of each channel in the band are counted to form a new matrix, in this matrix, the area below the baseline is negative and the area above the baseline is positive, and a broken line separating the positive and negative values is drawn according to the positive and negative of the area as the area zero-line of the band; the peak zero-line rotation angle is determined according to the angle difference between the peak zero-line of the T band and the peak zero-line of the R band; A magnetocardiogram analysis module, which is used to calculate the waveform deviation degree between the original magnetocardiogram and the reference diagram according to the plurality of time waveform features and the plurality of spatial waveform features, and obtain the analysis result of the original magnetocardiogram according to the waveform deviation degree; the waveform deviation degree includes time waveform deviation degree and spatial waveform deviation degree; When the magnetocardiogram analysis module is used to calculate the waveform deviation degree between the original magnetocardiogram and the reference diagram according to the plurality of time waveform features and the plurality of spatial waveform features, the magnetocardiogram analysis module is specifically used for: Calculating the superimposed waveform fraction between the original magnetocardiogram and the reference diagram according to the plurality of time waveform features, and determining the superimposed waveform fraction as the time waveform deviation degree; The calculation formula of the superimposed waveform fraction is: ovws = a11 × qrsttrt + max(a21 × strs, a22 × tsm) + max([a31 × ttflt, a32 × ttstpp, a33 × ttstpn, a34 × tmdf]); Wherein, ovws is the superimposed waveform fraction; a11, a21, a22, a31, a32, a33, a34 are preset coefficients; qrsttrt is the QRS / T average signal amplitude ratio; strs is the ST elevation fraction; tsm is the ST wave smoothness; ttflt is the peak length ratio of the T wave band; ttstpp is the positive wave shape fraction of the T wave band; ttstpn is the negative wave shape fraction of the T wave band; tmdf is the T wave peak alignment. Calculate the channel waveform fraction of the original magnetocardiogram and the reference diagram according to the multiple spatial waveform features, and determine the channel waveform fraction as the spatial waveform deviation degree. The calculation formula of the channel waveform fraction is: Wherein, cnws is the channel waveform fraction; b11, b12, b21, b22, b31, R0, R1, T0 are preset coefficients; zeroR is the R wave peak zero line diagonal property; zeroqrs is the QRS area zero line diagonal property; rtsgnum is the number of channels with different signs between the R wave peak and the T wave peak; zeroRTrot is the R wave peak T wave peak zero line rotation angle; zeroT is the T wave peak zero line diagonal property.

4. An electronic device, characterized in that, Including: A processor, a memory and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device runs, the processor communicates with the memory through the bus. When the machine-readable instructions are run by the processor, the steps of the magnetocardiogram analysis method based on spatio-temporal information according to any one of claims 1 to 2 are executed.

5. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium. When the computer program is run by the processor, the steps of the magnetocardiogram analysis method based on spatio-temporal information according to any one of claims 1 to 2 are executed.

Images