Cardiac ultrasonic image shooting suggestion generation method and system, medium and equipment

Abstract

The invention belongs to the technical field of cardiac ultrasonic image processing, and provides a cardiac ultrasonic image shooting suggestion generation method and system, a medium and equipment, and the technical scheme is as follows: extracting multi-scale spatial features of a cardiac ultrasonic image, and fusing the multi-scale spatial features to obtain multi-scale spatial fusion features; performing space attention enhancement and channel attention enhancement on the multi-scale space fusion features to obtain space attention enhanced and channel attention enhanced multi-scale space fusion features; calculating heart contour definition and echo uniformity based on the multi-scale space fusion features of space attention enhancement, and calculating heart valve definition based on the multi-scale space fusion features of channel attention enhancement; extracting features of the cardiac ultrasound image at different time points, and calculating the motion consistency of the left ventricular wall based on the features of the cardiac ultrasound image at different time points; image quality evaluation is carried out, shooting suggestions are generated according to evaluation results, and the extraction accuracy of important medical information is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of cardiac ultrasound image processing, and particularly relates to a method, a system, a medium and a device for generating cardiac ultrasound image shooting suggestions. Background Art

[0002] The statements in this part only provide background technical information related to the present invention, and do not necessarily constitute prior art.

[0003] Cardiac ultrasound images are widely used in the diagnosis and treatment of cardiovascular diseases. Through ultrasound images, the morphology of the heart cavity, the movement of the ventricular wall, the valve function, etc. can be observed. However, the current ultrasound image quality assessment method mainly relies on the doctor's experience for subjective judgment, which is easily affected by personal level and subjective factors, resulting in unstable image quality.

[0004] Cardiac shooting suggestions are crucial for obtaining key features of cardiac ultrasound images. The existing assessment methods lack automated analysis of cardiac ultrasound images and generation of optimized shooting suggestions, which may lead to omission of important medical information. Summary of the Invention

[0005] In order to solve at least one of the technical problems in the above background art, the present invention provides a method, a system, a medium and a device for generating cardiac ultrasound image shooting suggestions, which automatically evaluate the quality of cardiac ultrasound images and provide real-time shooting optimization suggestions to improve the extraction accuracy of important medical information.

[0006] To achieve the above object, the present invention adopts the following technical solutions: The first aspect of the present invention provides a method for generating cardiac ultrasound image shooting suggestions, including the following steps: Obtain cardiac ultrasound image data; Extract multi-scale spatial features of the cardiac ultrasound image, and obtain multi-scale spatial fusion features after fusing the multi-scale spatial features; Perform spatial attention enhancement and channel attention enhancement on the multi-scale spatial fusion features to obtain multi-scale spatial fusion features with spatial attention enhancement and channel attention enhancement; Calculate the clarity of the cardiac contour and the echo uniformity based on the multi-scale spatial fusion features with spatial attention enhancement, and calculate the clarity of the cardiac valve based on the multi-scale spatial fusion features with channel attention enhancement; Extract features at different time points of the cardiac ultrasound image, and calculate the left ventricular wall motion consistency based on the features at different time points of the cardiac ultrasound image; Extract the signal-to-noise ratio of the cardiac ultrasound image; The image quality is evaluated based on the clarity of the heart contour, the uniformity of the echo, the clarity of the heart valves, the consistency of the left ventricular wall movement, and the signal-to-noise ratio, and shooting suggestions are generated according to the evaluation results.

[0007] Further, the extraction of the multi-scale spatial features of the cardiac ultrasound image and the fusion of the multi-scale spatial features to obtain the multi-scale spatial fusion features include: Use a multi-scale convolutional neural network to extract the spatial features of the cardiac ultrasound image. By using convolutional kernels of various different sizes, feature maps extracted by different convolutional kernels are obtained, and then the feature maps of each scale are fused to obtain the fused feature map.

[0008] Further, the calculation formulas for enhancing the spatial attention and channel attention of the multi-scale spatial fusion features are: , , Among them, represents the importance degree of the pixels at each position, represents the multi-scale spatial fusion features, represents the weight assignment of the pixels at different positions when calculating the importance, represents the convolution operation. Through the convolution of and the weight matrix , the correlation information between different positions in the image is captured, and are bias terms used to adjust the offset of the calculation result; represents the importance degree of the pixels corresponding to each channel, is the weight vector of the channel attention, represents the dot product operation, is the activation function.

[0009] Further, the Canny edge detection method is used to detect the multi-scale spatial fusion features with enhanced spatial attention to identify the boundaries of the cardiac chambers, and the clarity of the heart contour is obtained.

[0010] Further, the multi-scale spatial fusion features with enhanced spatial attention are divided into multiple regions, the standard deviation and the average value of the pixel intensities in each region are calculated respectively, and the echo uniformity is obtained based on the standard deviation and the average value of the pixel intensities in each region.

[0011] Further, the method also includes maximizing the similarity of each feature of the similar samples through contrastive learning to obtain the optimized clarity of the heart contour, the echo uniformity, the clarity of the heart valves, the consistency of the left ventricular wall movement, and the signal-to-noise ratio.

[0012] Further, the loss function for maximizing the feature similarity of similar samples is as follows: , where and represent different sample indices, represents the sample and the sample is the inner product of the sums of the feature vectors, and are the norms of the sums of the feature vectors, respectively.

[0013] The second aspect of the present invention provides a cardiac ultrasound image shooting advice generation system, including: A data acquisition module for acquiring cardiac ultrasound image data; A multi-scale feature extraction module for extracting multi-scale spatial features of cardiac ultrasound images and obtaining multi-scale spatial fusion features after fusing the multi-scale spatial features; A multi-scale feature enhancement module for performing spatial attention enhancement and channel attention enhancement on the multi-scale spatial fusion features to obtain multi-scale spatial fusion features with spatial attention enhancement and channel attention enhancement; A feature extraction module for calculating the clarity of the cardiac contour and the echo uniformity based on the multi-scale spatial fusion features with spatial attention enhancement, calculating the clarity of the cardiac valve based on the multi-scale spatial fusion features with channel attention enhancement; extracting features at different time points of the cardiac ultrasound image, calculating the consistency of the left ventricular wall motion based on the features at different time points of the cardiac ultrasound image; extracting the signal-to-noise ratio of the cardiac ultrasound image; A shooting advice generation module for performing image quality evaluation based on the clarity of the cardiac contour, the echo uniformity, the clarity of the cardiac valve, the consistency of the left ventricular wall motion, and the signal-to-noise ratio, and generating shooting advice according to the evaluation results.

[0014] The third aspect of the present invention provides a computer-readable storage medium.

[0015] A computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the steps in the cardiac ultrasound image shooting advice generation method as described above.

[0016] The fourth aspect of the present invention provides a computer device.

[0017] A computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the program, it implements the steps in the cardiac ultrasound image shooting advice generation method as described above.

[0018] Compared with the prior art, the beneficial effects of the present invention are: According to the set image quality evaluation criteria, the present invention extracts multiple key cardiac features, evaluates the quality of cardiac ultrasound images based on the extracted multiple key cardiac features, and generates shooting suggestions according to the quality evaluation results, which can automatically evaluate the quality of cardiac ultrasound images and provide real-time shooting optimization suggestions, realizing efficient and accurate processing of cardiac ultrasound images.

[0019] Advantages of additional aspects of the present invention will be given in part in the following description, become apparent in part from the following description, or be learned through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings forming a part of this specification are used to provide a further understanding of the present invention. The schematic embodiments and descriptions thereof of the present invention are used to explain the present invention and do not constitute an improper limitation to the present invention.

[0021] Figure 1 is a flowchart of a method for generating shooting suggestions for cardiac ultrasound images provided by an embodiment of the present invention; Figure 2 is the preprocessing result of the input cardiac valve image provided by an embodiment of the present invention; Figure 3 is the two-dimensional cardiac ultrasound image before adjustment provided by an embodiment of the present invention; Figure 4 is the two-dimensional cardiac ultrasound image after adjustment provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0023] It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0024] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular forms are also intended to include the plural forms. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0025] Regarding the existing evaluation methods mentioned in the background art, the lack of automated analysis of cardiac ultrasound images and the generation of optimized shooting suggestions may lead to the omission of important medical information.

[0026] The present invention obtains cardiac ultrasound image data; extracts multi-scale spatial features of the cardiac ultrasound image, and obtains a multi-scale spatial fusion feature after fusing the multi-scale spatial features; performs spatial attention enhancement and channel attention enhancement on the multi-scale spatial fusion feature to obtain a multi-scale spatial fusion feature with enhanced spatial attention and channel attention; calculates the clarity of the cardiac contour and the echo uniformity based on the multi-scale spatial fusion feature with enhanced spatial attention, and calculates the clarity of the cardiac valve based on the multi-scale spatial fusion feature with enhanced channel attention; extracts features at different time points of the cardiac ultrasound image, and calculates the consistency of left ventricular wall motion based on the features at different time points of the cardiac ultrasound image; extracts the signal-to-noise ratio of the cardiac ultrasound image; performs image quality assessment based on the optimized cardiac contour clarity, echo uniformity, cardiac valve clarity, left ventricular wall motion consistency and signal-to-noise ratio, and generates shooting suggestions according to the assessment results.

[0027] Embodiment 1 As Figure 1 shown, this embodiment provides a method for generating cardiac ultrasound image shooting suggestions, including the following steps: Step 1: Obtain cardiac ultrasound image data , preprocess the obtained cardiac ultrasound image data to obtain preprocessed cardiac ultrasound image data; as Figure 2 shown is the preprocessed cardiac ultrasound image; Step 2: Extract multi-scale spatial features of the cardiac ultrasound image, and obtain a multi-scale spatial fusion feature after fusing the multi-scale spatial features; In this embodiment, a multi-scale convolutional neural network is used to extract the spatial features of the cardiac ultrasound image. By using convolution kernels of various different sizes, feature maps extracted by different convolution kernels are obtained, and then the feature maps of each scale are fused to obtain a fused feature map, expressed as: , where is the feature map extracted by the th convolution kernel, K is the number of convolution kernels, is the fusion weight, used to adjust the importance of each feature map; Step 3: Perform spatial attention enhancement and channel attention enhancement on the multi-scale spatial fusion feature to obtain a multi-scale spatial fusion feature with enhanced spatial attention and channel attention; Specifically, it includes the following steps: Step 301: Perform spatial attention enhancement on the multi-scale spatial fusion feature to obtain a multi-scale spatial fusion feature with enhanced spatial attention; In this embodiment, by calculating the importance of different positions in the image to focus on the cardiac edge region, especially when evaluating the clarity of the cardiac cavity contour, spatial attention can help the model accurately focus on the edge part. The specific formula is: , where, represents the importance degree of the pixels at each position, represents the weight assignment of the pixels at different positions when calculating the importance, represents the convolution operation. By convolving with the weight matrix , the correlation information between different positions in the image is captured, is the bias term, which is used to adjust the offset of the calculation result, is the activation function, which maps the result after convolution and bias calculation to a set interval, so that the output can represent the importance degree of each position.

[0028] According to the importance degree of each position, the final multi-scale spatial fusion feature map with enhanced spatial attention is obtained; Step 302: Enhance the channel attention of the multi-scale spatial fusion feature to obtain the multi-scale spatial fusion feature with enhanced channel attention; Reinforce the features related to the cardiac structure by channel weighting, and suppress the noise and irrelevant parts.

[0029] The calculation formula for the importance degree score of the pixels corresponding to each channel is: , where, represents the importance degree of the pixels corresponding to each channel, is the weight vector of the channel attention, which weights the data of different channels, represents the dot product operation. By taking the dot product with the weight vector , the data can be adjusted according to the importance of different channels, is the bias term, which is used to adjust the offset of the calculation result, is the activation function, which maps the calculation result to a set interval.

[0030] According to the importance degree of each channel, the final multi-scale spatial fusion feature map with enhanced channel attention is obtained; Step 4: Calculate the cardiac contour clarity and echo uniformity based on the multi-scale spatial fusion feature with enhanced spatial attention, and calculate the cardiac valve clarity based on the multi-scale spatial fusion feature with enhanced channel attention; Specifically, it includes the following steps: Step 401: Calculate the clarity of the cardiac contour and the echo uniformity based on the multi-scale spatial fusion features enhanced by spatial attention ; ; Among them, the Canny edge detection method is used to detect the multi-scale spatial fusion features enhanced by spatial attention to identify the boundaries of the cardiac chambers, and the clarity of the cardiac contour is obtained ; When evaluating the clarity of the cardiac chamber contour, the model needs to accurately identify the edges of the cardiac chambers. Through the above calculations, the spatial attention mechanism can highlight the importance of the edge regions in the image

[0031] For example, for the pixels at the edges of the cardiac chambers, their corresponding values may be higher, while the pixels far from the edges have lower values. In this way, the model can pay more attention to the features of the edge regions, thereby more accurately quantifying the clarity of the cardiac chamber contour

[0032] Among them, the calculation process of the echo uniformity includes: Divide the multi-scale spatial fusion features enhanced by spatial attention into multiple regions, calculate the standard deviation and average value of the pixel intensities in each region respectively, and obtain the echo uniformity based on the standard deviation and average value of the pixel intensities in each region ; , Among them, M represents the number of regions, and respectively represent the standard deviation and average value of the pixel intensities in the th local region

[0033] Step 402: Calculate the clarity of the cardiac valves based on the multi-scale spatial fusion features enhanced by channel attention and the texture features of the image ; Specifically, it includes the following steps: Step 4021: Extract the global features and local features in the image based on the multi-scale spatial fusion features enhanced by channel attention In this embodiment, a CNN is used to extract features from the valve region to capture the global and local features in the image Step 4022: Extract the texture features of the cardiac ultrasound image In this embodiment, first, preprocess the cardiac ultrasound image, including denoising and enhancement, to ensure that the image quality is suitable for feature extraction Next, the gray-level co-occurrence matrix (GLCM) method is used to extract texture features such as contrast, correlation, uniformity, etc. In addition, the local binary pattern (LBP) is used to analyze the texture features of local areas of the image; by selecting these texture features, the most representative features are selected for subsequent analysis.

[0034] Step 4033: Fuse the texture features with the global and local features in the image to evaluate the clarity of the valve area.

[0035] When assessing heart valve clarity, data from different channels may contain different information. Some channels may be more relevant to the characteristics of the heart valve structure, while others may contain more noise or irrelevant information. The channel attention mechanism, through the above calculations, can increase the weight of channels related to the heart valve structure, allowing the model to pay more attention to the data of these important channels when extracting features, thereby more accurately quantifying heart valve clarity.

[0036] Step 5: Extract features of the cardiac ultrasound image at different time points, and calculate the consistency of left ventricular wall motion based on the features of the cardiac ultrasound image at different time points; The specific steps include: Step 501: extracting features of cardiac ultrasound images at different time points; In this embodiment, the TCN (Temporal Convolutional Network) model is used for analysis. TCN performs a convolution operation on the data at each time step, which is expressed as: , in, is the temporal convolution kernel weight, is the length of the convolution kernel, For the input data sequence At time step The value at the moment represents the current time step The historical data that the upconvolution operation depends on.

[0037] Through this convolution operation, TCN can capture the dependencies between data at different time points. For example, during a cardiac cycle, the states of the ventricular wall at different times are correlated, with the state at earlier times affecting the state at later times. By applying convolution to these time series data, TCN can tap into this inherent temporal dependency.

[0038] Step 502: Calculate the consistency of left ventricular wall motion based on the features of the cardiac ultrasound image at different time points; In this embodiment, when extracting the motion consistency feature of the heart wall during the cardiac cycle, a TCN (Temporal Convolutional Network) model is used for analysis. The TCN is used to analyze the temporal features of the image, especially the dynamic changes of the left ventricular wall, and evaluate the motion consistency of the left ventricular wall by extracting temporal dependence features; Motion consistency feature of the left ventricular wall The calculation formula is: , where, represents the motion consistency of the left ventricular wall, and represent the ventricular wall feature vectors at time points and respectively.

[0039] Step 6: Extract the signal-to-noise ratio of the cardiac ultrasound image; The signal-to-noise ratio quantifies the ratio of the signal to the noise in the image and reflects the clarity of the effective information.

[0040] In this embodiment, the signal-to-noise ratio of the cardiac ultrasound image is calculated by calculating the signal and background noise in the selected area. The calculation formula is: , where, is the average value of the signal, is the standard deviation of the noise.

[0041] Step 7: Maximize the feature similarity of similar samples through contrastive learning to obtain optimized cardiac contour clarity, echo uniformity, cardiac valve clarity, left ventricular wall motion consistency, and signal-to-noise ratio; Specifically, it includes the following steps: Step 701: Define similar samples; For a pair of samples, if they are considered similar under a certain predefined similarity criterion, then , where can be some attributes based on the data itself or a similarity label determined by a certain algorithm. For example, in cardiac ultrasound image data, two samples can be judged whether they are similar according to some basic features of the image (such as imaging angle, general similarity of cardiac structure, etc.); Step 702: Construct a loss function that maximizes the feature similarity of each feature of similar samples: , where, and represent different sample indices, represents sample and sample The inner product of the sum of the eigenvectors. The value of the inner product reflects the similarity degree of the two eigenvectors in direction. and are the norms (lengths) of the eigenvectors and respectively. Through calculation, the cosine similarity of the two eigenvectors is obtained, and its value is between [-1, 1]. The closer it is to 1, the more similar the two eigenvectors are.

[0042] When it is hoped that the eigenvectors of two similar samples are as similar as possible, that is, the cosine similarity is as close to 1 as possible. Therefore, is used in the loss function to measure the difference degree of the features between similar samples. By summing all the sample pairs that meet the requirements, the loss function of contrastive learning is obtained.

[0043] Step 703: Minimize the loss function by continuously adjusting the parameters of the model (such as the weights of the convolutional kernels, etc.). The model gradually learns the feature expressions that can better distinguish similar samples and dissimilar samples. In this way, under the unsupervised condition, the model can learn the effective feature expressions for the cardiac ultrasound image data, providing a more accurate feature basis for subsequent image quality assessment and generation of shooting suggestions.

[0044] Step 8: Perform image quality assessment based on the optimized cardiac contour clarity, echo uniformity, cardiac valve clarity, left ventricular wall motion consistency, and signal-to-noise ratio, and generate shooting suggestions according to the assessment results; Specifically, it includes the following steps: Step 801: Perform image quality assessment based on the optimized cardiac contour clarity, echo uniformity, cardiac valve clarity, and left ventricular wall motion consistency to obtain the assessment result; In this embodiment, each image quality index is scored according to the set quantization scoring standard, and the scoring result is 0 - 5 points; Step 802: Generate shooting suggestions according to the obtained result; In this embodiment, the specific suggestions include: For cardiac contour clarity: If the score of the cardiac cavity contour clarity is less than 2 points, it is recommended to adjust the probe angle by at least 15 degrees to improve the visualization of the cardiac cavity edge; If the score of the cardiac cavity contour clarity is between 2 - 3 points, it is recommended to slightly adjust the probe angle by 5 - 10 degrees, or increase the image contrast to improve the edge sharpness; If the score of the cardiac cavity contour clarity is greater than 3 points, the current setting is sufficient and no adjustment is required.

[0045] For left ventricular wall motion consistency: If the score of the left ventricular wall motion consistency is less than 2 points, it is recommended to adjust the probe position to ensure comprehensive capture of the cardiac cycle, or change the imaging mode; If the score of the left ventricular wall motion consistency is between 2 and 3 points, it is recommended to slightly adjust the probe position or slightly adjust the capture timing of the cardiac cycle; If the score of the left ventricular wall motion consistency is greater than 3 points, the current settings are sufficient and no adjustment is required.

[0046] For echo uniformity; If the score of echo uniformity is less than 2 points, it is recommended to adjust the probe focusing settings or change the scanning depth to a more uniform area; If the score of echo uniformity is between 2 and 3 points, it is recommended to slightly adjust the transmit power to optimize tissue reflections at different depths; If the score of echo uniformity is greater than 3 points, the current settings are sufficient and no adjustment is required.

[0047] For cardiac valve clarity: If the score of echo uniformity is less than 2 points, it is recommended to adjust the probe angle by at least 20 degrees to better align with the valve area, or change the imaging frequency; If the score of echo uniformity is between 2 and 3 points, it is recommended to slightly adjust the probe angle by about 10 degrees or slightly adjust the imaging frequency; If the score of echo uniformity is greater than 3 points, the current settings are sufficient and no adjustment is required.

[0048] The system will monitor the scores of these indicators in real time and provide specific adjustment suggestions. The operator can adjust the device settings step by step according to the system prompts until the ideal image quality is achieved.

[0049] Such as Figure 3 and Figure 4 shown, Figure 3 is the two-dimensional cardiac ultrasound image before adjustment, and the set parameters are a probe angle of 0° and a focusing depth of 10 cm. Figure 4 is the two-dimensional cardiac ultrasound image after adjustment, and the set parameters are obtained by adjusting according to the present invention, with a probe angle of 10° and a focusing depth of 12 cm. By comparison, it can be seen that the ultrasound image obtained by the method of the present invention has better effects than the unadjusted image.

[0050] Embodiment 2 This embodiment provides a cardiac ultrasound image shooting advice generation system, including: A data acquisition module for acquiring cardiac ultrasound image data; A multi-scale feature extraction module for extracting multi-scale spatial features of cardiac ultrasound images and obtaining multi-scale spatial fusion features after fusing the multi-scale spatial features; A multi-scale feature enhancement module is used to perform spatial attention enhancement and channel attention enhancement on the multi-scale spatial fusion features to obtain multi-scale spatial fusion features with enhanced spatial attention and channel attention. A feature extraction module is used to calculate the clarity of the heart contour and the echo uniformity based on the multi-scale spatial fusion features with enhanced spatial attention, calculate the clarity of the heart valves based on the multi-scale spatial fusion features with enhanced channel attention; extract the features of different time points of the cardiac ultrasound image, and calculate the left ventricular wall motion consistency based on the features of different time points of the cardiac ultrasound image; extract the signal-to-noise ratio of the cardiac ultrasound image. A shooting suggestion generation module is used to perform image quality assessment based on the clarity of the heart contour, the echo uniformity, the clarity of the heart valves, the left ventricular wall motion consistency and the signal-to-noise ratio, and generate shooting suggestions according to the assessment results.

[0051] It should be noted that the specific implementation manner of the cardiac ultrasound image shooting suggestion generation system in the embodiments of the present invention is similar to the specific implementation manner of the cardiac ultrasound image shooting suggestion generation method in the embodiments of the present invention. For details, please refer to the description in the method part. To reduce redundancy, it will not be elaborated here.

[0052] Embodiment III This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the steps in the cardiac ultrasound image shooting suggestion generation method as described above are implemented.

[0053] Embodiment IV This embodiment provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the steps in the cardiac ultrasound image shooting suggestion generation method as described above are implemented.

[0054] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for generating cardiac ultrasound imaging shooting suggestions, characterized in that, It includes the following steps: Obtain cardiac ultrasound image data; Extract multi-scale spatial features of the cardiac ultrasound image, and after fusing the multi-scale spatial features, obtain multi-scale spatial fusion features; Perform spatial attention enhancement and channel attention enhancement on the multi-scale spatial fusion features to obtain multi-scale spatial fusion features with spatial attention enhancement and channel attention enhancement; Calculate the clarity of the cardiac contour and the echo uniformity based on the multi-scale spatial fusion features with spatial attention enhancement, and calculate the clarity of the cardiac valve based on the multi-scale spatial fusion features with channel attention enhancement; Extract the features of different time points of the cardiac ultrasound image, and calculate the consistency of the left ventricular wall motion based on the features of different time points of the cardiac ultrasound image; Extract the signal-to-noise ratio of the cardiac ultrasound image; Perform image quality assessment based on the clarity of the cardiac contour, the echo uniformity, the clarity of the cardiac valve, the consistency of the left ventricular wall motion, and the signal-to-noise ratio, and generate shooting suggestions according to the assessment results.

2. The method for generating a cardiac ultrasound imaging shooting suggestion according to claim 1, wherein The extraction of the multi-scale spatial features of the cardiac ultrasound image and the fusion of the multi-scale spatial features to obtain multi-scale spatial fusion features include: Use a multi-scale convolutional neural network to extract the spatial features of the cardiac ultrasound image. By using convolutional kernels of various different sizes, feature maps extracted by different convolutional kernels are obtained, and then the feature maps of each scale are fused to obtain a fused feature map.

3. The method for generating a cardiac ultrasound imaging shooting suggestion according to claim 1, wherein, The calculation formula for performing spatial attention enhancement and channel attention enhancement on the multi-scale spatial fusion features is: ， ， Among them, represents the importance degree of the pixels at each position, represents the multi-scale spatial fusion feature, represents the weight distribution of the pixels at different positions when calculating the importance, represents the convolution operation, through the convolution with the weight matrix to capture the correlation information between different positions in the image, and is the bias term, used to adjust the offset of the calculation result; represents the importance degree of the pixels corresponding to each channel, is the weight vector of the channel attention, represents the dot product operation, is the activation function.

4. The method for generating a cardiac ultrasound imaging shooting suggestion according to claim 1, wherein, Use the Canny edge detection method to detect the multi-scale spatial fusion features with spatial attention enhancement to identify the cardiac cavity boundary and obtain the clarity of the cardiac contour.

5. The method for generating a cardiac ultrasound imaging shooting suggestion according to claim 1, wherein, Divide the multi-scale spatial fusion features with spatial attention enhancement into multiple regions, calculate the standard deviation and average value of the pixel intensities within each region respectively, and obtain the echo uniformity based on the standard deviation and average value of the pixel intensities within each region.

6. The method for generating a cardiac ultrasound imaging shooting suggestion according to claim 1, wherein, The method also includes maximizing the similarity of each feature of similar samples through contrastive learning to obtain optimized clarity of the cardiac contour, echo uniformity, clarity of the cardiac valve, consistency of the left ventricular wall motion, and signal-to-noise ratio.

7. The method for generating a cardiac ultrasound imaging shooting suggestion according to claim 6, wherein, The loss function for maximizing the similarity of each feature of similar samples is: ， Among them, and represent different sample indices, represents the sample and the sample is the inner product of the sum of the feature vectors, and are the norms of the sum of the feature vectors respectively.

8. A cardiac ultrasound image capture recommendation generation system, characterized in that, It includes: A data acquisition module for obtaining cardiac ultrasound image data; A multi-scale feature extraction module for extracting multi-scale spatial features of the cardiac ultrasound image and obtaining multi-scale spatial fusion features after fusing the multi-scale spatial features; A multi-scale feature enhancement module for performing spatial attention enhancement and channel attention enhancement on the multi-scale spatial fusion features to obtain multi-scale spatial fusion features with spatial attention enhancement and channel attention enhancement; A feature extraction module for calculating the clarity of the cardiac contour and the echo uniformity based on the multi-scale spatial fusion features with spatial attention enhancement, and calculating the clarity of the cardiac valve based on the multi-scale spatial fusion features with channel attention enhancement; extracting the features of different time points of the cardiac ultrasound image and calculating the consistency of the left ventricular wall motion based on the features of different time points of the cardiac ultrasound image; extracting the signal-to-noise ratio of the cardiac ultrasound image; The shooting suggestion generation module is used to evaluate the image quality based on the clarity of the heart contour, the echo uniformity, the clarity of the heart valves, the consistency of the left ventricular wall movement, and the signal-to-noise ratio, and generate shooting suggestions according to the evaluation results.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the method for generating a shooting suggestion for cardiac ultrasound images as described in any one of claims 1-7.

10. A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the method for generating a shooting suggestion for cardiac ultrasound images as described in any one of claims 1-7.

Images