Heart color ultrasound image processing method based on enhancement processing

Abstract

The invention discloses a heart color ultrasound image processing method based on enhancement processing, belongs to the field of image enhancement, and aims to improve the structural definition and tissue recognition capability of heart color ultrasound images. Then calculating pixel local energy and establishing a signal excitation attention guiding module in combination with an energy difference driving strategy and a structure adjustment factor, then introducing a disturbance generation mechanism and local response difference modeling to construct a local disturbance divergence guiding mapping module, and optimizing image features through nonlinear projection; then, combining the response residual error, the local information entropy and the key regulation factor to construct a reversible response entropy coding module, controlling the enhancement intensity of different areas, and integrating all the modules to construct a heart color ultrasound image enhancement model; and finally, inputting the heart color Doppler ultrasound image into the model to output a heart color Doppler ultrasound enhanced image with high quality, rich details and high diagnostic readability.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of image enhancement, and particularly relates to a heart color ultrasound image processing method based on enhancement processing. BACKGROUND

[0002] At present, the quality of a heart color ultrasound image, which is an important medical image in clinical diagnosis and auxiliary analysis of heart lesions, affects the recognition accuracy of doctors on tissue boundaries and lesion areas. Traditional heart color ultrasound image enhancement methods rely on image processing techniques such as contrast enhancement, edge sharpening, and fixed rules of filter denoising. In complex tissue structures and low signal-to-noise ratio conditions, problems such as edge blurring, detail loss, and noise residue often occur, which limits the stability and accuracy of the enhanced results in intelligent diagnosis. Therefore, it is urgent to propose a processing method that can combine tissue response characteristics and detail expression to improve the structural clarity and tissue recognition ability of low-quality heart color ultrasound images.

[0003] Publication No. CN117379098A discloses a heart ultrasound image enhancement system, which intelligently determines the scanning area and collection quality by analyzing the real-time sound wave attenuation value, and performs automatic contrast and grayscale adjustment, target area automatic identification and signal preprocessing after image collection, thereby ensuring that the image quality meets the subsequent enhancement processing and overall improving the visibility of heart structures and image clarity. Publication No. CN112914610A discloses a contrast-enhanced heart motion graph wall thickness automatic analysis system based on deep learning, which first performs quality identification and tissue pre-labeling on the image, then evaluates the feasibility by the model and performs enhancement processing, and finally automatically measures the wall thickness, enhancing the resolution of edge recognition and structural details in the image. These methods have significantly improved the overall readability, edge clarity, and recognition of key structures in the heart color ultrasound scene.

[0004] However, the existing technology still has deficiencies in processing color ultrasound images with blurred boundaries and significant noise interference. The above-mentioned methods do not consider local response differences, disturbance characteristic modeling, and information entropy adjustment strategies at the same time, ignoring the influence of detail areas on the overall quality during the enhancement process. In complex backgrounds, the existing technology is prone to artifacts, detail loss, and inconsistent enhancement results, making it difficult to meet the requirements of clinical high-quality image enhancement. SUMMARY

[0005] The present application provides a heart color ultrasound image processing method based on enhancement processing, which aims to construct a signal excitation attention module, a local disturbance divergence guided mapping module, and a reversible response entropy coding module, combine regional structure modeling, disturbance propagation mechanism, and information entropy enhancement strategy, and realize the saliency modeling and detail enhancement of key tissue areas in heart color ultrasound images, thereby improving the accuracy of image recognition, the clarity of boundary expression, and the robustness of the overall processing model.

[0006] The present application aims to provide a color ultrasound image enhancement model, and provides a heart color ultrasound image processing method based on enhancement processing, which comprises the following steps: S1, collecting original heart color ultrasound image data to generate a heart color ultrasound image data set containing a target tissue region; S2, introducing a feature matrix, constructing a signal excitation guided attention module, and performing structural enhancement on the heart color ultrasound image data through the signal excitation guided attention module to generate an enhanced heart color ultrasound image; S3, introducing local perturbation response difference calculation, constructing a local perturbation divergence guided mapping module, and processing the enhanced heart color ultrasound image through the local perturbation divergence guided mapping module to generate an optimized heart color ultrasound image feature matrix; S4, designing a response residual and local information entropy calculation mechanism, introducing a key adjustment factor, constructing a reversible response entropy coding module, and enhancing the optimized heart color ultrasound image through the reversible response entropy coding module to generate a high-quality enhanced heart color ultrasound image; S5, integrating the signal excitation guided attention module, the local perturbation divergence guided mapping module and the reversible response entropy coding module to construct a heart color ultrasound image enhancement processing model; S6, inputting the heart color ultrasound image into the heart color ultrasound image enhancement processing model for image enhancement, and outputting a high-quality enhanced heart color ultrasound image.

[0007] Preferably, in S2, the signal excitation guided attention module is constructed and comprises the following steps: Step S21, defining a signal excitation intensity function for calculating the energy excitation intensity of each position in the input feature matrix , the mathematical model is: ; wherein, is the total number of spatial positions, is the number of channels, is the feature value of the position in the channel, as a signal excitation factor represents the local energy response of the position , providing basic input for subsequent connection matrix calculation.

[0008] Step S22, introducing an energy difference driven connection weight function for describing the similarity connection degree between any two positions and suppressing weakly related connections with large energy difference, the mathematical model is: ; wherein, is a temperature parameter, is an energy difference threshold, is an indicator function, the matrix The sparsity control is realized by exponential decay and threshold screening, so that the pixel positions with close energy produce strong connections and irrelevant areas are weakened.

[0009] Step S23, defining a structure response adjustment factor for enhancing the connection weight between high-energy positions and suppressing low-energy area connection, thereby improving the expression ability of structure features, the mathematical model is: ; wherein, is the maximum value of the energy excitation vector , is a nonlinear adjustment index. The adjustment factor amplifies the contribution of high-response areas to network feature enhancement through the normalization of energy products, while reducing the influence of background noise areas.

[0010] Step S24, combining the energy difference driven connection matrix with the structure response adjustment factor matrix , to construct a structure guided connection mapping matrix for realizing the reconstruction of global feature relationship and the enhancement of signal consistency, the mathematical model is: wherein, integrates the dual effects of sparse connection and energy adjustment, ensuring that high-correlation structure information is strengthened and providing guiding mapping for subsequent feature enhancement.

[0011] Step S25, using the structure guided connection mapping matrix to perform mapping operation on the input feature matrix , to obtain the output feature matrix , realizing the global enhancement of structure features and the suppression of noise interference, the mathematical model is: ; wherein, is the enhanced feature matrix, which is input to the subsequent local perturbation divergence guided mapping module (LPDI), effectively improving the recognition and enhancement effect of structure areas of color Doppler ultrasound images.

[0012] ​Preferably, the signal excitation guide attention module constructed by S2 excites attention, and the energy excitation factor is defined in S21 to accurately calculate the energy response of each pixel position in the cardiac color ultrasound image, thereby effectively capturing the correlation between the structural features and regions in the cardiac color ultrasound image, avoiding the deficiency of the traditional method that cannot dynamically capture local information, and improving the accuracy of structure enhancement; S22 introduces an energy difference driven connection weight function, which can adaptively adjust the connection strength between each position according to the energy excitation difference of the pixels, improving the flexibility and response ability of the cardiac color ultrasound image model; S23 further strengthens the connection between high-energy regions by combining a local structure response adjustment factor, reduces the influence of low-energy regions on the model, and enhances the transmission effect of signals in the cardiac color ultrasound image; The nonlinear correction factor of S24 further optimizes the weight of the connection matrix, so that the model can realize more accurate structure enhancement in the cardiac color ultrasound image; Overall, this module significantly improves the accuracy and robustness of structure enhancement in the cardiac color ultrasound image through the signal energy excitation and adaptive structure adjustment mechanism.

[0013] Preferably, in S3, the local disturbance divergence guide mapping module comprises the following steps: Step S31, the signal excitation guide feature matrix output by S2 Step S32, for each channel , define the disturbance response difference variance and , and the mathematical model is: ; ; Wherein is the disturbance amplitude, which is a fixed constant and will not be updated with training; represents the element level sign function, which is used to introduce positive and negative disturbances. In this way, two disturbance versions are generated to compare and calculate the response difference of each position in the image.

[0014] Step S32, for each channel , define the disturbance response difference variance , and the mathematical model is: ; Wherein, represents the response sensitivity of channel to disturbance, that is, its instability, and are the disturbance versions of position in the first channel, and the divergence reflects the connection between the channels The sensitivity of the perturbation, the greater the divergence value represents the channel response to the perturbation is stronger, the stability of the image region is poor. Through this calculation, we can evaluate the response stability of each channel under the perturbation, reflecting the local instability of the image.

[0015] Step S33, to realize the perturbation guided mapping, construct the structure guiding matrix The mathematical model is: ; Wherein, represents the structural correlation between the position and the position , based on the difference of their perturbation response And The matrix is calculated, which guides the signal flow by reflecting the difference of perturbation response, effectively enhancing the region with strong structural consistency.

[0016] Step S34, using the constructed structure guiding matrix Map the input feature matrix To get the final enhanced feature matrix The mathematical model is: ; Where the structure guiding matrix Adjust the flow of image information according to the difference of perturbation response, so as to optimize the transmission of information flow and enhance the structure expression of the more stable region in the image.

[0017] Step S35, use tanh function to carry out nonlinear projection transformation on the calculation result, get the final feature matrix, the mathematical model is: ; Where, tanh function enhances the dynamic range of perturbation response region, ensures that the details of the image are enhanced; To maintain the amplitude of the original signal, ensure the integrity of the original information; The final output maintains the features after the perturbation guidance, while ensuring the consistency of the image information.

[0018] Preferably, through the S3 local perturbation divergence guided mapping module, in S31, by defining a perturbation generation mechanism, the perturbation response at each pixel location in the image can be accurately calculated, thereby effectively capturing the stability differences of the cardiac ultrasound image region and avoiding the shortcomings of traditional methods that cannot dynamically adapt to local perturbations, thus improving the accuracy of structure enhancement; in S32, by introducing the perturbation response difference variance, the response sensitivity of each channel under perturbation can be accurately evaluated, thereby adjusting the propagation intensity of cardiac ultrasound image features according to the stability of the region, improving the adaptability and flexibility of the model; in S33, by constructing a structure guidance matrix and combining it with the perturbation response difference, the connection between regions with strong structural consistency is further enhanced, reducing the interference of regions with weak perturbation response to the model and optimizing the information flow transmission effect; in S34, by using the structure guidance matrix to perform weighted mapping of input features, the signal transmission is effectively guided, ensuring that the features of stable regions in the image are enhanced; in S35, the nonlinear correction factor further optimizes the weight of the feature matrix, enabling the model to achieve more accurate feature enhancement in cardiac ultrasound images; overall, this module uses a local perturbation difference guidance mechanism and a structural response adjustment mechanism.

[0019] Preferably, in S4, constructing the reversible response entropy encoding module specifically includes the following steps: Step S41: To accurately assess the reliability of the image structure, a locally reconstructed estimated echocardiogram image is first constructed to estimate the reconstructed value of each pixel in its neighborhood. This reconstruction process uses median filtering to avoid noise interference and the influence of local inconsistencies. The mathematical model is as follows: ; in In pixels The 3x3 neighborhood centered on the center, The set of neighboring pixels after removing the center pixel. This step, by removing local anomalies, provides a more stable basis for subsequent response calculations, resulting in a reconstructed estimate.

[0020] Step S42: Calculate the response residual between each pixel location and its reconstructed value. It is used to reflect the differences and instabilities of image regions, and its mathematical model is as follows: ; in, In response to the residual, a large residual value indicates that there is a significant change in the image content of the pixel area, while a small residual value indicates that the area is relatively smooth and stable.

[0021] Step S43: To measure the amount of information in the neighborhood of each pixel, define local entropy. , which reflects the complexity and information density of the image region, and its mathematical model is: ; wherein, is the normalized probability distribution of the pixel values in the neighborhood, is the neighborhood set, represents the local entropy value of the pixel region, which can represent the complexity of the local region of the image, and a high entropy value represents that the region is rich in information, and a low entropy value represents that the region is less in information.

[0022] Step S44, combining the response residual and the local information entropy, the response entropy score of each pixel is calculated , which represents that the pixel belongs to which information region, and its mathematical model is: ; wherein, is a very small constant, used to prevent division by zero error, and the response entropy score is used to distinguish the high information region and the low information region of the image, and for the low information region, the value is large, and for the high information region, the value is small.

[0023] Step S45, constructing a key adjustment factor is used to dynamically adjust the intensity of image enhancement according to the reflection intensity and structural complexity of the color Doppler ultrasound image. The adjustment factor is based on the reflection intensity change and the structural feature reconstruction error, so as to ensure that in the image enhancement process, the low information region is not over-enhanced, and the high information region is properly enhanced, and its mathematical model is: ; wherein, is the global reference adjustment intensity, which controls the overall intensity of enhancement; represents the ultrasonic reflection intensity at the position , which reflects the reflection characteristics of the tissue in the image; and are the mean and standard deviation of the ultrasonic reflection intensity image respectively, which represent the global distribution of the reflection intensity in the image; is a very small constant, used to prevent division by zero error; represents the structural reconstruction error at the position , which reflects the detail complexity of the region, in the embodiment, the structural reconstruction error is a mathematical model: ; is the average structural reconstruction error of the whole image, which represents the global structural complexity of the image, and in the embodiment, the average structural reconstruction error is a mathematical model: 。。

[0024] Step S46, constructing the final defogging mask , which is used to control the difference of enhancement intensity between structure region and low reflection region in the image enhancement process. The mask is generated by combining the response entropy score and the adjustment factor , which generates an adjustable enhancement factor based on the local information of the image, strengthens the high reflection region and suppresses the over-enhancement of the low reflection region, and its mathematical model is: ; wherein, is the adjustment factor, which dynamically controls the image enhancement intensity and depends on the ultrasonic reflection intensity and the structure complexity, and the detailed calculation is shown in step S45; is the response entropy score, which is used to measure the combination degree of the response residual and the local information entropy of each pixel, and indicates whether the pixel belongs to the high information region, and the specific calculation is shown in step S44; the mask For the high information region, the response entropy score is small, the adjustment factor is large, so the mask will keep a small value, reducing the over-enhancement of the image; for the low reflection region, the response entropy score is large, the adjustment factor is small, so the mask will increase, enhancing the detail and structure performance of these regions; the mask obtained by calculation is used as the adjustment factor of image enhancement, which automatically adjusts the enhancement intensity of different regions, ensuring that the structure information in the image is moderately enhanced, while avoiding the over-enhancement of the low reflection region.

[0025] Step S47, using the response entropy encoding mask to perform enhancement processing on the input image, obtaining the enhanced image , and the target of this step is to automatically adjust the enhancement intensity according to the different reflection characteristics and structure complexity of the image region, so as to highlight the key information region in the image, while avoiding the over-enhancement of the low reflection region, and its mathematical model is: ; wherein, is the original pixel value of the input image; is the mask value calculated according to the response entropy score and the adjustment factor; is the pixel value of the enhanced image; for the high reflection region, since is small, the enhanced image will keep a relatively stable pixel value, avoiding over-enhancement; for the low reflection region, The larger, enhanced image will enhance the details in these areas, enhance the contrast and structural clarity of the image; in this step, the image is enhanced by weighting, and the enhanced image is finally obtained The image has been emphasized in the structure area and suppressed in the background area, and the response entropy coding mask is obtained It plays a key role in the enhancement process, which can automatically adjust the enhancement process according to the characteristics of the image content, thereby improving the quality and recognizability of the image.

[0026] Preferably, by defining a local reconstruction estimation mechanism in S41, the S4 reversible response entropy coding module can accurately calculate the reconstruction error of each pixel position in the cardiac color ultrasound image, thereby effectively evaluating the detail complexity of the image area, avoiding the shortcomings of traditional methods that cannot dynamically adapt to local changes, and improving the accuracy of cardiac color ultrasound image enhancement; S42 can accurately evaluate the change degree of each region in the cardiac color ultrasound image by calculating the response residual, thereby enhancing the structural detail area and suppressing the excessive enhancement of the low information area; S43 introduces the calculation of local information entropy to accurately measure the complexity of the local area of the cardiac color ultrasound image, providing an effective basis for subsequent cardiac color ultrasound image enhancement, so that the model can better adapt to the diversity and complexity of the cardiac color ultrasound image; S44 can effectively distinguish between high information areas and low information areas in the cardiac color ultrasound image by combining response entropy scores with response residuals, providing accurate control for the enhancement process and ensuring the prominent performance of high information areas while avoiding excessive strengthening of low information areas; S45 introduces the illumination-component-driven adjustment factor, dynamically adjusts the enhancement intensity based on the illumination information and structural complexity of the image, so that the cardiac color ultrasound image enhancement process can better adapt to the characteristics of different regions, ensuring the local adaptability and global consistency of the enhancement effect; S46 can adaptively adjust the enhancement intensity according to the response characteristics of the cardiac color ultrasound image area by calculating the response entropy coding mask, further optimizing the structural performance and detail highlighting of the cardiac color ultrasound image, and ensuring the preservation of key information and the moderate suppression of low reflection areas in the enhancement process; S47 finally enhances the cardiac color ultrasound image through the response entropy coding mask, accurately adjusts the enhancement degree of different regions, and finally obtains the enhanced cardiac color ultrasound image, improving the quality and recognizability of the cardiac color ultrasound image. Overall, the module optimizes the response residual, local information entropy and adjustment factor to ensure the detail preservation and noise suppression in the cardiac color ultrasound image enhancement process, and improves the structural clarity and contrast of the cardiac color ultrasound image.

[0027] Preferably, in S5, the cardiac color ultrasound image enhancement processing model is constructed, which specifically includes the following steps: Step S51, first, the input cardiac color ultrasound image is formatted and preprocessed, the features of the cardiac color ultrasound image are extracted, and a feature matrix is constructed. The matrix is used for subsequent image enhancement processing to ensure that the image data meets the model input requirements.

[0028] Step S52, by constructing a signal excitation attention guiding module, the structural information in the cardiac color ultrasound image is enhanced based on the energy distribution of the cardiac color ultrasound image. The module constructs a connection matrix by calculating the signal excitation intensity of each position in the image, and adjusts the connection strength through a response adjustment factor, thereby enhancing the distinguishability of the key information region in the cardiac color ultrasound image.

[0029] Step S53, the local perturbation divergence guided mapping module calculates the response difference of each region in the image through a perturbation generation mechanism, and adjusts the feature transmission strength of the cardiac color ultrasound image according to the perturbation response difference. By introducing the perturbation response difference matrix, the structural consistency of the cardiac color ultrasound image is optimized, the details of the target region are strengthened, and the interference of the low reflection region on the enhancement result is reduced.

[0030] Step S54, in the reversibility response entropy coding module, the enhancement mask is constructed by calculating the response residual and the local information entropy, and finally the high-quality enhanced cardiac color ultrasound image is output. The module evaluates the detail region of the cardiac color ultrasound image according to the response residual and the information entropy score of the region of the cardiac color ultrasound image, and dynamically adjusts the enhancement strength of the cardiac color ultrasound image combined with the adjustment factor, thereby optimizing the quality and structural performance of the cardiac color ultrasound image.

[0031] In summary, due to the adoption of the technical solution, compared with the prior art, the beneficial effects of the present application are: the signal excitation attention guiding module successfully extracts important structural information in the cardiac color ultrasound image by performing energy excitation calculation on the color ultrasound image; the local perturbation divergence guided mapping module effectively adjusts the propagation strength of the features of the cardiac color ultrasound image by performing perturbation response difference calculation on the cardiac color ultrasound image, thereby optimizing the enhancement effect of the structural region; the reversibility response entropy coding module dynamically adjusts the enhancement strength of the cardiac color ultrasound image by combining the response residual and the local information entropy, so that the image details and high information regions are precisely enhanced. Overall, the model realizes high-quality enhancement of the cardiac color ultrasound image through the fusion design of structure guidance, response difference adjustment and information entropy driving, makes the target region more prominent, the details clear, and significantly improves the distinguishability and diagnostic value of the cardiac color ultrasound image. BRIEF DESCRIPTION OF DRAWINGS

[0032] Figure 1 A cardiac color ultrasound image processing method based on enhancement processing

[0033] Figure 2 A signal excitation attention guiding module structure diagram.

[0034] Figure 3 This is a structural diagram of the local perturbation divergence guided mapping module.

[0035] Figure 4 This is a diagram of the entropy encoding module for reversible response.

[0036] Figure 5 This is a diagram showing the overall structure of the color Doppler ultrasound image enhancement processing model.

[0037] Figure 6 Comparison of color Doppler ultrasound images before and after enhancement using a color Doppler ultrasound image enhancement processing model. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] Please see the appendix Figure 1 -Appendix Figure 6 This invention provides a method for processing cardiac color Doppler ultrasound images based on enhancement processing.

[0040] As attached Figure 1 As shown in the flowchart, this invention proposes a method for processing cardiac ultrasound images based on enhancement processing. The specific implementation method includes the following steps: S1. Acquire color Doppler ultrasound image data and generate a raw dataset of cardiac color Doppler ultrasound images containing the target tissue region.

[0041] Furthermore, as shown in the appendix Figure 1 The acquisition of cardiac ultrasound image data described in S1 specifically involves maintaining the stability of the acquisition device and controlling the imaging parameters, using a high-resolution ultrasound device to perform a non-destructive scan of the target area, and generating a cardiac ultrasound image dataset of 500 images. The dataset is in PNG format, and each image is uniformly preprocessed to a size of 512×512 pixels.

[0042] S2. Introduce a feature matrix and construct a signal-excited guided attention module. The echocardiogram image is structurally enhanced through the signal-excited guided attention module to generate an enhanced echocardiogram image.

[0043] Furthermore, as shown in the appendix Figure 1 The construction of the signal-induced attention module described in S2 is illustrated in the attached figure. Figure 2 As shown, the specific implementation of the module includes the following steps: Further, step S21, define signal excitation intensity function , for calculating the energy excitation intensity of each position in the input feature matrix The mathematical model is: ; Wherein, the size of the input cardiac color Doppler image is 512x512 pixels, and 3 channels are adopted, The total number of spatial positions, so L=512x512=262144 (i.e. the total number of pixels in the image), The number of channels, so C=3 (indicating that the color Doppler image uses three channels), The feature value of the i-th Position of the j-th Channel, As a signal excitation factor, it represents the local energy response of the position , providing basic input for subsequent connection matrix calculation.

[0044] Further, step S22, introduce energy difference driven connection weight function For describing the similarity connection degree between any two positions and inhibiting weakly related connections with large energy difference, the mathematical model is: ; Wherein, Temperature parameter, initial value is set to 0.1, which is gradually reduced according to the convergence, Energy difference threshold, threshold Used to control which pixels to establish connection, set threshold =2.0, which means that only when the signal excitation difference of two pixels is less than 2, it is considered that they are related, Indicator function, the matrix Achieve sparse control through exponential decay and threshold screening, so that pixels with close energy produce strong connection and irrelevant areas are weakened.

[0045] Further, step S23, define structure response adjustment factor , for enhancing the connection weight between high energy positions and inhibiting low energy area connection, so as to improve the expression ability of structure characteristics, the mathematical model is: ; Wherein, The maximum value of energy excitation vector , Nonlinear adjustment index, here Set to 0.5, used to avoid excessive adjustment, the adjustment factor​ The contribution of the high response area to the network feature enhancement is amplified by the normalization of the energy product, while the influence of the background noise area is reduced.

[0046] Further, step S24, based on the energy difference driven connection matrix Combined with the structure response adjustment factor matrix , the structure guided connection mapping matrix is constructed to realize the reconstruction of global feature relationship and signal consistency enhancement, and the mathematical model is: ; Among them, The dual effects of sparse connection and energy regulation are integrated to ensure that high correlation structure information is strengthened and provides guidance for subsequent feature enhancement. The structure guided connection mapping matrix is used to strengthen the connection of high correlation areas and suppress the connection of low correlation areas in the cardiac color Doppler ultrasound image, thereby realizing the reconstruction and enhancement of the cardiac color Doppler ultrasound image features.

[0047] Further, step S25, using the structure guided connection mapping matrix Map operation is performed on the input feature matrix to obtain the output feature matrix , realize the global enhancement of structure features and suppress noise interference, and the mathematical model is: ; Among them, The enhanced feature matrix is used as the input of the subsequent local perturbation divergence guided mapping module (LPDI), which effectively improves the recognition degree and enhancement effect of the structure area of the color Doppler ultrasound image.

[0048] S3, by introducing local perturbation response difference calculation, constructing a local perturbation divergence guided mapping module, and processing the enhanced cardiac color Doppler ultrasound image through the local perturbation divergence guided mapping module, an optimized cardiac color Doppler ultrasound image feature matrix is generated.

[0049] Further, as described in S3 of the attached Figure 3 , the specific construction steps of the local perturbation divergence guided mapping module are shown in the attached Figure 3 , and the specific implementation of the module includes the following steps: Further, step S31, the signal excitation guide feature matrix output by S2 is perturbed to obtain a perturbation response matrix. By introducing a small perturbation at each position, two perturbation versions and are obtained, and the mathematical model is: ; ; wherein is the perturbation amplitude, a fixed constant, set to 0.05 in this implementation, and does not update with training; is an element-wise sign function to introduce positive and negative perturbations, by which two perturbed versions are generated to compare and calculate the response difference of each position in the image.

[0050] Further, in step S32, for each channel , the perturbation response difference variance is defined, whose mathematical model is: ; wherein represents the response sensitivity of channel to the perturbation, i.e., its instability degree, and are the perturbed versions of position in the first channel, which are generated by step S31, representing the versions of the cardiac color ultrasound image feature matrix after adding and subtracting the perturbation, and the divergence reflects the sensitivity of channel to the perturbation, and a greater divergence value indicates that the channel has a stronger response to the perturbation and the stability of the cardiac color ultrasound image region is poor. Through this calculation, we can evaluate the response stability of each channel under the perturbation and reflect the local instability of the cardiac color ultrasound image.

[0051] Further, in step S33, to realize the perturbation guided mapping, a structure guiding matrix is constructed, whose mathematical model is: ; wherein represents the structural correlation degree between position and position , which is calculated based on the perturbation response difference and , and the matrix guides the signal flow by reflecting the perturbation response difference, effectively enhancing the regions with strong structural consistency.

[0052] Further, in step S34, the constructed structure guiding matrix is used to map the input feature matrix , and the final enhanced feature matrix is obtained by weighting the response features of each pixel, and its mathematical model is: ; Among them, through the structural guiding matrix By adjusting the flow of image information according to the differences in perturbation response, the transmission of information flow is optimized and the structural expression of more stable regions in the image is enhanced.

[0053] Further, in step S35, the calculation result is subjected to a nonlinear projection transformation using the tanh function to obtain the final feature matrix, the mathematical model of which is: ; The tanh function enhances the dynamic range of the perturbation response region, ensuring that the details of the image are enhanced. When applied to the enhanced feature matrix, it can enhance the discernibility of information by performing nonlinear transformations on the features. To maintain the amplitude of the original signal and ensure the integrity of the original information, the final output retains the characteristics after perturbation guidance while ensuring the consistency of image information.

[0054] S4. Design a response residual and local information entropy calculation mechanism, introduce an adjustment factor, construct a reversible response entropy encoding module, and enhance the optimized cardiac ultrasound image through the reversible response entropy encoding module to generate a high-quality enhanced cardiac ultrasound image.

[0055] Furthermore, as shown in the appendix Figure 4 The reversible response entropy coding module described in S4 is constructed using the following steps: Figure 4 As shown, the specific implementation of the module includes the following steps: Furthermore, in step S41, to accurately assess the reliability of the image structure, a local reconstruction estimation of the cardiac ultrasound image is first constructed to estimate the reconstruction value of each pixel in its neighborhood. This reconstruction process uses median filtering to avoid the effects of noise interference and local inconsistencies. Its mathematical model is as follows: in In pixels The 3x3 neighborhood centered on the center, To obtain the set of neighboring pixels after removing the center pixel, a fixed 3×3 window is used for median filtering. This means the filtering operation considers the values ​​of each pixel and its eight neighboring pixels. This step, by removing local anomalies, provides a more stable basis for subsequent response calculations, resulting in a reconstructed estimate.

[0056] Further, in step S42, the response residual between each pixel location and its reconstructed value is calculated. It is used to reflect the differences and instabilities of image regions, and its mathematical model is as follows: ; in, To respond to the residual, when the residual value is larger, it indicates that there is obvious image content change in the pixel region, while when the residual value is smaller, it indicates that the region is relatively smooth and stable.

[0057] Further, in step S43, in order to measure the information amount of each pixel in its neighborhood, the local entropy is defined, which reflects the complexity and information density of the image region, and its mathematical model is: ; wherein, is the normalized probability distribution of the pixel values in the neighborhood, is the neighborhood set, which is a 3x3 window, represents the local entropy value of the pixel region, which can represent the complexity of the local region of the image, and a high entropy value indicates that the region is information-rich, and a low entropy value indicates that the region is information-poor.

[0058] Further, in step S44, the response entropy score of each pixel is calculated by combining the response residual and the local information entropy, which represents which information region the pixel belongs to, and its mathematical model is: ; wherein, is a very small constant, which is used to prevent division by zero error, and is set to 0.000001, and the response entropy score is used to distinguish the high information region and the low information region of the image, and for the low information region, the value is larger, while for the high information region, the value is smaller.

[0059] Further, in step S45, the key adjustment factor is constructed, which is used to dynamically adjust the intensity of image enhancement according to the reflection intensity and structural complexity of the color Doppler ultrasound image, and the adjustment factor is based on the reflection intensity change and the structural feature reconstruction error, so as to ensure that in the image enhancement process, the low information region is not over-enhanced, and the high information region is appropriately enhanced, and its mathematical model is: ; wherein, is the global reference adjustment intensity, which controls the overall intensity of enhancement; represents the ultrasound reflection intensity at the position , which reflects the reflection characteristics of the tissue in the image; and are the mean and standard deviation of the ultrasound reflection intensity map, respectively, which represent the global distribution of the reflection intensity in the image; is a very small constant, which is used to prevent division by zero error; represents the structural reconstruction error at the position , which reflects the detail complexity of the region, and in the embodiment, the structural reconstruction error The mathematical model is: ; is the average structure reconstruction error of the whole image, representing the global structure complexity of the image. In this embodiment, the average structure reconstruction error The mathematical model is: .

[0060] Further, step S46, constructing the final dehazing mask for controlling the enhancement intensity difference between the structure region and the low reflection region in the image enhancement process.

[0061] The mask generates an adjustable enhancement factor based on the local information of the image by combining the response entropy score and the adjustment factor , which strengthens the high reflection region and suppresses the excessive enhancement of the low reflection region. The mathematical model is: ; wherein, is the adjustment factor, which dynamically controls the image enhancement intensity and depends on the ultrasonic reflection intensity and the structure complexity. The detailed calculation is shown in step S45; is the response entropy score, which is used to measure the combination degree of the response residual and the local information entropy of each pixel, indicating whether the pixel belongs to a high information region. The specific calculation is shown in step S44; the mask For the high information region, including the edge, tissue interface, and blood vessels, the response entropy score is small, the adjustment factor is large, so the mask will remain a small value, reducing the excessive enhancement of the image; for the low reflection region, the response entropy score is large, the adjustment factor is small, so the mask will increase, enhancing the detail and structure performance of these regions; the mask obtained by calculation is used as the adjustment factor of image enhancement, automatically adjusting the enhancement intensity of different regions, ensuring that the structure information in the image is moderately enhanced, while avoiding the excessive enhancement of the low reflection region; the input weighted adjustment factor and the response entropy score are both 2D arrays, representing the feature matrix of the image.

[0062] Further, step S47, using the response entropy encoding mask to perform enhancement processing on the input image, obtaining the enhanced image , the purpose of this step is to automatically adjust the enhancement intensity according to the different reflection characteristics and structural complexity of the image region, to highlight the key information area in the image, while avoiding over-enhancing the low reflection area, the mathematical model is: ; wherein, is the pixel value of the enhanced image; for high reflection area, the enhanced image will maintain a relatively stable pixel value, avoiding over-enhancement; for low reflection area, the enhanced image will strengthen the details in these areas, enhance the contrast and structural clarity of the image; in this step, by weighting enhancement on the image, the enhanced image , the image has been emphasized in the structure area and suppressed in the background area, responding to the entropy coding mask plays a key role in it, so that the enhancement process can automatically adjust according to the characteristics of the image content, thereby improving the quality and recognizability of the image.

[0063] S5, integrate the signal excitation attention guiding module, the local perturbation divergence guiding mapping module and the reversibility response entropy coding module to build a heart color ultrasound image enhancement processing model.

[0064] Further, as described in S5 of the accompanying Figure 5 , build a heart color ultrasound image enhancement processing model, the specific construction steps of the model are shown in the accompanying Figure 5 , the specific implementation of the module includes the following steps: Further, step S51, first, format the input heart color ultrasound image for preprocessing, extract the features of the heart color ultrasound image and build a feature matrix, which is used for subsequent heart color ultrasound image enhancement processing to ensure that the heart color ultrasound image data meets the model input requirements.

[0065] Further, step S52, by constructing the signal excitation attention guiding module, based on the energy distribution of the heart color ultrasound image, the structural information in the heart color ultrasound image is enhanced. This module constructs a connection matrix by calculating the signal excitation intensity of each position in the heart color ultrasound image, and adjusts the connection strength through the response adjustment factor, so as to enhance the distinguishability of the key information area in the heart color ultrasound image.

[0066] Further, step S53, the local perturbation divergence guiding mapping module calculates the response difference of each region in the image through the perturbation generation mechanism, and adjusts the feature transmission strength of the heart color ultrasound image according to the perturbation response difference. By introducing the perturbation response difference matrix, the structural consistency of the heart color ultrasound image is optimized, the details of the target area are strengthened, and the interference of the low reflection area on the enhancement result is reduced.

[0067] Further, in step S54, the reversible response entropy coding module constructs an enhanced mask and finally outputs the enhanced cardiac color ultrasound image by calculating the response residual and the local information entropy. The module evaluates the image detail area according to the response residual and the information entropy score of the cardiac color ultrasound image area, and dynamically adjusts the cardiac color ultrasound image enhancement intensity in combination with the key adjustment factor, so as to optimize the quality and structure performance of the cardiac color ultrasound image.

[0068] Further, as shown in the cardiac color ultrasound image enhancement model shown in the accompanying Figure 1 Further, as shown in the cardiac color ultrasound image enhancement model shown in the accompanying Figure 6 Further, as shown in the cardiac color ultrasound image enhancement model shown in the accompanying Further, in step S6, the operating system platform used by the model is Linux system, the language is Python3.9.1, the processor used is Jetson Xavier, the image processing library used is OpenCV, PIL, the server hardware memory is 64G, and the data set contains a total of 500 cardiac color ultrasound images taken under different lighting conditions.

[0069] The above is only the preferred embodiment of the present application, and it should be pointed out that for those skilled in the art, without departing from the inventive concept, a number of modifications and improvements can be made, which are all within the scope of the present application.

Claims

1. A method for processing cardiac color Doppler ultrasound images based on enhancement processing, characterized in that, Includes the following steps: S1. Collect raw echocardiogram image data and generate an echocardiogram image dataset containing the target tissue region; S2. Introduce a feature matrix and construct a signal-excited guided attention module. The echocardiogram image data is structurally enhanced through the signal-excited guided attention module to generate an enhanced echocardiogram image. S3. Introduce local perturbation response difference calculation, construct local perturbation divergence guided mapping module, and process the enhanced cardiac ultrasound image through the local perturbation divergence guided mapping module to generate an optimized cardiac ultrasound image feature matrix. S4. Design a response residual and local information entropy calculation mechanism, introduce key adjustment factors, construct a reversible response entropy encoding module, and enhance the optimized cardiac ultrasound image through the reversible response entropy encoding module to generate a high-quality enhanced cardiac ultrasound image. S5. Integrating the signal excitation-guided attention module, the local perturbation divergence-guided mapping module, and the reversible response entropy coding module, a cardiac color Doppler ultrasound image enhancement processing model is constructed. S6. Input the echocardiogram image into the echocardiogram image enhancement processing model for image enhancement, and output a high-quality enhanced echocardiogram image.

2. The method for processing cardiac ultrasound images based on enhancement processing according to claim 1, characterized in that, In step S2, constructing the signal-induced guided attention module includes the following steps: S21. Define the signal excitation intensity function. Used to calculate the input feature matrix Each position The energy excitation intensity, mathematically modeled as follows: ； in, The total number of spatial locations, For the number of channels, For the first Position No. The eigenvalues ​​of the channel, As a signal excitation factor, it represents the location. The local energy response; S22. Introducing an energy difference-driven connection weighting function. The mathematical model describes the degree of similarity connectivity between any two locations and suppresses weakly correlated connections with large energy differences: ； in, For temperature parameters, The energy difference threshold, For indicator functions; S23. Define the structural response modulation factor This enhances the connection weights between high-energy locations and suppresses connections in low-energy regions, thereby improving the expressive power of structural features. The mathematical model is as follows: ； in, Energy excitation vector The maximum value, It is a non-linear adjustment index; S24, Connection matrix driven by energy difference With structural response adjustment factor matrix The combination of these elements constructs a structure-guided connection mapping matrix. The mathematical model is: ; S25 uses the structure to guide the connection mapping matrix For the input feature matrix Perform mapping operations to obtain the output feature matrix. .

3. The method for processing cardiac color Doppler ultrasound images based on enhancement processing according to claim 2, characterized in that, In step S3, constructing the local perturbation divergence guided mapping module includes the following steps: S31, Excitation and guidance feature matrix of the signal output from S2 Perturbation generation is performed to obtain the perturbation response matrix. This is done by introducing small perturbations at each location. Two perturbation versions were obtained. and Its mathematical model is: ; ； in The amplitude of the disturbance; Symbolic functions at the element level; S32, For each channel Define the variance of the disturbance response difference. Its mathematical model is: ； in, Indicates channel Sensitivity to disturbances and Positions Located in the The perturbation version of the channel.

4. The method for processing echocardiographic images based on enhancement processing according to claim 3, characterized in that, In step S3, constructing the local perturbation divergence guided mapping module includes the following steps: S301. To achieve perturbation-guided mapping, construct a structure-guided matrix. Its mathematical model is: ； in, Indicates position and location The degree of structural correlation between them; S302, Using the constructed structure guidance matrix For the input feature matrix The mapping is performed to obtain the final enhanced feature matrix. Its mathematical model is: ； S303. Use the tanh function to perform a nonlinear projection transformation on the calculation results to obtain the final feature matrix. Its mathematical model is: ； in, To preserve the amplitude of the original signal.

5. The method for processing cardiac color Doppler ultrasound images based on enhancement processing according to claim 4, characterized in that, In step S4, a reversible response entropy encoding module is constructed. Includes the following steps; S41. Construct a locally reconstructed and estimated echocardiogram image. Median filtering is used to avoid the effects of noise interference and local inconsistencies. The mathematical model is as follows: ； in In pixels The neighborhood centered on, The set of neighboring pixels after removing the center pixel. The reconstructed estimated value; S42. Calculate the response residual between each pixel location and its reconstructed value. Its mathematical model is: ； S43. Define local entropy Its mathematical model is: ； in, This represents the normalized probability distribution of pixel values ​​within the neighborhood. For neighborhood set, This represents the local entropy value of the pixel region; S44. Combining the response residual and local information entropy, calculate the response entropy score for each pixel. Its mathematical model is: ； in, It is a very small constant.

6. The method for processing cardiac color Doppler ultrasound images based on enhancement processing according to claim 5, characterized in that, In step S4, constructing the reversible response entropy encoding module includes the following steps: S401, Constructing Key Regulatory Factors Based on the changes in reflection intensity and the reconstruction error of structural features, its mathematical model is as follows: ； in, Adjust the intensity as a global benchmark. Indicates position Ultrasonic reflection intensity at the location, and These represent the mean and standard deviation of the ultrasound reflection intensity map, respectively. It is a very small constant. Indicates position Structural reconstruction error at the location, The average structure reconstruction error of the entire image; S402, Construct the final defogging mask. This mask is achieved by combining the response entropy score. and regulatory factors A modulotable enhancement factor based on local image information is generated, and its mathematical model is as follows: ； in, As a regulating factor, For response entropy scoring; S403, Use response entropy coding mask The input image is enhanced to obtain the enhanced image. Its mathematical model is: ； in, These are the raw pixel values ​​of the input image; The mask value is calculated based on the response entropy score and the adjustment factor; These are the enhanced image pixel values.

7. The method for processing cardiac color Doppler ultrasound images based on enhancement processing according to claim 6, characterized in that, In step S5, constructing the cardiac ultrasound image enhancement processing model includes the following steps: S51. Input the echocardiogram image dataset into the signal-guided attention module. Calculate the weight coefficient matrix based on the difference between local response intensity and global mean. Perform weighted processing on the input feature matrix to obtain the enhanced echocardiogram image. ; S52. Input the enhanced echocardiogram image into the local perturbation divergence guided mapping module. Introduce positive and negative perturbations of fixed amplitude at each pixel location, calculate the perturbation response difference, and standardize to generate a perturbation divergence weight matrix. Apply this weight matrix to feature optimization to obtain the optimized echocardiogram image feature matrix. ; S53. Optimize the feature matrix of the echocardiogram image. The reversible response entropy encoding module is input to calculate the pixel residual matrix and local information entropy. A key adjustment factor is introduced to correct the weights, completing the reversible encoding enhancement and obtaining a high-quality enhanced echocardiogram image. .

Images