Coronary artery analysis processing method and system based on magnetic resonance CNR value

Abstract

The invention relates to the technical field of medical data analysis, in particular to a coronary artery analysis processing method and system based on a magnetic resonance CNR value, and the method comprises the following steps: S1, collecting coronary artery image data, carrying out the preprocessing of the collected coronary artery image data, obtaining the preprocessing data, inputting the preprocessing data into an ROI recognition model, and obtaining a target ROI; and S2, based on the target ROI, using an MRCA method to obtain an MRCA image of the target ROI, and using a final dynamic adaptability formula to obtain a CNR value of an associated reference region in the MRCA image. According to the method, the efficiency and the precision of target region extraction are improved by adopting the image enhancement and region extraction technology combining the edge-preserving normal gradient vector flow and the active contour model, the coronary artery stenosis degree is diagnosed by using the method combining the MRCA image and the CNR value, and the working efficiency of imaging doctors is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical data analysis, and particularly to a coronary artery analysis and processing method and system based on the CNR value of magnetic resonance. Background Art

[0002] Magnetic resonance coronary angiography (MRCA) technology is a non-invasive, non-radiative and non-enhanced coronary angiography examination, which has good adaptability and clinical guiding significance for early coronary heart disease screening. With the progress of magnetic resonance hardware technology, its imaging quality has been improving year by year. The main coronary artery and proximal branch can be clearly displayed, providing a safe and reliable means for non-invasive examination of coronary heart disease, realizing early screening and accurate diagnosis of coronary heart disease, and having very important practical significance for preventing fatal "sudden death" and taking scientific and effective treatment; However, the MRCA technology has problems such as relatively complex operation, low success rate and low image spatial resolution. Moreover, at present, MRCA still relies solely on the subjective judgment of diagnosticians to determine the degree of coronary artery stenosis. This method has problems such as strong subjectivity, low accuracy and large difference. And since MRCA is a magnetic resonance examination, if the lumen diameter of the coronary artery is reduced and the signal intensity is decreased solely by subjective judgment, the degree of coronary artery stenosis cannot be accurately diagnosed. Therefore, it is very necessary to relieve the working pressure of radiologists in reconstructing images and assist them to complete image diagnosis objectively and efficiently. Summary of the Invention

[0003] In order to overcome the drawback of being unable to accurately diagnose the degree of coronary artery stenosis, the present invention provides a coronary artery analysis and processing method and system based on the CNR value of magnetic resonance.

[0004] Technical Solution: A coronary artery analysis and processing method based on the CNR value of magnetic resonance includes the following steps: S1: Collect coronary artery image data, preprocess the collected coronary artery image data to obtain preprocessed data, and input the preprocessed data into an ROI recognition model to obtain a target ROI; S2: Based on the target ROI, use the MRCA method to obtain an MRCA image of the target ROI, and use the final dynamic adaptability formula to obtain the CNR value of the associated reference region in the MRCA image; S3: Obtain a comprehensive CNR value by using a weighted average method according to the CNR value of the associated reference region in the MRCA image; S4: Perform stenosis ratio feature analysis according to the comprehensive CNR value to obtain stenosis degree data of the corresponding cross-section.

[0005] Preferably, the collecting coronary artery image data, preprocessing the collected coronary artery image data to obtain preprocessed coronary artery images, and inputting the preprocessed coronary artery images into an ROI recognition model to obtain a target ROI include: using a multi-plane imaging and multi-angle reconstruction method based on the criteria of minimizing artifacts and anatomical coverage, screening the collected coronary artery image data to obtain standard coronary artery images, and performing preprocessing of data enhancement and contrast enhancement on the standard coronary artery images to obtain preprocessed coronary artery images, and inputting the preprocessed coronary artery images into the ROI recognition model to obtain a target ROI.

[0006] Preferably, the inputting the preprocessed coronary artery images into the ROI recognition model to obtain a target ROI includes: the ROI recognition model uses a method combining edge-preserving normal gradient vector flow and an active contour model to extract the cardiac region. Specifically, historical coronary artery image data is used to train a convolutional neural network model, the preprocessed coronary artery images are enhanced using edge-preserving normal gradient vector flow to obtain enhanced images, and based on the enhanced images combined with the active contour model, the target contour of the image is determined by minimizing the energy functional to remove tissues of non-interest; based on the regularized flow structure network of the affine coupling layer, coronary artery segmentation is performed, and using manually annotated coronary artery data, the network parameters are iteratively trained using a gradient descent optimization algorithm; the centerline of the blood vessel is detected based on the Kalman filter algorithm, and combined with a prior knowledge-driven blood vessel identification and re-identification program, the automatic segmentation result is corrected and optimized, and finally a target ROI is obtained.

[0007] Preferably, the using the MRCA method based on the target ROI to obtain an MRCA image of the target ROI and using the final dynamic adaptability formula to obtain the CNR value of the associated reference region in the MRCA image includes: selecting the normal blood vessel segment closest to the lesion but not affected as the proximal reference point, calculating the proximal reference point based on the MRCA image and the final dynamic adaptability formula to obtain the CNR value of the proximal reference lumen of the lesion; selecting the normal blood vessel segment farthest from the lesion but not affected as the distal reference point, calculating the distal reference point based on the MRCA image and the final dynamic adaptability formula to obtain the CNR value of the distal reference lumen of the lesion.

[0008] Preferably, the using the MRCA method based on the target ROI to obtain an MRCA image of the target ROI and using the final dynamic adaptability formula to obtain the CNR value of the associated reference region in the MRCA image includes: obtaining the relevant data of the target patient, the relevant data of the proximal reference point, and the relevant data of the distal reference point, and adjusting the final dynamic adaptability formula according to the relevant data of the target patient, where the final dynamic adaptability formula is: ; In the formula, is the contrast noise ratio of the cross-section to be measured; is the signal intensity in the lumen of the cross-section of the coronary artery segment to be measured; is the absolute value of the signal intensity of the chest wall muscle; is the standard deviation of the air background noise signal intensity; is the dynamic adjustment factor for adjusting the CNR calculation according to the individual characteristics of the patient.

[0009] Preferably, obtaining the relevant data of the target patient, the relevant data of the proximal reference point, and the relevant data of the distal reference point, and adjusting the final dynamic adaptability formula according to the relevant data of the target patient includes: the relevant data of the target patient includes the age, weight, and basic disease data of the target patient, and using the dynamic adjustment formula according to the relevant data of the target patient to obtain the dynamic adjustment factor, where the dynamic adjustment formula is: ; In the formula, is the dynamic adjustment factor; is the weight adjustment coefficient; is the age of the target patient; is the average age of the target population; is the weight of the target patient; is the average weight of the target population; is the binary marker of the th disease of the target patient; is the th disease weight.

[0010] Preferably, obtaining the comprehensive CNR value using the weighted average method according to the CNR value of the associated reference region in the MRCA image includes: obtaining the comprehensive CNR value using the comprehensive calculation formula according to the CNR value of the proximal reference lumen of the lesion and the CNR value of the distal reference lumen of the lesion, where the comprehensive calculation formula is: ; In the formula, is the comprehensive CNR value; is the dynamic weight factor; is the CNR value of the proximal reference lumen of the lesion; is the CNR value of the distal reference lumen of the lesion.

[0011] Preferably, the stenosis ratio feature analysis is performed according to the comprehensive CNR value to obtain the stenosis degree data of the corresponding cross-section, including: obtaining the stenosis ratio according to the comprehensive CNR value and the normal lumen CNR value, dividing the cross-section stenosis degree into a preset number of levels according to the stenosis ratio; and judging the reliability of the stenosis rate formula through dot-plot statistical analysis.

[0012] Preferably, judging the reliability of the stenosis rate formula through dot-plot statistical analysis includes: analyzing the distribution mean, variance and skewness of the CNR values on different segments to evaluate the applicability of the formula; and analyzing the classification ability of the comprehensive CNR value to predict the stenosis rate through the ROC curve, and calculating the AUC value to verify the formula performance.

[0013] Preferably, a coronary artery analysis and processing system based on magnetic resonance CNR value includes: A data acquisition and preprocessing module, configured to collect coronary artery image data, screen valid images for standardization processing; and use multi-slice imaging and multi-angle reconstruction methods to screen and obtain standard coronary artery images; An interest area recognition module, configured to use a convolutional neural network and an active contour model, combined with an edge-preserving normal gradient vector flow, to segment the heart area and coronary blood vessels; A CNR value calculation module, configured to use a dynamically adjusted formula to obtain the CNR value of the associated reference area in the MRCA image; A comprehensive CNR analysis module, configured to obtain a comprehensive CNR value according to the CNR value of the associated reference area in the MRCA image by using a weighted average method; A stenosis degree analysis module, configured to analyze the stenosis ratio feature of the target segment according to the comprehensive CNR value and classify it, and judge the reliability of the stenosis rate formula.

[0014] The beneficial effects of the present invention are: 1. By adopting an image enhancement and region extraction technology that combines an edge-preserving normal gradient vector flow and an active contour model, the present invention improves the recognition accuracy of the region of interest. At the same time, the coronary blood vessels are accurately segmented by an affine coupling regularization network, improving the efficiency and accuracy of target region extraction; 2. By introducing a dynamically adjusted factor and combining the individual characteristics of the patient, such as age, weight and underlying diseases, the present system dynamically optimizes the CNR calculation, improving the personalized adaptation ability of the CNR value calculation and being able to more accurately reflect the coronary artery stenosis characteristics of different patients; 3. By comprehensively calculating the CNR values of the proximal and distal reference areas through a weighted average formula, a comprehensive analysis of the stenosis degree of the coronary artery segment is ensured, the deviation caused by a single CNR value is eliminated, and the stability of the stenosis ratio analysis is improved; 4. The system comprehensively verifies the classification performance and applicability of the stenosis rate formula through dot-plot statistical analysis of the CNR value distribution characteristics and ROC curve analysis to calculate the AUC value, effectively improving the reliability and stability of stenosis rate analysis; 5. As a non-invasive analysis method based on MRCA, it has no radiation risk and is easy to operate, reducing the risks and discomfort of patients during the diagnosis process, and avoiding complications caused by invasive operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flowchart of the coronary artery analysis and processing method based on the magnetic resonance CNR value of the present invention; Figure 2 It is a flowchart of the coronary artery analysis and processing system based on the magnetic resonance CNR value of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0017] Embodiment 1: A coronary artery analysis and processing method based on the magnetic resonance CNR value, as Figure 1 and Figure 2 shown, includes the following steps: S1: Collect coronary artery image data, preprocess the collected coronary artery image data to obtain preprocessing data, and input the preprocessing data into the ROI recognition model to obtain the target ROI; Use the multi-slice imaging and multi-angle reconstruction method based on the criteria of minimizing artifacts and anatomical coverage to screen the collected coronary artery image data to obtain standard coronary artery images, and perform preprocessing of data enhancement and contrast enhancement on the standard coronary artery images to obtain preprocessed coronary artery images, and input the preprocessed coronary artery images into the ROI recognition model to obtain the target ROI.

[0018] It should be noted that through the magnetic resonance coronary angiography (MRCA) device, the coronary artery image data of the patient is collected, and each image is subjected to artifact detection, and the images with artifacts greater than the threshold, such as 20% signal change, are removed; according to the anatomical coverage criteria, the clear imaging data of the coronary artery main trunk area is preferentially selected; additional data is generated by applying translation, rotation, and symmetry transformations to each image; the local blood vessel contrast is improved through contrast stretching and histogram equalization; finally, the preprocessed coronary artery image dataset is obtained.

[0019] The ROI recognition model uses a method that combines the edge-preserving normal gradient vector flow and the active contour model to extract the cardiac region. Specifically, the convolutional neural network model is trained using historical coronary artery image data. The preprocessed coronary artery image is enhanced using the edge-preserving normal gradient vector flow to obtain an enhanced image. Based on the enhanced image and combined with the active contour model, the target contour of the image is determined by minimizing the energy functional, and the tissues of non-interest are removed. Based on the regularization flow structure network of the affine coupling layer, coronary artery segmentation is performed. Using the manually labeled coronary artery data, the network parameters are iteratively trained using the gradient descent optimization algorithm. The centerline of the blood vessel is detected based on the Kalman filter algorithm, and combined with the prior knowledge-driven blood vessel identification re-identification program, the automatic segmentation result is corrected and optimized, and finally the target ROI is obtained.

[0020] It should be noted that for the input preprocessed coronary artery image, the edge-preserving GVF algorithm is used for enhancement processing. The normal gradient vector flow of each pixel is calculated, the strong gradient regions, such as the edges of the blood vessel wall, are retained, and the weak gradient regions are suppressed. The output enhanced image presents clearer coronary artery boundary features. Based on the enhanced image, the following energy functional is minimized using the active contour model. , where is the image intensity energy, is the contour smoothing energy, is the balance coefficient, and the extraction result includes the preliminary contour of the cardiac region and the coronary artery trunk. The coronary artery segmentation is completed using the affine coupling regularization flow network. The network parameters are optimized by gradient descent. During training, the manually labeled blood vessel contour data is used, and the resulting image contains the segmentation mask of the blood vessel region. Based on the Kalman filter algorithm, the position of the blood vessel centerline is detected and false segmentation is eliminated. The prior knowledge-driven re-identification program is used to correct the misidentification of the blood vessel branch label, and finally the target ROI is obtained, including the complete coronary artery region and the centerline information.

[0021] S2: Based on the target ROI, use the MRCA method to obtain the MRCA image of the target ROI, and use the final dynamic adaptability formula to obtain the CNR value of the associated reference region in the MRCA image; Select the normal blood vessel segment closest to the lesion but not affected as the proximal reference point, and calculate the proximal reference lumen CNR value of the lesion based on the MRCA image and the final dynamic adaptability formula. Select the normal blood vessel segment farthest from the lesion but not affected as the distal reference point, and calculate the distal reference lumen CNR value of the lesion based on the MRCA image and the final dynamic adaptability formula.

[0022] It should be noted that, upstream of the lesion area is selected from the MRCA image, and the normal vascular segment closest to the lesion and not affected by the lesion is used as the proximal reference point; downstream of the lesion area is selected from the MRCA image, and the normal vascular segment farthest from the lesion and not affected by the lesion is used as the distal reference point. In one adopted scheme in this embodiment, the distance from the proximal reference point to the diseased vascular segment is equal to the distance from the distal reference point to the diseased vascular segment.

[0023] Obtain the relevant data of the target patient, the relevant data of the proximal reference point, and the relevant data of the distal reference point, and adjust the final dynamic adaptability formula according to the relevant data of the target patient, where the final dynamic adaptability formula is: ; In the formula, is the contrast-to-noise ratio of the cross-section to be measured; is the signal intensity in the lumen of the cross-section of the coronary artery segment to be measured; is the absolute value of the signal intensity of the chest wall muscle; is the standard deviation of the air background noise signal intensity; is the dynamic adjustment factor for adjusting the CNR calculation according to the individual characteristics of the patient.

[0024] The obtaining of the relevant data of the target patient, the relevant data of the proximal reference point, and the relevant data of the distal reference point, and the adjustment of the final dynamic adaptability formula according to the relevant data of the target patient include: The relevant data of the target patient includes the age, weight, and underlying disease data of the target patient. According to the relevant data of the target patient, use the dynamic adjustment formula to obtain the dynamic adjustment factor, where the dynamic adjustment formula is: ; In the formula, is the dynamic adjustment factor; is the weight adjustment coefficient; is the age of the target patient; is the average age of the target population; is the weight of the target patient; is the average weight of the target population; is the binary marker of the th disease of the target patient; is the th disease weight.

[0025] It should be noted that, for example, collect the personalized characteristics of the target patient, age: 60 years old; weight: 75 kg; whether there are underlying diseases: hypertension (yes), diabetes (no); adjust the CNR formula according to the target patient data to obtain the dynamic adjustment factor of the target patient; proximal reference point selection: use the MRCA image to select the proximal reference point of the target ROI and measure the signal intensity and background noise of the relevant area; distal reference point selection: similarly select the distal reference point of the target ROI and measure its signal intensity and background noise.

[0026] S3: Obtain the comprehensive CNR value using the weighted average method according to the CNR value of the associated reference area in the MRCA image; Obtain the comprehensive CNR value according to the CNR value of the proximal reference lumen of the lesion and the CNR value of the distal reference lumen of the lesion using the comprehensive calculation formula, where the comprehensive calculation formula is: ; In the formula, is the comprehensive CNR value; is the dynamic weight factor; is the CNR value of the proximal reference lumen of the lesion; is the CNR value of the distal reference lumen of the lesion.

[0027] It should be noted that use the dynamically adjusted CNR value obtained above: that is, the CNR value of the proximal reference lumen of the lesion and the CNR value of the distal reference lumen of the lesion. According to the distance weight of the reference point in the lesion area, in the formula is equal to ; In this embodiment, the specific process of using the comprehensive calculation formula is: measure the CNR value of the coronary lumen with a diameter greater than or equal to 2 mm for each segment, clarify the position of the chest wall muscle tissue and the standard deviation position of the air background noise signal intensity on the MRCA image data, and set both to 0.5, that is, the comprehensive calculation formula becomes: After that, use the comprehensive calculation formula to obtain the comprehensive CNR value.

[0028] S4: Perform stenosis ratio feature analysis according to the comprehensive CNR value to obtain the stenosis degree data of the corresponding cross-section.

[0029] Obtain the stenosis ratio according to the comprehensive CNR value and the CNR value of the normal lumen, divide the cross-section stenosis degree into a preset number of grades according to the stenosis ratio; and judge the reliability of the stenosis rate formula through point-by-point statistical analysis.

[0030] Analyze the distribution mean, variance and skewness of the CNR values on different segments to evaluate the applicability of the formula; and analyze the classification ability of the comprehensive CNR value to predict the stenosis rate through the ROC curve, and calculate the AUC value to verify the formula performance.

[0031] It should be noted that the normal lumen CNR value is measured by selecting lesion-free areas in the healthy coronary artery segments of the target patient through MRCA images; the comprehensive CNR value is obtained according to the above comprehensive calculation formula; for example, the ratio of the comprehensive CNR value to the normal lumen CNR value is used as the stenosis ratio. According to the preset stenosis degree grading standard, the cross-sectional stenosis degree is divided into the following four grades: Grade 0 (no stenosis): the stenosis ratio is greater than 0.9; Grade 1 (mild stenosis): the stenosis ratio is greater than 0.75 and less than or equal to 0.9; Grade 2 (moderate stenosis): the stenosis ratio is greater than or equal to 0.51 and less than or equal to 0.75; Grade 3 (severe stenosis): the stenosis ratio is greater than or equal to 0.31 and less than or equal to 0.5; Grade 4 (very severe stenosis): the stenosis ratio is greater than or equal to 0.01 and less than or equal to 0.3; Grade 5 (occlusion): the stenosis ratio is equal to 0; The reliability of the stenosis rate formula is analyzed using point plotting statistics and the ROC curve. The point plotting statistics analysis is as follows: Segment division: The coronary artery is divided into three segments: proximal, middle, and distal; Statistical parameter extraction: The mean, variance, and skewness coefficient of the CNR value distribution are calculated separately in each segment; According to the mean and variance, the CNR value distribution in the proximal segment is more concentrated and the data skewness is smaller, so the formula has good applicability; while the skewness in the distal segment increases and the formula's prediction ability is limited; The ROC curve analysis is as follows: Real data collection: By pairing the actual clinical coronary artery stenosis rate grading data with the comprehensive CNR value, classification samples are generated; ROC curve plotting: Using the comprehensive CNR value as the classifier input, the sensitivity and specificity are calculated, and the ROC curve is plotted; AUC value calculation: The AUC value is calculated according to the ROC curve.

[0032] Example 2: On the basis of Example 1, a coronary artery analysis and processing system based on the magnetic resonance CNR value includes: A data acquisition and preprocessing module, which is used to collect coronary artery image data, screen effective images for standardization processing; and use multi-slice imaging and multi-angle reconstruction methods to screen and obtain standard coronary artery images; An interest area recognition module, which is used to use a convolutional neural network and an active contour model, combined with an edge-preserving normal gradient vector flow, to segment the heart area and coronary blood vessels; A CNR value calculation module, which is used to use a dynamically adjusted formula to obtain the CNR value of the associated reference area in the MRCA image; A comprehensive CNR analysis module, which is used to obtain a comprehensive CNR value according to the CNR value of the associated reference area in the MRCA image using a weighted average method; A stenosis degree analysis module, which is used to analyze the stenosis ratio characteristics of the target segment according to the comprehensive CNR value and classify them, and judge the reliability of the stenosis rate formula.

[0033] It should be understood that this embodiment is only used to illustrate the present invention and not to limit the scope of the present invention. In addition, it should be understood that after reading the content taught by the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms also fall within the scope defined by the appended claims of this application.

Claims

1. A coronary artery analysis and processing method based on magnetic resonance CNR value, characterized in that: The following steps are involved: S1: collecting coronary artery image data, preprocessing the collected coronary artery image data to obtain preprocessed data, and inputting the preprocessed data into the ROI recognition model to obtain the target ROI; S2: using the MRCA method based on the target ROI, obtaining an MRCA image of the target ROI, and using a final dynamic adaptive formula to obtain a CNR value of an associated reference area in the MRCA image; S3: obtaining a comprehensive CNR value by using a weighted average method according to the CNR value of the associated reference area in the MRCA image; S4: Perform stenosis ratio characteristic analysis based on the comprehensive CNR value to obtain stenosis degree data of the corresponding cross section.

2. The method for coronary artery analysis and processing based on magnetic resonance CNR value according to claim 1, characterized in that: The collecting of coronary image data, preprocessing the collected coronary image data to obtain a preprocessed coronary image, and inputting the preprocessed coronary image into a ROI recognition model to obtain a target ROI, includes: using multi-slice imaging and multi-angle reconstruction methods to screen the collected coronary image data to obtain a standard coronary image based on artifact minimization and anatomical coverage standards, preprocessing the standard coronary image for data enhancement and contrast enhancement to obtain a preprocessed coronary image, and inputting the preprocessed coronary image into the ROI recognition model to obtain a target ROI.

3. The method for coronary artery analysis and processing based on magnetic resonance CNR value according to claim 2, characterized in that: The method comprises: the ROI recognition model uses a method combining edge-preserving normal gradient vector flow and active contour model to extract the heart area, specifically, using historical coronary image data to train the convolutional neural network model, using edge-preserving normal gradient vector flow to enhance the preprocessed coronary image to obtain an enhanced image, and based on the enhanced image combined with the active contour model, the image target contour is determined by minimizing the energy functional to remove tissues of non-interest; coronary vessel segmentation is performed based on a regularized flow structure network of an affine coupling layer, and the network parameters are iteratively trained using a gradient descent optimization algorithm using manually labeled coronary vessel data; the vessel centerline is detected based on the Kalman filter algorithm, and the automatic segmentation result is corrected and optimized in combination with a vessel marker re-identification program driven by prior knowledge, so as to finally obtain the target ROI.

4. The method for coronary artery analysis and processing based on magnetic resonance CNR value according to claim 1, characterized in that: The method uses the MRCA method based on the target ROI to obtain the MRCA image of the target ROI, and uses the final dynamic adaptability formula to obtain the CNR value of the associated reference area in the MRCA image, including: selecting a normal blood vessel segment that is closest to the lesion but not affected as a proximal reference point, calculating the proximal reference point based on the MRCA image and the final dynamic adaptability formula, and obtaining the CNR value of the proximal reference lumen of the lesion; selecting a normal blood vessel segment that is farthest from the lesion but not affected as a distal reference point, calculating the distal reference point based on the MRCA image and the final dynamic adaptability formula, and obtaining the CNR value of the distal reference lumen of the lesion.

5. The method for coronary artery analysis and processing based on magnetic resonance CNR value according to claim 4, characterized in that: The method of using the MRCA method based on the target ROI to obtain the MRCA image of the target ROI, and using the final dynamic adaptability formula to obtain the CNR value of the associated reference area in the MRCA image, includes: obtaining relevant data of the target patient, relevant data of the proximal reference point, and relevant data of the distal reference point, and adjusting the final dynamic adaptability formula according to the relevant data of the target patient, wherein the final dynamic adaptability formula is: ； In the formula, is the ratio of the cross section to be measured to noise; The signal intensity within the lumen of the coronary artery segment cross section should be measured; is the absolute value of the chest wall muscle signal intensity; is the standard deviation of the air background noise signal intensity; Dynamic adjustment factors for adjusting CNR calculations according to individual patient characteristics.

6. The method for coronary artery analysis and processing based on magnetic resonance CNR value according to claim 5, characterized in that: The step of obtaining relevant data of the target patient, relevant data of the proximal reference point, and relevant data of the distal reference point, and adjusting the final dynamic adaptability formula according to the relevant data of the target patient includes: the relevant data of the target patient includes the age, weight, and underlying disease data of the target patient, and using a dynamic adjustment formula according to the relevant data of the target patient to obtain a dynamic adjustment factor, wherein the dynamic adjustment formula is: ； In the formula, is the dynamic adjustment factor; is the weight adjustment coefficient; is the age of the target patient; is the average age of the target population; is the target patient's weight; is the average weight of the target population; For target patients A binary marker for a disease; For the The weight of the disease.

7. The method for coronary artery analysis and processing based on magnetic resonance CNR value according to claim 4, characterized in that: The method of obtaining a comprehensive CNR value by using a weighted average method according to the CNR value of the associated reference area in the MRCA image includes: obtaining a comprehensive CNR value by using a comprehensive calculation formula according to the CNR value of the proximal reference lumen of the lesion and the CNR value of the distal reference lumen of the lesion, wherein the comprehensive calculation formula is: ; In the formula, is the comprehensive CNR value; is the dynamic weight factor; is the CNR value of the reference lumen proximal to the lesion; It is the CNR value of the reference lumen distal to the lesion.

8. The method for coronary artery analysis and processing based on magnetic resonance CNR value according to claim 1, characterized in that: The method of performing stenosis ratio characteristic analysis based on the comprehensive CNR value to obtain stenosis degree data of the corresponding cross section includes: obtaining a stenosis ratio based on the comprehensive CNR value and the CNR value of the normal lumen, and dividing the stenosis degree of the cross section into a preset number of levels based on the stenosis ratio; and determining the reliability of the stenosis rate formula through point-plotting statistical analysis.

9. The method for coronary artery analysis and processing based on magnetic resonance CNR value according to claim 8, characterized in that: The reliability of the stenosis rate formula is determined by the point-plotting statistical analysis, including: analyzing the mean, variance and skewness of the CNR value distribution on different segments to evaluate the applicability of the formula; and analyzing the classification ability of the comprehensive CNR value to predict the stenosis rate through the ROC curve, and calculating the AUC value to verify the performance of the formula.

10. A coronary artery analysis and processing system based on magnetic resonance CNR value, according to a coronary artery analysis and processing method based on magnetic resonance CNR value according to any one of claims 1 to 9, characterized in that: include: Data acquisition and preprocessing module, used to collect coronary image data and screen valid images for standardized processing; Multi-section imaging and multi-angle reconstruction methods were used to screen and obtain standard coronary artery images; Region of Interest Recognition module for segmentation of cardiac regions and coronary vessels using convolutional neural networks and active contour models combined with edge-preserving normal gradient vector flow; A CNR value calculation module, used for obtaining the CNR value of the associated reference area in the MRCA image using a dynamic adjustment formula; A comprehensive CNR analysis module, used for obtaining a comprehensive CNR value by using a weighted average method according to the CNR value of the associated reference area in the MRCA image; The stenosis degree analysis module is used to analyze and classify the stenosis ratio characteristics of the target segment according to the comprehensive CNR value, and to determine the reliability of the stenosis rate formula.

Images