Cardiac abnormality analysis method and system based on ultrasound imaging

Abstract

The present invention provides a cardiac abnormality analysis method and system based on ultrasonic imaging, and relates to the technical field of ultrasonic image analysis. The present invention uses multi-angle scanning and image registration technology, combined with information compensation and image optimization processing, to significantly improve the comprehensiveness and clarity of cardiac structure and functional status, enhance the accuracy of feature extraction, and thus effectively improve the accuracy and detection sensitivity of cardiac image analysis. Moreover, by comparing and analyzing the main image and the compensated optimized cardiac image detection results, the reliability of abnormality identification is further verified, the credibility of the diagnostic conclusion is enhanced, and more comprehensive and detailed diagnostic information is provided to doctors, thereby assisting them in making more accurate clinical decisions.

Description

Technical Field

[0001] The present invention relates to the technical field of ultrasonic image analysis, and in particular to a cardiac abnormality analysis method and system based on ultrasonic images. Background Art

[0002] In the field of clinical medicine, early diagnosis and precise treatment of heart disease are crucial to the patient's prognosis. However, traditional cardiac imaging diagnostic methods often rely on scanning images from a single angle or a single phase, which to a certain extent limits doctors' comprehensive assessment of the heart's structure and functional status. Specifically, a single-angle scan may not be able to capture abnormal changes in certain hidden areas of the heart, while a single-phase image can hardly fully reflect the dynamic changes of the heart during the cardiac cycle;

[0003] In addition, cardiac images are often interfered by a variety of factors during the acquisition process, such as changes in patient position, respiratory movement, equipment noise, etc. These factors may lead to a decline in image quality and affect subsequent feature extraction and abnormality detection. Although modern medical imaging technology has made significant progress, in the diagnosis of complex heart diseases, how to further improve image quality and enhance the accuracy of feature extraction is still a technical problem that needs to be solved urgently;

[0004] On the other hand, there are many types of heart diseases, and different diseases may have overlapping or similar imaging manifestations, which poses a huge challenge to doctors' diagnosis. Traditional diagnostic methods often rely on doctors' experience and subjective judgment, lack objective and quantitative evaluation criteria, and are prone to misdiagnosis or missed diagnosis.

[0005] Therefore, it is necessary to provide a cardiac abnormality analysis method and system based on ultrasound imaging to solve the above technical problems. Summary of the invention

[0006] To solve the above technical problems, the present invention provides a cardiac abnormality analysis method and system based on ultrasound imaging, which achieves the beneficial effects of improving the accuracy and detection sensitivity of cardiac image analysis, verifying the reliability of abnormality identification, and enhancing the credibility of diagnostic conclusions.

[0007] The present invention provides a method for analyzing cardiac abnormalities based on ultrasound images, and the method comprises the following steps:

[0008] S1. Scanning a heart region to obtain a heart image as a main image, and scanning the same heart region from different angles to obtain at least one heart image as a secondary image;

[0009] S2, performing image registration on the secondary image and the primary image, and based on the image registration result, performing information compensation on the primary image to obtain a compensated cardiac image;

[0010] S3, performing image optimization processing on the compensated cardiac image, specifically including denoising, sharpening and contrast enhancement, to obtain a compensated optimized cardiac image;

[0011] S4, extracting features from the main image and the compensated optimized cardiac image respectively, specifically including the geometric shape of the cardiac structure and blood flow information, to obtain main image features and compensated optimized cardiac image features;

[0012] S5. Based on the features extracted from the main image and the compensated optimized cardiac image, use a preset abnormal analysis model to detect and identify the main image and the compensated optimized cardiac image respectively, and output the detection results of the main image and the compensated optimized cardiac image;

[0013] S6. Compare the detection results of the main image and the compensated optimized heart image to obtain a heart abnormality comparison result.

[0014] Preferably, the step S2 specifically includes the following steps:

[0015] S2.1, preprocessing the secondary image, specifically including removing artifacts and correcting geometric distortion;

[0016] S2.2, using an image registration algorithm to calibrate and align the secondary image with the primary image to obtain a registration result;

[0017] S2.3. According to the registration result, the missing or blurred information in the secondary image is identified, and the information of the primary image is compensated by using compensation and fusion technology to obtain a compensated cardiac image.

[0018] Preferably, the step S4 specifically includes the following steps:

[0019] S4.1. Extract the heart contour using image segmentation technology and mark the heart contour structure;

[0020] S4.2, extracting geometric shape features based on the marked heart contour structure, including ventricle size and atrial volume;

[0021] S4.3. Use Doppler ultrasound technology to analyze and identify hemodynamic characteristics, including blood flow velocity, direction and eddy flow hemodynamic characteristics.

[0022] Preferably, the abnormal analysis model training process preset in step S5 includes:

[0023] Collect historical cardiac image data, specifically including cardiac images of normal hearts and known abnormal states;

[0024] Preprocessing the historical cardiac image data, specifically including cleaning the data and standardizing the image, to obtain cleaned cardiac image data;

[0025] A convolutional neural network was selected as the model architecture, and the cleaned cardiac image data was used as a training set to input into the convolutional neural network model architecture to obtain an abnormality analysis model through training.

[0026] Preferably, the step S5 specifically includes the following steps:

[0027] S5.1. Input the main image features and the compensated optimized cardiac image features into the abnormality analysis model for detection and recognition, and output the main image detection result and the compensated optimized cardiac image detection result;

[0028] S5.2. Record and store the main image detection results and the compensated optimized heart image detection results.

[0029] Preferably, the step S6 specifically includes the following steps:

[0030] S6.1. displaying the main image detection result and the compensated optimized heart image detection result side by side;

[0031] S6.2, comparing the consistency of the main image detection result and the compensated optimized cardiac image detection result in abnormality detection to obtain a comparison result;

[0032] S6.3. Generate a cardiac abnormality comparison report based on the comparison analysis results.

[0033] A cardiac abnormality analysis system based on ultrasound images, the analysis system comprising:

[0034] A scanning acquisition module, used for scanning the heart area to obtain a heart image as a main image, and scanning the same area of ​​the heart from different angles to obtain at least one heart image as a secondary image;

[0035] A registration and compensation module, used to perform image registration on the secondary image and the primary image, and based on the image registration result, perform information compensation on the primary image to obtain a compensated cardiac image;

[0036] An optimization processing module is used to perform image optimization processing on the compensated cardiac image, specifically including denoising, sharpening and contrast enhancement, to obtain a compensated optimized cardiac image;

[0037] A feature extraction module is used to extract features from the main image and the compensated optimized cardiac image respectively, specifically including the geometric shape of the cardiac structure and blood flow information, to obtain the main image features and the compensated optimized cardiac image features;

[0038] A detection and recognition module, which is used to detect and recognize the main image and the compensated optimized cardiac image respectively based on the features extracted from the main image and the compensated optimized cardiac image using a preset abnormal analysis model, and output the detection results of the main image and the compensated optimized cardiac image;

[0039] The comparison module is used to compare the detection results of the main image and the compensated optimized cardiac image to obtain a cardiac abnormality comparison result.

[0040] Compared with the related art, the cardiac abnormality analysis method and system based on ultrasound imaging provided by the present invention has the following beneficial effects:

[0041] The present invention uses multi-angle scanning and image registration technology, combined with information compensation and image optimization processing, to significantly improve the comprehensiveness and clarity of cardiac structure and functional status, enhance the accuracy of feature extraction, and thereby effectively improve the accuracy and detection sensitivity of cardiac image analysis. Furthermore, by comparing and analyzing the main image and the compensated optimized cardiac image detection results, the reliability of abnormality identification is further verified, the credibility of the diagnostic conclusion is enhanced, and doctors are provided with more comprehensive and detailed diagnostic information, thereby assisting them in making more accurate clinical decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 A flowchart of the method for analyzing cardiac abnormalities based on ultrasound images provided by the present invention;

[0043] Figure 2 This is a system block diagram of the ultrasound imaging-based cardiac abnormality analysis system provided by the present invention. DETAILED DESCRIPTION

[0044] The present invention will be further described below in conjunction with the accompanying drawings and implementation modes.

[0045] Embodiment 1

[0046] like Figure 1 As shown, this embodiment provides a cardiac abnormality analysis method based on ultrasound imaging, comprising the following steps:

[0047] S1. Scan the heart region to obtain a heart image as a main image, and scan the same heart region from different angles to obtain at least one heart image as a secondary image.

[0048] Specifically, the patient's heart area is scanned initially by ultrasound imaging equipment, and the heart image obtained by the initial scan is used as the main image. The same heart area of ​​the initial scan is scanned multiple times from different angles, specifically from the side and oblique angles, to obtain multiple heart images at different angles, and the multiple heart images at different angles are used as secondary images. The main image is used as a reference to provide a basis for subsequent image processing and feature extraction; the secondary image is used as a supplement to help enhance the understanding and analysis of the complex structure of the heart.

[0049] In this embodiment, the patient is suspected of having ventricular wall thickening. The patient's heart area is initially scanned by ultrasonic imaging equipment to obtain a main image. The main image clearly shows the general shape of the ventricle from the front, but the details are not obvious enough. Subsequently, multiple secondary images are obtained by scanning from the left and right sides.

[0050] S2. Performing image registration on the secondary image and the primary image, and performing information compensation on the primary image based on the image registration result to obtain a compensated cardiac image.

[0051] Specifically, the secondary image is spatially aligned with the main image, and based on the registration result, the information in the secondary image is used to compensate for the missing or blurred parts of the main image by using compensation and fusion technology to obtain a compensated cardiac image. Image registration and information compensation can integrate complementary information from different views, improve the overall quality and information content of the image, and lay a more solid foundation for subsequent feature extraction and anomaly detection.

[0052] In this embodiment, due to angle limitation in the main image, the thickened portion of the ventricular wall is not displayed clearly. Through registration and information compensation, the clear thickened area information in the secondary image is used to enhance the display effect of the corresponding area in the main image, making the entire heart image more complete and accurate, and obtaining a compensated heart image.

[0053] S3. Performing image optimization processing on the compensated cardiac image, specifically including denoising, sharpening and contrast enhancement, to obtain a compensated optimized cardiac image.

[0054] Specifically, the compensated cardiac image is denoised to eliminate noise points or blurred areas in the image; sharpening is performed to enhance the edges and details of the image; finally, the contrast is adjusted to make different structures in the image more distinct and easy to distinguish, which can significantly improve the image quality and make the cardiac structure clearer, facilitating subsequent automatic or manual feature extraction and anomaly detection.

[0055] In this embodiment, after image optimization processing, in the compensated heart image, the thickened area that was originally slightly blurred due to noise interference becomes clearer, the edge becomes sharper, and the contrast is enhanced, thereby obtaining a compensated and optimized heart image.

[0056] S4. Extract features from the main image and the compensated optimized heart image respectively, including the geometric shape of the heart structure and blood flow information, to obtain main image features and compensated optimized heart image features.

[0057] Specifically, the Sobel operator in the edge detection algorithm is used to identify the boundary lines in the main image and compensate and optimize the heart image. The image segmentation technology is used to extract the heart contour and mark the heart contour structure. Then, the geometric shape features are extracted from the heart contour structure, including the size of the ventricle and the volume of the atrial chamber. The Doppler imaging technology is used to extract blood flow information features, including blood flow velocity, direction and eddy flow hemodynamic characteristics.

[0058] In this embodiment, the contours of the ventricular wall and the ventricular septum are identified, and the compensated optimized heart image is segmented and marked into contour structures such as the left ventricle, right ventricle and atrium using image segmentation technology, and the geometric shape features such as the size and volume of each contour structure are extracted. At the same time, color Doppler imaging technology is used to extract the blood flow velocity characteristics of the left ventricle.

[0059] S5. Based on the features extracted from the main image and the compensated optimized heart image, a pre-set abnormal analysis model is used to detect and identify the main image and the compensated optimized heart image respectively, and the detection results of the main image and the compensated optimized heart image are output.

[0060] In this embodiment, the abnormal analysis model detects and identifies features in the main image and the compensated optimized heart image, and finds that the left ventricle in the main image is of normal size but has a slightly increased wall thickness, while the compensated optimized heart image shows a slowed blood flow velocity during the left ventricle ejection phase.

[0061] S6. Compare the detection results of the main image and the compensated optimized heart image to obtain a heart abnormality comparison result.

[0062] Specifically, the detection results of the main image and the compensated optimized heart image are compared and analyzed to comprehensively evaluate whether there are abnormalities in the state and function of the heart; based on the comparison results, a final heart abnormality comparison result report is output for the doctor's reference and decision-making. By comparing the detection results of different images, a more comprehensive understanding of the state and functional changes of the heart can be obtained.

[0063] In this embodiment, the final report pointed out that the patient's heart had abnormal conditions of myocardial hypertrophy and heart failure, and further examination and treatment were recommended.

[0064] In the specific implementation process, the step S2 specifically includes the following steps:

[0065] S2.1. Preprocessing the secondary image, specifically including removing artifacts and correcting geometric distortion.

[0066] Specifically, artifacts are removed from the secondary image through filtering, non-cardiac structure interference in the secondary image caused by equipment, patients or environmental factors is eliminated, and the spatial position relationship in the secondary image is adjusted to make it consistent with the actual heart structure.

[0067] S2.2. Use an image registration algorithm to align the secondary image with the primary image to obtain a registration result.

[0068] Specifically, a feature-based registration algorithm is selected to first extract common features in the two images, then calculate the correspondence between these features, and apply a projective transformation model to align the secondary image to the primary image.

[0069] S2.3. According to the registration result, the missing or blurred information in the secondary image is identified, and the information of the primary image is compensated by using compensation and fusion technology to obtain a compensated cardiac image.

[0070] In the specific implementation process, the step S4 specifically includes the following steps:

[0071] S4.1. Use image segmentation technology to extract the heart contour and mark the heart contour structure.

[0072] Specifically, the Sobel operator in the edge detection algorithm is used to identify the boundary lines in the main image and compensate and optimize the heart image, and the threshold segmentation in the image segmentation technology is used to extract the heart contour. At the same time, the noise, breaks or redundant parts of the extracted heart contour are eliminated, including smoothing the contour edges, filling small holes, removing outliers and other operations to obtain a more accurate and complete heart contour.

[0073] S4.2. Based on the marked heart contour structure, extract geometric shape features, including ventricle size and atrial volume.

[0074] Specifically, based on marking the heart contour structure, the pixel counting and distance measurement in the image measurement tool are used to calculate the cross-sectional area, major axis and minor axis length and other parameters of the ventricle; the volume of the atrium is calculated by measuring the area of ​​the atrium on multiple sections and combining it with a polyhedron model.

[0075] S4.3. Use Doppler ultrasound technology to analyze and identify hemodynamic characteristics, including blood flow velocity, direction and eddy flow hemodynamic characteristics.

[0076] Specifically, Doppler ultrasound imaging technology is used to obtain color-coded images of blood flow in the heart, and by measuring the blood flow velocity in different areas of the color-coded image, the ejection velocity and filling velocity of each chamber of the heart are evaluated. Secondly, analyzing the direction of blood flow can reveal whether there are abnormal phenomena such as reflux and vortex in the heart.

[0077] In the specific implementation process, the abnormal analysis model training process preset in step S5 includes:

[0078] Collect historical cardiac image data, specifically including cardiac images of normal hearts and known abnormal states;

[0079] Preprocessing the historical cardiac image data, specifically including cleaning the data and standardizing the image, to obtain cleaned cardiac image data;

[0080] A convolutional neural network was selected as the model architecture, and the cleaned cardiac image data was used as a training set to input into the convolutional neural network model architecture to obtain an abnormality analysis model through training.

[0081] In the specific implementation process, the step S5 specifically includes the following steps:

[0082] S5.1. Input the main image features and the compensated optimized cardiac image features into the abnormality analysis model for detection and recognition, and output the main image detection result and the compensated optimized cardiac image detection result;

[0083] S5.2. Record and store the main image detection results and the compensated optimized heart image detection results.

[0084] In the specific implementation process, the step S6 specifically includes the following steps:

[0085] S6.1. Display the main image detection result and the compensated optimized heart image detection result side by side.

[0086] Specifically, the main image detection results and the compensated optimized heart image detection results are displayed through a report generation tool, and the two detection results contain evaluation information on the heart structure, function, blood flow status, etc.

[0087] S6.2, comparing the consistency of the main image detection result and the compensated optimized cardiac image detection result in abnormality detection to obtain a comparison result;

[0088] S6.3. Generate a cardiac abnormality comparison report based on the comparison analysis results.

[0089] Specifically, the consistency and difference between the main image detection results and the compensated optimized cardiac image detection results in abnormality detection are listed in detail, including the comparison of the location, size, shape and other features of the abnormal area, as well as quantitative indicators of the similarity or difference degree, so as to generate a cardiac abnormality comparison report.

[0090] Embodiment 2

[0091] like Figure 2 As shown, this embodiment provides a cardiac abnormality analysis system based on ultrasound images applied to a cardiac abnormality analysis method based on ultrasound images, comprising:

[0092] The scanning acquisition module is used to scan the heart area to obtain a heart image as a main image, and scan the same area of ​​the heart from different angles to obtain at least one heart image as a secondary image.

[0093] Specifically, the scanning acquisition module is used to acquire the cardiac image obtained by the ultrasound imaging device during the initial scan of the patient's cardiac area, and use the cardiac image obtained by the initial scan as the main image, and perform multiple scans on the same cardiac area of ​​the initial scan from different angles, specifically from the side and oblique angles, to obtain multiple cardiac images at different angles, and use the multiple cardiac images at different angles as secondary images. The primary image is used as a reference to provide a basis for subsequent image processing and feature extraction; the secondary image is used as a supplement to help enhance the understanding and analysis of the complex structure of the heart.

[0094] The registration and compensation module is used to perform image registration on the secondary image and the primary image, and based on the image registration result, perform information compensation on the primary image to obtain a compensated cardiac image.

[0095] Specifically, the registration and compensation module is used to align the secondary image with the main image in space. Based on the registration result, the information in the secondary image is used to compensate for the missing or blurred parts of the main image by using compensation and fusion technology to obtain a compensated cardiac image. Image registration and information compensation can integrate complementary information from different views, improve the overall quality and information content of the image, and lay a more solid foundation for subsequent feature extraction and anomaly detection.

[0096] The optimization processing module is used to perform image optimization processing on the compensated cardiac image, specifically including denoising, sharpening and contrast enhancement, to obtain a compensated optimized cardiac image.

[0097] Specifically, the optimization processing module is used to perform denoising on the compensated cardiac image to eliminate noise points or blurred areas in the image; perform sharpening to enhance the edges and details of the image; finally, by adjusting the contrast, different structures in the image are made more distinct and easy to distinguish, which can significantly improve the image quality and make the cardiac structure clearer, facilitating subsequent automatic or manual feature extraction and anomaly detection.

[0098] The feature extraction module is used to extract features from the main image and the compensated optimized cardiac image respectively, specifically including the geometric shape of the cardiac structure and blood flow information, to obtain the main image features and the compensated optimized cardiac image features.

[0099] Specifically, the feature extraction module is used to identify the main image and compensate and optimize the boundary lines in the heart image through the Sobel operator in the edge detection algorithm, and the image segmentation technology is used to extract the heart contour and mark the heart contour structure. Then, the geometric shape features are extracted from the heart contour structure, including the size of the ventricle and the volume of the atrial; the blood flow information features are extracted through Doppler imaging technology, including the blood flow velocity, direction and eddy flow hemodynamic characteristics.

[0100] The detection and recognition module is used to detect and recognize the main image and the compensated optimized cardiac image respectively based on the features extracted from the main image and the compensated optimized cardiac image using a preset abnormal analysis model, and output the detection results of the main image and the compensated optimized cardiac image.

[0101] The comparison module is used to compare the detection results of the main image and the compensated optimized cardiac image to obtain a cardiac abnormality comparison result.

[0102] Specifically, the comparison module is used to compare and analyze the detection results of the main image and the compensated optimized heart image, and comprehensively evaluate whether there are abnormalities in the state and function of the heart; based on the comparison results, the final heart abnormality comparison result report is output for the doctor's reference and decision-making. By comparing the detection results of different images, a more comprehensive understanding of the state and functional changes of the heart can be obtained.

[0103] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0104] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0105] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

Claims

1. A method for analyzing cardiac abnormalities based on ultrasound images, characterized in that: The analytical method comprises the following steps: S1. Scanning a heart region to obtain a heart image as a main image, and scanning the same heart region from different angles to obtain at least one heart image as a secondary image; S2, performing image registration on the secondary image and the primary image, and based on the image registration result, performing information compensation on the primary image to obtain a compensated cardiac image; Specifically, the secondary image is spatially aligned with the primary image, and based on the registration result, the information in the secondary image is used to compensate for the missing or blurred parts of the primary image by using compensation and fusion technology, so as to obtain a compensated cardiac image; S3, performing image optimization processing on the compensated cardiac image, specifically including denoising, sharpening and contrast enhancement, to obtain a compensated optimized cardiac image; S4, extracting features from the main image and the compensated optimized cardiac image respectively, specifically including the geometric shape of the cardiac structure and blood flow information, to obtain main image features and compensated optimized cardiac image features; Specifically, the Sobel operator in the edge detection algorithm is used to identify the boundary lines in the main image and compensate and optimize the heart image. The image segmentation technology is used to extract the heart contour and mark the heart contour structure. The geometric shape features are extracted from the heart contour structure, including the size of the ventricle and the volume of the atrial chamber. The Doppler imaging technology is used to extract the blood flow information features, including the blood flow velocity, direction and eddy flow hemodynamic characteristics. S5. Based on the features extracted from the main image and the compensated optimized cardiac image, use a preset abnormal analysis model to detect and identify the main image and the compensated optimized cardiac image respectively, and output the detection results of the main image and the compensated optimized cardiac image; S6. Compare the detection results of the main image and the compensated optimized heart image to obtain a heart abnormality comparison result.

2. The method for analyzing cardiac abnormalities based on ultrasound images according to claim 1, characterized in that: The step S2 specifically includes the following steps: S2.1, preprocessing the secondary image, specifically including removing artifacts and correcting geometric distortion; S2.2, using an image registration algorithm to calibrate and align the secondary image with the primary image to obtain a registration result; S2.

3. According to the registration result, the missing or blurred parts in the main image are compensated by compensation and fusion technology to obtain a compensated cardiac image.

3. The method for analyzing cardiac abnormalities based on ultrasound images according to claim 1, characterized in that: The step S4 specifically comprises the following steps: S4.

1. Extract the heart contour using image segmentation technology and mark the heart contour structure; S4.2, extracting geometric shape features based on the marked heart contour structure, including ventricle size and atrial volume; S4.

3. Use Doppler ultrasound technology to analyze and identify hemodynamic characteristics, including blood flow velocity, direction and eddy flow hemodynamic characteristics.

4. The method for analyzing cardiac abnormalities based on ultrasound images according to claim 1, characterized in that: The abnormal analysis model training process preset in step S5 includes: Collect historical cardiac image data, specifically including cardiac images of normal hearts and known abnormal states; Preprocessing the historical cardiac image data, specifically including cleaning the data and standardizing the image, to obtain cleaned cardiac image data; A convolutional neural network was selected as the model architecture, and the cleaned cardiac image data was used as a training set to input into the convolutional neural network model architecture to obtain an abnormality analysis model through training.

5. The method for analyzing cardiac abnormalities based on ultrasound images according to claim 1, characterized in that: The step S5 specifically comprises the following steps: S5.

1. Input the main image features and the compensated optimized cardiac image features into the abnormality analysis model for detection and recognition, and output the main image detection result and the compensated optimized cardiac image detection result; S5.

2. Record and store the main image detection results and the compensated optimized heart image detection results.

6. The method for analyzing cardiac abnormalities based on ultrasound images according to claim 1, characterized in that: The step S6 specifically comprises the following steps: S6.

1. displaying the main image detection result and the compensated optimized heart image detection result side by side; S6.2, comparing the consistency of the main image detection result and the compensated optimized cardiac image detection result in abnormality detection to obtain a comparison result; S6.

3. Generate a cardiac abnormality comparison report based on the comparison analysis results.

7. A cardiac abnormality analysis system based on ultrasound imaging, applied to a cardiac abnormality analysis method based on ultrasound imaging as claimed in any one of claims 1 to 6, characterized in that: The analysis system comprises: A scanning acquisition module, used for scanning the heart area to obtain a heart image as a main image, and scanning the same area of ​​the heart from different angles to obtain at least one heart image as a secondary image; A registration and compensation module, used to perform image registration on the secondary image and the primary image, and based on the image registration result, perform information compensation on the primary image to obtain a compensated cardiac image; An optimization processing module is used to perform image optimization processing on the compensated cardiac image, specifically including denoising, sharpening and contrast enhancement, to obtain a compensated optimized cardiac image; A feature extraction module is used to extract features from the main image and the compensated optimized cardiac image respectively, specifically including the geometric shape of the cardiac structure and blood flow information, to obtain the main image features and the compensated optimized cardiac image features; A detection and recognition module, which is used to detect and recognize the main image and the compensated optimized cardiac image respectively based on the features extracted from the main image and the compensated optimized cardiac image using a preset abnormal analysis model, and output the detection results of the main image and the compensated optimized cardiac image; The comparison module is used to compare the detection results of the main image and the compensated optimized cardiac image to obtain a cardiac abnormality comparison result.

Images