Method and system for extracting distributed parameters of magnetocardiogram

Abstract

The present invention discloses a method and system for extracting distribution parameters of a magnetocardiogram, comprising: obtaining a magnetic field map at z moments in a TT segment of the magnetocardiogram; determining a value line of positive distribution at a position M pixels to the right of the magnetic field map i (i≤z); p , calculate the value line p S for each pixel rgb‑p,j Value; determine a negative pole distribution value line at a position N pixels away from the left side of the magnetic field map i n , calculate the value line n S for each pixel rgb‑n,j value; according to the magnetic field diagram i on the value line p , S for each pixel on n rgb‑p,j , S rgb‑n,j The distribution parameters of the magnetic field map i are calculated according to the distribution parameters of the magnetic field map at the zth moment, and the distribution parameters of the magnetocardiogram are calculated. The distribution parameters extracted by the present invention can well represent the pole distribution information in the magnetic field map, and the distribution parameters are applied to the coronary artery blockage degree assessment method based on the magnetocardiogram, which can effectively improve the prediction performance of the model.

Description

Technical Field

[0001] The present invention relates to the field of magnetocardiogram image processing, and more specifically, to a method and system for extracting distribution parameters of a magnetocardiogram. Background Art

[0002] Magnetocardiography (MCG) is a technology used to measure and record the electrical activity of the heart. Unlike the traditional electrocardiogram (ECG), MCG obtains the electrophysiological information of the heart by detecting the weak magnetic field generated by the heart. It can detect weaker signals than ECG and is suitable for certain cases that are difficult to detect with ECG.

[0003] Magnetocardiography can provide more accurate spatial distribution information of cardiac electrical activity, which helps doctors better understand the electrophysiological characteristics of the heart. However, precisely because this information is richer, it is more difficult to interpret the magnetocardiogram with the naked eye. Currently, there is a tendency to use an artificial intelligence (AI)-based magnetocardiogram severity prediction model for auxiliary diagnosis. Therefore, improving prediction accuracy is a goal that researchers are striving for.

[0004] The two-dimensional magnetic field map is a typical visualization method of the magnetocardiograph. The distribution of the poles in the magnetic field map is particularly important for distinguishing the severity of coronary artery disease. Usually, the magnetic field map of a normal person consists of a positive pole located in the upper left and a negative pole located in the lower right, distributed at approximately 45°.

[0005] Artificial intelligence is data-driven and cannot perceive image distribution information as intuitively as the human eye. It needs to reasonably and accurately extract feature parameters for model training and achieve sufficient accuracy after testing before it can be used for reasoning and prediction of the severity of magnetocardiograms. Deep learning can automatically extract distribution information from images, but it usually requires tens of thousands or even millions of data for training to achieve a good accuracy rate, and the decision-making process is poorly interpretable. There is currently no public magnetocardiogram dataset available for training; traditional machine learning methods, such as logistic regression and decision trees, require less training data and are relatively easy to analyze the decision-making process, but manual feature selection is required. There are currently no available parameters that can describe image distribution for traditional machine learning training, so this type of machine learning model is not very sensitive to image distribution. When there is an extremely weak pole in the lower left or upper right and the magnetic field angle is approximately 45°, traditional machine learning methods often cannot predict it correctly because they do not have relevant distribution parameters. Summary of the invention

[0006] An object of the present invention is to provide a method and system for extracting distributed parameters of a magnetocardiogram to at least solve the above-mentioned problem.

[0007] In order to achieve the purpose and other advantages of the present invention, a method for extracting distributed parameters of a magnetocardiogram is provided, comprising:

[0008] Obtain the magnetic field map at the zth moment in the magnetocardiogram TT segment, and set the magnetic field map at the ith moment as magnetic field map i, where i ≤ z and is an integer. Suppose the magnetic field map i is l × l Pixel distribution;

[0009] A vertical line is determined at a position M pixels to the right of the magnetic field map i as the value line of the positive distribution p , and the value line p Without passing through the pole, M< l / 2 and is an integer, in the value line p Calculate the composite RGB value of each pixel j, denoted as S rgb-p,j ;

[0010] A vertical line is determined at a position N pixels to the left of the magnetic field map i as the value line of the negative pole distribution. n , and the value line n Not passing through the pole, N< l / 2 and is an integer, in the value line n Calculate the composite RGB value of each pixel j, denoted as S rgb-n,j ;

[0011] According to the magnetic field diagram i in the value line p S for each pixel j on rgb-p,j and in the value line n S for each pixel j on rgb-n,j , calculate the distribution parameter Γ of the magnetic field diagram i pi , Γ ni ;

[0012]

[0013]

[0014]

[0015]

[0016] According to the magnetic field distribution parameter Γ at z moments pi , Γ ni , calculate the magnetocardiogram distribution parameter Γ TT,p , Γ TT,n ;

[0017]

[0018] .

[0019] Preferably, S rgb-p,j= a 1 ×R + a 2 × G + a 3 × B, where a 1 、a 2 、a 3 is the channel coefficient, and R, G, and B are distributed as the red value, green value, and blue value of the corresponding pixel.

[0020] Preferably, S rgb-n,j = b 1 ×R + b 2 × G + b 3 × B, where b 1 、b 2 、b 3 is the channel coefficient, and R, G, and B are distributed as the red value, green value, and blue value of the corresponding pixel.

[0021] The present invention also provides a magnetocardiogram distribution parameter extraction system, comprising:

[0022] The preprocessing module is used to obtain the magnetic field map at the zth moment in the magnetocardiogram TT segment, and set the magnetic field map at the i-th moment as the magnetic field map i, i ≤ z and is an integer, and the magnetic field map i is l × l Pixel distribution;

[0023] A magnetic field map distribution parameter extraction module is used to determine a vertical line at a position M pixels to the right of the magnetic field map i as a value line for the positive distribution. p , and the value line p Without passing through the pole, M< l / 2 and is an integer, in the value line p Calculate the composite RGB value of each pixel j, denoted as S rgb-p,j ; Determine a vertical line at a position N pixels to the left of the magnetic field map i as the value line of the negative pole distribution n , and the value line n Without passing through the pole, N< l / 2 and is an integer, in the value line n Calculate the composite RGB value of each pixel j, denoted as S rgb-n,j According to the magnetic field diagram i in the value line p S for each pixel j on rgb-p,j and in the value line n S for each pixel j on rgb-n,j , calculate the distribution parameter Γ of the magnetic field diagram i pi , Γ ni ;

[0024]

[0025]

[0026]

[0027]

[0028] The magnetocardiogram distribution parameter extraction module is used to extract the magnetic field distribution parameter Γ at z moments. pi , Γ ni , calculate the magnetocardiogram distribution parameter Γ TT,p , Γ TT,n ;

[0029]

[0030] .

[0031] The present invention also provides a method for assessing the degree of coronary artery blockage based on magnetocardiography, comprising:

[0032] Obtain a magnetocardiogram dataset, randomly divide it into a training set and a test set, and extract the characteristic parameters and distribution parameters of each magnetocardiogram;

[0033] According to the coronary angiography result corresponding to each magnetocardiogram in the training set, a label is given to each magnetocardiogram;

[0034] Using the training set, taking the feature parameters and distribution parameters as input and the labels as output, training a traditional machine learning model to obtain a trained classification model;

[0035] The test set and the classification model are used to predict the severity of coronary artery blockage in each magnetocardiogram with the characteristic parameters and distribution parameters as input; wherein,

[0036] The characteristic parameters include angle change information and position change information in the magnetic field diagram and current diagram of the magnetocardiogram TT segment, and amplitude information in the butterfly diagram of the magnetocardiogram TT segment;

[0037] The distribution parameters are extracted using the above-mentioned magnetocardiogram distribution parameter extraction method.

[0038] The present invention also provides a coronary artery blockage degree assessment system based on magnetocardiogram, comprising:

[0039] An acquisition module is used to acquire a magnetocardiogram dataset, randomly divide it into a training set and a test set, and extract characteristic parameters and distribution parameters from each magnetocardiogram;

[0040] A labeling module, which is used to give a label to each magnetocardiogram according to the coronary angiography result corresponding to each magnetocardiogram in the training set;

[0041] A training module, which is used to use the training set, the feature parameters and the distribution parameters as input, and the label as output to train the traditional machine learning model to obtain a trained classification model;

[0042] A testing module is used to use the test set and the classification model, and take the characteristic parameters and distribution parameters as input to predict the severity of coronary artery blockage in each magnetocardiogram; wherein,

[0043] The characteristic parameters include angle change information and position change information in the magnetic field diagram and current diagram of the magnetocardiogram TT segment, and amplitude information in the butterfly diagram of the magnetocardiogram TT segment;

[0044] The distribution parameters are extracted using the above-mentioned magnetocardiogram distribution parameter extraction method.

[0045] The present invention also provides an electronic device, comprising: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the programs include methods for executing the above-mentioned method.

[0046] The present invention also provides a computer-readable storage medium storing a computer program used in combination with an electronic device, wherein the computer program can be executed by a processor to implement the above method.

[0047] The present invention has at least the following beneficial effects:

[0048] The present invention proposes a method for extracting distribution parameters of a magnetocardiogram, and uses the extracted distribution parameters together with the existing magnetocardiogram feature parameters as the input of a traditional machine learning model, making the input parameters more comprehensive, effectively solving the problem that the traditional machine learning model cannot distinguish image distribution, and improving the prediction performance of the model.

[0049] Other advantages, objectives and features of the present invention will be embodied in part through the following description, and in part will be understood by those skilled in the art through study and practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is a schematic diagram of value lines, pixels, and poles in a magnetic field diagram i according to an embodiment of the present invention;

[0051] Figure 2 It is a flowchart of a method for assessing the degree of coronary artery blockage based on magnetocardiography according to an embodiment of the present invention. DETAILED DESCRIPTION

[0052] The present invention is further described in detail below in conjunction with embodiments and drawings so that those skilled in the art can implement the invention with reference to the description.

[0053] It should be understood that the terms such as “having”, “including” and “comprising” used herein do not exclude the existence or addition of one or more other elements or combinations thereof.

[0054] It should be noted that the experimental methods described in the following embodiments are conventional methods unless otherwise specified, and the reagents and materials can be obtained from commercial channels unless otherwise specified.

[0055] In one aspect of the present invention, Figure 1 As shown, a method for extracting distribution parameters of a magnetocardiogram is provided, comprising:

[0056] Step 1: Obtain the magnetic field diagram at the zth moment in the magnetocardiogram TT segment, and set the magnetic field diagram at the ith moment as magnetic field diagram i, where i ≤ z and is an integer. Suppose the magnetic field diagram i is l × l Pixel distribution.

[0057] For example, z ranges from 5 to 20. l The value ranges from 300 to 800.

[0058] Step 2: Determine a vertical line at a position M pixels to the right of the magnetic field map i as the value line of the positive distribution p , and the value line p Without passing through the pole, M< l / 2 and is an integer, in the value line p Calculate the composite RGB value of each pixel j, denoted as S rgb-p,j .

[0059] Furthermore, S rgb-p,j = a 1 ×R + a 2 × G + a 3 × B, where a 1 、a 2 、a 3 is the channel coefficient, and R, G, and B are distributed as the red value, green value, and blue value of the corresponding pixel.

[0060] Step 3: Determine a vertical line at a position N pixels to the left of the magnetic field map i as the value line of the negative pole distribution. n , and the value line n Not passing through the pole, N< l / 2 and is an integer, in the value line n Calculate the composite RGB value of each pixel j, denoted as S rgb-n,j.

[0061] Furthermore, S rgb-n,j = b 1 ×R + b 2 ×G + b 3 ×B, where b 1 、b 2 、b 3 is the channel coefficient, and R, G, and B are distributed as the red value, green value, and blue value of the corresponding pixel.

[0062] Step 4: According to the magnetic field diagram i, the value line p S for each pixel j on rgb-p,j and in the value line n S for each pixel j on rgb-n,j , calculate the distribution parameter Γ of the magnetic field diagram i pi , Γ ni ;

[0063]

[0064]

[0065]

[0066] .

[0067] Step 5: Calculate the distribution parameter Γ based on the magnetic field diagram at the zth moment pi , Γ ni , calculate the magnetocardiogram distribution parameter Γ TT,p , Γ TT,n ;

[0068]

[0069] .

[0070] In this embodiment, the magnetic field diagrams at z moments in the TT segment of the magnetocardiogram are obtained, and the magnetic field diagrams at each moment are processed respectively, the distribution parameters of the magnetic field diagrams at each moment are extracted, and the distribution parameters of the magnetic field diagrams at z moments are averaged to obtain the distribution parameters of the magnetocardiogram. The extraction process is simple, and the obtained distribution parameters can well represent the pole distribution information in the magnetic field diagram.

[0071] Another aspect of the present invention further provides a magnetocardiogram distribution parameter extraction system, comprising:

[0072] The preprocessing module is used to obtain the magnetic field map at the zth moment in the magnetocardiogram TT segment, and set the magnetic field map at the i-th moment as the magnetic field map i, i ≤ z and is an integer, and the magnetic field map i is l ×l Pixel distribution.

[0073] A magnetic field map distribution parameter extraction module is used to determine a vertical line at a position M pixels to the right of the magnetic field map i as a value line for the positive distribution. p , and the value line p Without passing through the pole, M< l / 2 and is an integer, in the value line p Calculate the composite RGB value of each pixel j, denoted as S rgb-p,j ; Determine a vertical line at a position N pixels to the left of the magnetic field map i as the value line of the negative pole distribution n , and the value line n Without passing through the pole, N< l / 2 and is an integer, in the value line n Calculate the composite RGB value of each pixel j, denoted as S rgb-n,j According to the magnetic field diagram i in the value line p S for each pixel j on rgb-p,j and in the value line n S for each pixel j on rgb-n,j , calculate the distribution parameter Γ of the magnetic field diagram i pi , Γ ni ;

[0074]

[0075]

[0076]

[0077]

[0078] The magnetocardiogram distribution parameter extraction module is used to extract the magnetic field distribution parameter Γ at z moments. pi , Γ ni , calculate the magnetocardiogram distribution parameter Γ TT,p , Γ TT,n ;

[0079]

[0080] .

[0081] The preprocessing module, magnetic field map distribution parameter extraction module and magnetocardiogram distribution parameter extraction module of this embodiment are all constructed by software programs, and the magnetocardiogram distribution parameter extraction system of this embodiment is constructed based on the same inventive concept as the above-mentioned magnetocardiogram distribution parameter extraction method, which will not be repeated here.

[0082] In another aspect of the present invention, Figure 2 As shown, a method for assessing the degree of coronary artery blockage based on magnetocardiography is provided, comprising:

[0083] S1. Obtain a magnetocardiogram dataset, randomly divide it into a training set and a test set, and extract characteristic parameters and distribution parameters of each magnetocardiogram. The characteristic parameters include angle change information and position change information in the magnetocardiogram TT segment magnetic field diagram and current diagram, and amplitude information in the magnetocardiogram TT segment butterfly diagram. The distribution parameters are extracted using the above-mentioned magnetocardiogram distribution parameter extraction method.

[0084] Specifically, the patient's magnetocardiogram data is collected by a magnetocardiograph and the patient's coronary angiography results are correspondingly obtained to form a magnetocardiogram data set; the method for extracting the characteristic parameters is an existing technology and will not be repeated here.

[0085] Exemplarily, the characteristic parameters include T peak magnetic field angle, T peak current angle, T peak maximum-minimum ratio, TT maximum current angle, TT maximum magnetic field angle, TT minimum current angle, TT minimum magnetic field angle, positive pole area change standard deviation, and negative pole area change standard deviation.

[0086] S2. According to the coronary angiography result corresponding to each magnetocardiogram in the training set, a label is given to each magnetocardiogram.

[0087] For example, if the degree of coronary angiography blockage is 70% as the boundary for dividing negative and positive, the magnetocardiogram with a coronary angiography result of less than 70% is marked as negative, and the magnetocardiogram with a coronary angiography result greater than or equal to 70% is marked as positive.

[0088] S3. Using the training set, taking the feature parameters and distribution parameters as input and the label as output, the traditional machine learning model is trained to obtain a trained classification model.

[0089] Exemplarily, the traditional machine learning models include a random forest model (RF) and an extreme gradient boosting model (XGB).

[0090] S4. Using the test set and the classification model, taking the characteristic parameters and distribution parameters as input, the severity of coronary artery blockage of each magnetocardiogram is predicted.

[0091] In this embodiment, the distribution parameters extracted by the method of the present invention and the existing magnetocardiogram feature parameters are used as the input of the traditional machine learning model to train the traditional machine learning model, so that the training parameters of the model are more comprehensive, which effectively solves the problem that the traditional machine learning model cannot distinguish the image distribution and improves the prediction performance of the model.

[0092] In another aspect, the present invention provides a system for assessing the degree of coronary artery blockage based on magnetocardiography, comprising:

[0093] An acquisition module is used to acquire a magnetocardiogram dataset, randomly divide it into a training set and a test set, and extract characteristic parameters and distribution parameters from each magnetocardiogram;

[0094] A labeling module, which is used to give a label to each magnetocardiogram according to the coronary angiography result corresponding to each magnetocardiogram in the training set;

[0095] A training module, which is used to use the training set, the feature parameters and the distribution parameters as input, and the label as output to train the traditional machine learning model to obtain a trained classification model;

[0096] A testing module is used to use the test set and the classification model, and take the characteristic parameters and distribution parameters as input to predict the severity of coronary artery blockage in each magnetocardiogram; wherein,

[0097] The characteristic parameters include angle change information and position change information in the magnetic field diagram and current diagram of the magnetocardiogram TT segment, and amplitude information in the butterfly diagram of the magnetocardiogram TT segment;

[0098] The distribution parameters are extracted using the above-mentioned magnetocardiogram distribution parameter extraction method.

[0099] The acquisition module, labeling module, training module and testing module of this embodiment are all constructed by software programs, and the coronary artery blockage assessment system based on magnetocardiography of this embodiment is constructed based on the same inventive concept as the above-mentioned coronary artery blockage degree assessment method based on magnetocardiography, which will not be repeated here.

[0100] On the other hand, the present invention provides an electronic device, comprising: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the program includes a method for executing the above-mentioned method.

[0101] Exemplarily, the electronic device is a device including a processor (CPU / MCU / SOC) and a memory (ROM / RAM), such as a desktop computer, a portable computer, a smart phone, etc.

[0102] In one aspect, the present invention provides a computer-readable storage medium storing a computer program for use in combination with an electronic device, wherein the computer program can be executed by a processor to implement the above method.

[0103] Exemplarily, the computer-readable storage medium includes: ROM, RAM, magnetic disk, optical disk, and other media that can store program codes.

[0104] The present invention will be further described below with specific embodiments:

[0105] Embodiment 1:

[0106] The method for extracting distribution parameters of a magnetocardiogram includes:

[0107] (1) Obtain the magnetic field diagram at the zth moment in the TT segment of the magnetocardiogram, where z is 10, and set the magnetic field diagram at the ith moment to be magnetic field diagram i, where i ≤ 10 and is an integer. Set the magnetic field diagram i to be l × l Pixel distribution, l The value is 500;

[0108] (2) Determine a vertical line at a position 50 pixels to the right of the magnetic field map i as the value line of the positive distribution p , on the value line p Calculate the composite RGB value of each pixel j, denoted as S rgb-p,j ;

[0109] S rgb-p,j = a 1 ×R + a 2 × G + a 3 × B

[0110] Among them, a 1 、a 2 、a 3 is the channel coefficient, which takes values ​​of 0, 1, and 0 respectively, and the R, G, and B distributions are the red value, green value, and blue value of the corresponding pixel;

[0111] (3) Determine a vertical line at a position 50 pixels to the left of the magnetic field map i as the value line of the negative pole distribution n , on the value line n Calculate the composite RGB value of each pixel j, denoted as S rgb-n,j ;

[0112] S rgb-n,j = b 1 ×R + b 2 × G + b 3 × B

[0113] Among them, b 1 、b 2 、b 3 is the channel coefficient, which takes values ​​of 0, 1, and 0 respectively, and the R, G, and B distributions are the red value, green value, and blue value of the corresponding pixel;

[0114] (4) Value line p and value line n Each has l pixels, then the distribution parameter Γ of the magnetic field map i pi , Γ ni for:

[0115]

[0116]

[0117]

[0118] ;

[0119] (5) The distribution parameter in the TT segment of the magnetocardiogram is Γ TT,p , Γ TT,n , Γ TT,p is the magnetic field diagram at 10 moments in the TT segment, i.e., the distribution parameter Γ of the 10 magnetic field diagrams pi The average value of Γ TT,n is the magnetic field diagram at 10 moments in the TT segment, i.e., the distribution parameter Γ of the 10 magnetic field diagrams ni The average value of is:

[0120]

[0121] .

[0122] Embodiment 2:

[0123] A method for assessing the degree of coronary artery blockage based on magnetocardiography, comprising:

[0124] (1) Obtaining a magnetocardiogram data set, randomly dividing it into a training set and a test set, and extracting the distribution parameters in each magnetocardiogram using the method of Example 1, and extracting the characteristic parameters of the magnetic field diagram, current diagram and butterfly diagram in the magnetocardiogram using the prior art for later use, wherein the characteristic parameters are T peak magnetic field angle, T peak current angle, T peak maximum-minimum ratio, TT maximum current angle, TT maximum magnetic field angle, TT minimum current angle, TT minimum magnetic field angle, positive pole area change standard deviation, and negative pole area change standard deviation;

[0125] (2) Based on the coronary angiography results corresponding to each MCG in the training set, the coronary angiography blockage degree of 70% was used as the boundary to divide the MCG into positive and negative, and a label was given to each MCG;

[0126] (3) Using the training set, feature parameters and distribution parameters as input, and labels as output, the RF model is trained to obtain a trained RF classification model;

[0127] (4) Using the test set and the trained RF classification model, with feature parameters and distribution parameters as input, the severity of coronary artery blockage in each magnetocardiogram was predicted to obtain the corresponding negative and positive prediction results of coronary artery blockage. Negative means that the blockage degree is less than 70%, and positive means that the blockage degree is greater than or equal to 70%.

[0128] The above evaluation method was used to train the model on 739 cases of magnetic cardiology data, of which 592 cases were randomly divided as training sets and 147 cases were used as test sets. When the distribution parameters were not added, the accuracy of the RF model on the test set was 87.98%, the sensitivity was 88.21%, and the specificity was 84.33%. After adding the distribution parameters, the accuracy, sensitivity, and specificity of the RF model on the test set were 89.57%, 91.36%, and 87.80%, respectively.

[0129] Embodiment 3:

[0130] A method for assessing the degree of coronary artery blockage based on magnetocardiography, comprising:

[0131] (1) Obtaining a magnetocardiogram data set, randomly dividing it into a training set and a test set, and extracting the distribution parameters in each magnetocardiogram using the method of Example 1, and extracting the characteristic parameters of the magnetic field diagram, current diagram and butterfly diagram in the magnetocardiogram using the prior art for later use, wherein the characteristic parameters are T peak magnetic field angle, T peak current angle, T peak maximum-minimum ratio, TT maximum current angle, TT maximum magnetic field angle, TT minimum current angle, TT minimum magnetic field angle, positive pole area change standard deviation, negative pole area change standard deviation, RT angle, sum of positive pole position changes, and sum of negative pole position changes;

[0132] (2) Based on the coronary angiography results corresponding to each MCG in the training set, the coronary angiography blockage degree of 70% was used as the boundary to divide the MCG into positive and negative, and a label was given to each MCG;

[0133] (3) Using the training set, feature parameters and distribution parameters as input, and labels as output, the XGB model is trained to obtain a trained XGB classification model;

[0134] (4) Using the test set and the trained XGB classification model, with feature parameters and distribution parameters as input, the severity of coronary artery occlusion in each magnetocardiogram was predicted to obtain the corresponding negative and positive prediction results of coronary artery occlusion. Negative means that the degree of occlusion is less than 70%, and positive means that the degree of occlusion is greater than or equal to 70%.

[0135] The above evaluation method was used to train the model on 513 cases of magnetic cardiology data, of which 410 cases were randomly divided as training sets and 103 cases were used as test sets. When the distribution parameters were not added, the accuracy of the XGB model on the test set was 81.75%, the sensitivity was 82.54%, and the specificity was 80.95%; after adding the distribution parameters, the accuracy, sensitivity, and specificity of the XGB model on the test set were 86.51%, 88.89%, and 84.13%, respectively.

[0136] The number of devices and processing scales described here are used to simplify the description of the present invention. The application, modification and variation of the method and system for extracting distributed parameters of magnetocardiogram of the present invention will be obvious to those skilled in the art.

[0137] Although the embodiments of the present invention have been disclosed as above, they are not limited to the applications listed in the specification and the implementation modes, and they can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and the illustrations shown and described herein.

Claims

1. A method for extracting distribution parameters of a magnetocardiogram, characterized in that: include: Obtain the magnetic field map at the zth moment in the magnetocardiogram TT segment, and set the magnetic field map at the ith moment as magnetic field map i, where i ≤ z and is an integer. Suppose the magnetic field map i is l × l Pixel distribution; A vertical line is determined at a position M pixels to the right of the magnetic field map i as the value line of the positive distribution p , and the value line p Without passing through the pole, M < l / 2 and is an integer, in the value line p Calculate the composite RGB value of each pixel j, denoted as S rgb-p,j ; A vertical line is determined at a position N pixels to the left of the magnetic field map i as the value line of the negative pole distribution. n , and the value line n Not passing through the pole, N < l / 2 and is an integer, in the value line n Calculate the composite RGB value of each pixel j, denoted as S rgb-n,j ; According to the magnetic field diagram i in the value line p S for each pixel j on rgb-p,j and in the value line n S for each pixel j on rgb-n,j , calculate the distribution parameter Γ of the magnetic field diagram i pi , Γ ni ; According to the magnetic field distribution parameter Γ at z moments pi , Γ ni , calculate the magnetocardiogram distribution parameter Γ TT,p , Γ TT,n ; 。 2. The method for extracting distributed parameters of a magnetocardiogram according to claim 1, characterized in that: S rgb-p,j = a1×R + a2×G + a3×B, where a1, a2, and a3 are channel coefficients, and R, G, and B are distributed as the red, green, and blue values ​​of the corresponding pixels.

3. The method for extracting distributed parameters of a magnetocardiogram according to claim 1, characterized in that: S rgb-n,j = b1×R + b2× G + b3 × B, where b1, b2, b3 are channel coefficients, and R, G, B are distributed as the red, green, and blue values ​​of the corresponding pixels.

4. Magnetocardiogram distribution parameter extraction system, characterized in that: include: The preprocessing module is used to obtain the magnetic field map at the zth moment in the magnetocardiogram TT segment, and set the magnetic field map at the i-th moment as the magnetic field map i, i ≤ z and is an integer, and the magnetic field map i is l × l Pixel distribution; A magnetic field map distribution parameter extraction module is used to determine a vertical line at a position M pixels to the right of the magnetic field map i as a value line for the positive distribution. p , and the value line p Without passing through the pole, M < l / 2 and is an integer, in the value line p Calculate the composite RGB value of each pixel j, denoted as S rgb-p,j ; Determine a vertical line at a position N pixels to the left of the magnetic field map i as the value line of the negative pole distribution n , and the value line n Not passing through the pole, N < l / 2 and is an integer, in the value line n Calculate the composite RGB value of each pixel j, denoted as S rgb-n,j According to the magnetic field diagram i in the value line p S for each pixel j on rgb-p,j and in the value line n S for each pixel j on rgb-n,j , calculate the distribution parameter Γ of the magnetic field diagram i pi , Γ ni ; The magnetocardiogram distribution parameter extraction module is used to extract the magnetic field distribution parameter Γ at z moments. pi , Γ ni , calculate the magnetocardiogram distribution parameter Γ TT,p , Γ TT,n ; 。 5. A method for assessing the degree of coronary artery blockage based on magnetocardiography, characterized in that: include: Obtain a magnetocardiogram dataset, randomly divide it into a training set and a test set, and extract the characteristic parameters and distribution parameters of each magnetocardiogram; According to the coronary angiography result corresponding to each magnetocardiogram in the training set, a label is given to each magnetocardiogram; Using the training set, taking the feature parameters and distribution parameters as input and the labels as output, training a traditional machine learning model to obtain a trained classification model; Using the test set and the classification model, taking the characteristic parameters and distribution parameters as input, predicting the severity of coronary artery occlusion of each magnetocardiogram; in, The characteristic parameters include angle change information and position change information in the magnetic field diagram and current diagram of the magnetocardiogram TT segment, and amplitude information in the butterfly diagram of the magnetocardiogram TT segment; The distribution parameters are extracted using the magnetocardiogram distribution parameter extraction method as described in any one of claims 1 to 3.

6. A coronary artery blockage assessment system based on magnetocardiography, characterized in that: include: An acquisition module is used to acquire a magnetocardiogram dataset, randomly divide it into a training set and a test set, and extract characteristic parameters and distribution parameters from each magnetocardiogram; A labeling module, which is used to give a label to each magnetocardiogram according to the coronary angiography result corresponding to each magnetocardiogram in the training set; A training module, which is used to use the training set, the feature parameters and the distribution parameters as input, and the label as output to train the traditional machine learning model to obtain a trained classification model; A testing module, which is used to use the test set and the classification model, and take the characteristic parameters and distribution parameters as input to predict the severity of coronary artery blockage of each magnetocardiogram; in, The characteristic parameters include angle change information and position change information in the magnetic field diagram and current diagram of the magnetocardiogram TT segment, and amplitude information in the butterfly diagram of the magnetocardiogram TT segment; The distribution parameters are extracted using the magnetocardiogram distribution parameter extraction method as described in any one of claims 1 to 3.

7. An electronic device, characterized in that include: one or more processors; Memory; and, one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs including instructions for executing the method of claim 5.

8. A computer-readable storage medium, characterized in that A computer program stored in an electronic device for use in combination, wherein the computer program can be executed by a processor to implement the method of claim 5.

Images