Ultrasound echocardiogram image processing method, system, medium, and device

Abstract

The embodiment of the application discloses an echocardiogram image processing method, system, medium and equipment. The method comprises the following steps: performing image segmentation on a first image; performing phase analysis on the segmented image region to obtain an analysis region of an end diastole and an end systole; and finally, measuring the analysis region to obtain an image measurement result. The method avoids manual selection of the end diastole and the end systole and the introduction of manual errors, greatly reduces the influence of subjective factors on the measurement result, improves the accuracy of the measurement result, reduces the workload of doctors, and improves the measurement efficiency.

Description

Technical Field

[0001] This invention relates to the field of medical technology, and in particular to a method, system, medium, and device based on echocardiographic image processing. Background Technology

[0002] When examining a patient's heart, the parasternal left ventricular long-axis waveform measurement method is commonly used to measure the patient's echocardiogram. The measurement of the parasternal left ventricular long-axis waveform mainly includes measuring the interventricular septum thickness at end-diastole and end-systole, the left ventricular diameter, and the thickness of the left ventricular posterior wall.

[0003] The timing of end-diastolic and end-systolic measurements can reflect the heart's preload or afterload, and are of great significance for assessing left ventricular diastolic and systolic function and diagnosing heart diseases such as cardiomyopathy and heart failure.

[0004] The existing method for measuring the long-axis ventricular complex of the left ventricle parasternally first obtains a two-dimensional echocardiogram. Then, a professional physician locates the end-diastolic and end-systolic phases and measures the interventricular septum thickness, left ventricular diameter, and left ventricular posterior wall thickness at these phases. Based on the measurement results, the physician determines the specific condition of the patient's heart and makes a diagnosis accordingly.

[0005] However, existing measurement methods rely on manual measurement by professional physicians. The results of manual measurement are affected by subjective factors, which leads to errors. Furthermore, the measurement process requires repeated measurements, resulting in a very large workload and very low efficiency.

[0006] Secondly, since manual measurement relies mainly on the doctor's experience, different doctors may make different judgments during the measurement process, which can easily lead to misdiagnosis in the diagnosis of a patient's heart. Once a misdiagnosis occurs, firstly, the patient may not receive timely treatment, and secondly, the treatment direction may be incorrect, affecting the patient's health. Summary of the Invention

[0007] Based on this, it is necessary to address the above-mentioned problems. This application proposes a method, system, medium, and device based on echocardiographic image processing to avoid errors caused by manual measurement of the interventricular septum thickness at end-diastole and end-systole, the left ventricular diameter, and the left ventricular posterior wall thickness. This reduces the influence of subjective factors on measurement results and improves the accuracy of measurement results.

[0008] This invention provides a method based on echocardiographic image processing, the method comprising:

[0009] Acquire echocardiogram images to be detected, and generate a first image based on the echocardiogram images;

[0010] The first image is segmented to obtain segmented image regions, which include: the interventricular septum region, the left ventricle region, and the posterior wall region of the left ventricle.

[0011] The end-diastolic and end-systolic phases in the segmented image region are analyzed by temporal analysis to obtain the analysis region, and the analysis region is measured to obtain the image measurement results.

[0012] In at least one embodiment of this application, the specific steps of acquiring the echocardiogram image to be detected and generating a first image based on the echocardiogram image further include:

[0013] The echocardiogram image is marked with M-shaped sampling lines to generate the first image;

[0014] The first image is segmented to obtain segmented image regions;

[0015] Temporal analysis is performed on the segmented image region to obtain the analysis region corresponding to the maximum and minimum time of the left ventricular intra-aortic diameter, and the analysis region is measured to obtain image measurement results.

[0016] In at least one embodiment of this application, the method further includes:

[0017] The first image is filtered and denoised to obtain the second image;

[0018] The second image is segmented to obtain segmented image regions.

[0019] In at least one embodiment of this application, the specific steps of filtering and denoising the first image to obtain the second image include:

[0020] A two-dimensional Log-Gabor filter is used to perform a first filtering process on the first image to extract the ultrasonic phase features of the first image;

[0021] The first image was denoised using speckle suppression anisotropic diffusion filtering to obtain a denoised echocardiogram.

[0022] The denoised echocardiogram image is multiplied by the ultrasound phase feature to obtain a filtered and denoised image.

[0023] The filtered and denoised image is processed using bilateral filtering to distinguish the left ventricular region from the interventricular septum region and the left ventricular posterior wall region, thereby obtaining the second image after enhanced processing and differentiation.

[0024] In at least one embodiment of this application, the specific steps of using speckle suppression anisotropic diffusion filtering to denoise the first image to obtain a denoised echocardiogram include:

[0025] Obtain the gradient and local variance of the first image;

[0026] The diffusion coefficient of the speckle suppression anisotropic diffusion filter is determined based on the gradient and local variance of the first image.

[0027] In at least one embodiment of this application, the step of performing image segmentation on the second image to obtain segmented image regions includes:

[0028] The second image is segmented using adaptive thresholding to initially segment the interventricular septum region, the left ventricular region, and the posterior wall region of the left ventricle.

[0029] In at least one embodiment of this application, the method further includes:

[0030] The preliminary segmented region is refined by using distance regularization level set evolution to obtain a precisely segmented image region.

[0031] A system based on echocardiographic image processing, the system comprising:

[0032] The acquisition module is used to acquire the echocardiogram images to be detected.

[0033] The image generation module generates a first image based on the echocardiogram image;

[0034] The image segmentation module segments the first image to obtain segmented image regions;

[0035] The temporal analysis module analyzes the segmented image regions to obtain the analysis region;

[0036] The measurement module measures the analysis area to obtain image measurement results;

[0037] The echocardiogram processing system performs the following steps:

[0038] Acquire echocardiogram images to be detected, and generate a first image based on the echocardiogram images;

[0039] The first image is segmented to obtain segmented image regions, which include: the interventricular septum region, the left ventricle region, and the posterior wall region of the left ventricle.

[0040] The end-diastolic and end-systolic phases in the segmented image region are analyzed by temporal analysis to obtain the analysis region, and the analysis region is measured to obtain the image measurement results.

[0041] A computer device includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the following steps:

[0042] Acquire echocardiogram images to be detected, and generate a first image based on the echocardiogram images;

[0043] The first image is segmented to obtain segmented image regions, which include: the interventricular septum region, the left ventricle region, and the posterior wall region of the left ventricle.

[0044] The end-diastolic and end-systolic phases in the segmented image region are analyzed by temporal analysis to obtain the analysis region, and the analysis region is measured to obtain the image measurement results.

[0045] A computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the following steps:

[0046] Acquire echocardiogram images to be detected, and generate a first image based on the echocardiogram images;

[0047] The first image is segmented to obtain segmented image regions, which include: the interventricular septum region, the left ventricle region, and the posterior wall region of the left ventricle.

[0048] The end-diastolic and end-systolic phases in the segmented image region are analyzed by temporal analysis to obtain the analysis region, and the analysis region is measured to obtain the image measurement results.

[0049] Implementing the method, system, medium, and device based on echocardiographic image processing of the present invention will have at least the following beneficial effects:

[0050] This invention provides a method, system, medium, and device based on echocardiographic image processing. By segmenting a first image and then using temporal analysis to analyze the segmented image regions, the analysis areas of end-diastole and end-systole are obtained. Finally, the analysis areas are measured to obtain image measurement results. This avoids the human error introduced by manually selecting end-diastole and end-systole, greatly reduces the influence of subjective factors on the measurement results, improves the accuracy of the measurement results, reduces the workload of doctors, and improves the efficiency of measurement. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] in:

[0053] Figure 1 This is a flowchart of an embodiment of a method based on echocardiographic image processing;

[0054] Figure 2 for Figure 1 A flowchart of another embodiment based on an echocardiogram image processing method;

[0055] Figure 3 for Figure 2 A detailed flowchart of the method based on echocardiographic image processing in China;

[0056] Figure 4 This is a block diagram of a system based on echocardiography image processing in one embodiment;

[0057] Figure 5 This is a structural block diagram of a computer device in one embodiment.

[0058] 100. Based on echocardiogram image processing system; 110. Acquisition module; 120. Image generation module; 130. Image segmentation module; 140. Temporal analysis module; 150. Measurement module. Detailed Implementation

[0059] 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.

[0060] This invention provides a method based on echocardiographic image processing, the method comprising:

[0061] S101. Acquire the echocardiogram image to be detected, and generate a first image based on the echocardiogram image;

[0062] S102. Perform image region segmentation on the first image to obtain segmented image regions, wherein the segmented image regions include: interventricular septum region, left ventricle region, and left ventricular posterior wall region.

[0063] S103. Analyze the end-diastolic and end-systolic phases in the segmented image region through temporal analysis to obtain the analysis region, and measure the analysis region to obtain image measurement results.

[0064] Please refer to Figure 1 In this embodiment, after the system acquires a standard parasternal left ventricular long-axis section two-dimensional image from the probe of an external ultrasound device, it generates an echocardiogram based on the standard parasternal left ventricular long-axis section two-dimensional image, and then generates a first image based on the echocardiogram. In one embodiment, the first image is an M-mode echocardiogram.

[0065] The system performs image segmentation on the first image to automatically segment the interventricular septum region, left ventricular region, and left ventricular posterior wall, etc., to obtain the segmented image region.

[0066] Then, the system automatically locates the end-diastolic and end-systolic images in the segmented image region through temporal analysis, filters out the end-diastolic and end-systolic images, marks the images in the end-diastolic and end-systolic phases as the analysis region, measures the interventricular septal thickness, left ventricular diameter, and left ventricular posterior wall thickness parameters at the end-diastolic and end-systolic phases, and generates image measurement results based on the measured parameters.

[0067] The measurement results can be used to evaluate the overall cardiac function, provide reference data for the diagnosis and analysis of cardiac diseases, avoid errors caused by manual measurement of echocardiograms, solve the influence of subjective factors caused by existing manual measurements, and improve the efficiency and accuracy of measurement results.

[0068] It should be noted that during the acquisition of echocardiogram images, the patient should be in a supine or slightly left lateral decubitus position with their head tilted slightly to the left to bring the heart closer to the chest wall. The probe should be placed in the anterior chest wall region of the 3rd-4th intercostal space next to the left sternal border, with the marker point facing the right shoulder. The detection plane should be roughly level with the line connecting the right shoulder to the left rib. The ultrasound probe beam should be pointed towards the patient's back. The probe should be gently tilted upwards or downwards to ensure that the sound beam passes through the long axis of the left ventricle. This method can locate the long axis section of the left ventricle next to the sternum.

[0069] This allows the acquired echocardiograms to simultaneously observe the left atrium, left ventricle, mitral valve, left ventricular posterior wall, and interventricular septum, ensuring that the mitral valve and the inner cavity of the left ventricle are clearly displayed in the images, and that the ventricular wall and interventricular septum exhibit good contrast.

[0070] It should be further explained that by using an external electrocardiogram (ECG) to obtain an ECG, and then analyzing the positions of the corresponding R and T waves in the ECG, the end-diastolic and end-systolic phases of the heart can be located. Furthermore, by measuring the timing of the end-diastolic and end-systolic phases in the echocardiogram, the corresponding interventricular septum thickness, left ventricular diameter, and left ventricular posterior wall thickness can be determined, thus providing data reference for subsequent diagnosis, treatment, monitoring, and screening.

[0071] In at least one embodiment of this application, the specific steps of acquiring the echocardiogram image to be detected and generating a first image based on the echocardiogram image further include:

[0072] S201. Mark the echocardiogram with M-shaped sampling lines to generate the first image;

[0073] S202. Perform image segmentation on the first image to obtain segmented image regions;

[0074] S203. Perform temporal analysis on the segmented image region to obtain the analysis region corresponding to the maximum and minimum time of the left ventricular intra-aortic diameter, and measure the analysis region to obtain image measurement results.

[0075] Please refer to Figures 2-3 In this embodiment, after acquiring an echocardiogram (i.e., a two-dimensional image of the left ventricle along its long axis), an M-shaped sampling line is placed on the echocardiogram image (the M-shaped sampling line is placed using a manual marking method), so that the sampling line is placed at the level of the chordae tendineae and perpendicular to the interventricular septum and the posterior wall of the left ventricle to generate a first image, which is the M-shaped echocardiogram. The M-shaped echocardiogram image is then segmented to automatically segment the interventricular septum region, the left ventricle region, and the posterior wall of the left ventricle in the image, thereby obtaining the segmented image region.

[0076] Generally, the left ventricular intraventricular diameter is at its maximum, corresponding to the end of diastole in the cardiac cycle, and the left ventricular intraventricular diameter is at its minimum, corresponding to the end of systole in the cardiac cycle. Temporal analysis is performed on the segmented image regions to select the image regions at the moments of maximum and minimum left ventricular intraventricular diameter in vertical distance. Based on the regions corresponding to the moments of maximum and minimum left ventricular intraventricular diameter (i.e., the regions corresponding to end-diastole and end-systole), the analysis region is obtained.

[0077] The analysis area is measured to determine the interventricular septum thickness, left ventricular diameter, and left ventricular posterior wall thickness at end-systole and end-diastole, thus providing data reference for subsequent diagnosis, treatment, monitoring, and screening.

[0078] Based on the automatically located end-diastolic and end-systolic times, the interventricular septum thickness, left ventricular diameter, and left ventricular posterior wall thickness at the corresponding times can be measured in the segmented first image, providing more accurate measurement results. Simultaneously, the method automatically segments, filters, and measures the images, avoiding errors caused by manual measurement of these parameters at end-diastolic and end-systolic times. This reduces the influence of subjective factors on measurement results and improves accuracy. Furthermore, the automatic extraction of feature region images and measurement results significantly improves measurement efficiency.

[0079] In at least one embodiment of this application, the method further includes:

[0080] S301. Filter and denoise the first image to obtain the second image;

[0081] S302. Perform image segmentation on the second image to obtain segmented image regions.

[0082] In this embodiment, after filtering and denoising the first image, a second image is obtained. By reducing noise and artifacts in the left ventricle, the contrast between the ventricular wall and the interventricular septum is enhanced, making subsequent image segmentation easier. Furthermore, the edge regions and image details in the image are improved to further enhance the accuracy of the image segmentation process.

[0083] In at least one embodiment of this application, the specific steps of filtering and denoising the first image to obtain the second image include:

[0084] S303. The first image is subjected to a first filtering process using a two-dimensional Log-Gabor filter to extract the ultrasonic phase features of the first image;

[0085] S304. The first image is denoised by speckle suppression anisotropic diffusion filtering to obtain a denoised echocardiogram.

[0086] S305. Multiply the denoised echocardiogram with the ultrasound phase feature to obtain a filtered and denoised image.

[0087] S306. The filtered and denoised image is processed by bilateral filtering to distinguish the left ventricular region from the interventricular septum region and the left ventricular posterior wall region, so as to obtain the second image after enhanced processing and differentiation.

[0088] In this embodiment, during filtering, the presence of noise or artifacts in the first image can lead to low accuracy in image segmentation. Therefore, a two-dimensional Log-Gabor filter is first used to perform multi-scale and multi-directional filtering on the first image to extract the ultrasonic phase features of the first image.

[0089] A two-dimensional Log-Gabor filter is used to perform multi-scale and multi-directional first filtering processing on the first image to enhance and filter the image, thereby reducing noise and artifacts in the left ventricular cavity image region, thereby enhancing the contrast between the ventricular wall and the interventricular septum, and enhancing the edge contour between the ventricular wall and the interventricular septum, so as to facilitate subsequent image segmentation between the ventricular wall and the interventricular septum.

[0090] The two-dimensional Log-Gabor filter is as follows:

[0091]

[0092] Where W is the frequency, corresponding to multi-scale features; denoted by angle, corresponding to multi-directional characteristics; k is the bandwidth of the filter; W0 is the center frequency of the filter; The direction of the filter; This refers to the directional bandwidth parameter.

[0093] The first image is denoised using a speckle-suppressed anisotropic diffusion filter, which preserves the edges and details of the image.

[0094] Among them, the speckle suppression anisotropic diffusion filter is:

[0095] Where c(x, y, t) is the diffusion coefficient.

[0096] Where I(x, y, t+Δt) is the image grayscale value at position (x, y) and time t; This represents the gradient operation; diV() represents the divergence operation.

[0097] Simultaneously, speckle noise present in the image is removed to eliminate speckle noise in the first image, thereby avoiding the problem of image segmentation accuracy being affected by the presence of speckle noise during subsequent image segmentation, and thus ensuring the accuracy of the measurement results.

[0098] The image after anisotropic diffusion filtering and noise reduction is multiplied by the phase features obtained by filtering with a two-dimensional Log-Gabor filter to reduce noise and artifacts in the first image, while enhancing the structural information of the edges in the first image.

[0099] The system employs bilateral filtering to process the filtered and denoised image. Based on the anechoic, hypoechoic, and hyperechoic regions in the filtered and denoised image, the interventricular septum and the thickness of the left ventricular posterior wall are distinguished from other regions. Since the content displayed in the first image, from top to bottom, consists of the right ventricular anterior wall, interventricular septum, left ventricular cavity, and left ventricular posterior wall, the left ventricle, interventricular septum, and left ventricular posterior wall are then distinguished through grayscale differences and the aforementioned location regions to obtain a second image after enhancement processing, thereby differentiating the different regions in the first image.

[0100] In at least one embodiment of this application, the specific steps of using speckle suppression anisotropic diffusion filtering to denoise the first image to obtain a denoised echocardiogram include:

[0101] S307. Obtain the gradient and local variance of the first image;

[0102] S308. Determine the diffusion coefficient of the speckle suppression anisotropic diffusion filter based on the gradient and local variance of the first image.

[0103] In this embodiment, the system obtains the image gradient and local variance of the first image, and calculates the diffusion coefficient based on the gradient and local variance. The diffusion coefficient is calculated as follows:

[0104] in, The gradient of the M-mode echocardiogram, V(x, y, t) / C0). -1 The local variance of the M-mode cardiac image.

[0105] in, Let V(x, y, t) be the gradient magnitude of the image at position (x, y) and time t; let V(x, y, t) be the local variance of the image at position (x, y); and let K and C0 be the parameters that control the degree of gradient and variance diffusion, respectively.

[0106] Traditional diffusion coefficients are calculated based solely on the image gradient, without considering the influence of local image variance. Furthermore, the gradient only reflects the rate of change of image pixel values ​​and cannot distinguish whether these changes are caused by noise or by the image's texture or structure.

[0107] Within high gradient regions, areas with large gradients are considered edges, resulting in a reduced diffusion coefficient and protecting these regions. However, in the presence of noise, some noise points may also have large gradients, leading to misjudgments where noise is treated as important edges, thus suppressing diffusion.

[0108] In low-gradient regions, areas with small gradients are perceived as flat and have a large diffusion coefficient. This can lead to the destruction of texture or important details in the image during the smoothing process.

[0109] This can cause speckle noise to generate high gradients in a local area. Gradient judgment may misclassify these noises as edges and suppress their diffusion, resulting in the noise not being effectively removed.

[0110] Gradients in complex texture regions may be small, but local variance is high. Relying solely on gradients may result in over-smoothing of these texture regions, destroying texture information.

[0111] This results in poor filtering performance, which in turn affects the accuracy of image segmentation and ultimately leads to large measurement errors.

[0112] In this embodiment, the diffusion coefficient is calculated by combining gradient and local variance. The improved diffusion coefficient is not only based on the gradient of the image, but also on the local variance of the image. The diffusion coefficient is calculated by using gradient and local variance to achieve a better balance between image denoising and detail preservation, thereby better adjusting the diffusion intensity and improving the effect of diffusion filtering.

[0113] Therefore, the use of gradient combined with local variance in the anisotropic diffusion filtering for speckle suppression can effectively remove the influence of speckle noise in the image while ensuring the preservation of image edges and details. This enhances edge structure information and enables more accurate segmentation of different regions in the first image during image segmentation, thereby obtaining a precisely segmented image and ensuring the accuracy of the measurement results.

[0114] In at least one embodiment of this application, the step of performing image segmentation on the second image to obtain segmented image regions includes:

[0115] S309. Adaptive threshold segmentation is used to segment the second image to segment the interventricular septum region, the left ventricular region, and the posterior wall region of the left ventricular region into preliminary segmented regions.

[0116] In this embodiment, the system uses an adaptive threshold to perform preliminary segmentation on the second image. Since the second image has been preliminarily distinguished from the different regions in the first image after bilateral filtering, the adaptive threshold segmentation can perform preliminary segmentation on the second image to roughly segment the interventricular septum, left ventricle, and posterior wall of the left ventricle in the first image, thus obtaining the preliminary segmentation region.

[0117] In at least one embodiment of this application, the method further includes:

[0118] S310. The preliminary segmented region is refined by using distance regularization level set evolution to obtain a precisely segmented image region.

[0119] In this embodiment, after the system performs initial segmentation of the second image using an adaptive threshold, the system further refines the segmented region using a distance-regularized level set evolution segmentation method. The threshold segmentation results of the interventricular septum, left ventricle, and left ventricular posterior wall are used to generate initial level set functions. Then, the system uses the distance-regularized level set evolution segmentation method to segment the precise interventricular septum, left ventricle, and left ventricular posterior wall regions to obtain accurately segmented image regions.

[0120] The energy function of the evolution curve in the distance-regularized level set evolution is:

[0121]

[0122] Among them, R p Here, μ is the regularization term, and μ is the regularization weight parameter. External energy function; The term is the weighted length term, and λ is the control parameter. Here, α is the weighted area term, and α is the control parameter.

[0123] The partial differential equation in the energy function of the evolution curve is:

[0124]

[0125] Where p is the energy density function;

[0126]

[0127] g(I) is the edge indicator function:

[0128]

[0129] Preferably, in the process of performing distance regularization level evolution segmentation, in order to accelerate the rate of level set contour evolution, only the values ​​near the contour can be updated, without updating the contour values ​​of the entire image, so as to improve the efficiency of segmentation.

[0130] Among them, G σ It is a Gaussian function with variance σ; * represents convolution operation; ρ is a parameter that controls the convergence rate of the curve.

[0131] Preferably, when performing time-domain analysis based on the first image:

[0132] The system first acquires an echocardiogram (i.e., a two-dimensional image of the left ventricle in the long axis section beside the sternum). Then, an M-shaped sampling line is placed on the echocardiogram image (the M-shaped sampling line is placed by manual marking) so that the sampling line is at the level of the chordae tendineae and perpendicular to the interventricular septum and the posterior wall of the left ventricle to generate the first image (the first image here is an M-shaped echocardiogram).

[0133] The system then applies a two-dimensional Log-Gabor filter to the first image to perform image enhancement and filtering, thereby obtaining ultrasonic phase features.

[0134] The system then uses speckle suppression anisotropic diffusion filtering to denoise the first image, obtaining an anisotropic diffusion filtering denoised echocardiogram.

[0135] The system performs a dot product between the anisotropic diffusion-filtered and denoised echocardiogram and the ultrasound phase features to obtain the filtered and denoised image.

[0136] The system uses bilateral filtering to process the filtered and denoised image, obtaining a second image after enhancement and differentiation.

[0137] During the segmentation process, the system first uses an adaptive threshold segmentation method to segment the second image, so as to segment the interventricular septum region, the left ventricular region, and the posterior wall region of the left ventricular region into preliminary segmentation areas.

[0138] Then, distance regularization level set evolution is used to refine the initial segmentation region, so as to accurately segment the interventricular septum, left ventricle, and left ventricular posterior wall region, and obtain the accurately segmented image region.

[0139] Finally, the system performs temporal analysis on the segmented image regions, and selects the image regions at the moments when the left ventricular region has the largest and smallest diameter in terms of vertical distance. Based on the regions corresponding to the moments when the left ventricular region has the largest and smallest diameter, the analysis region is obtained.

[0140] The analysis area is measured to obtain the corresponding interventricular septum thickness, left ventricular diameter, and left ventricular posterior wall thickness, thereby providing data reference for subsequent diagnosis, treatment, monitoring, and screening.

[0141] An echocardiogram image processing system 100, the system 100 comprising:

[0142] Acquisition module 110 is used to acquire the echocardiogram image to be detected;

[0143] Image generation module 120 generates a first image based on echocardiogram images;

[0144] Image segmentation module 130 segments the first image to obtain segmented image regions;

[0145] The temporal analysis module 140 analyzes the segmented image regions to obtain the analysis region;

[0146] Measurement module 150 measures the analysis area to obtain image measurement results;

[0147] The echocardiogram processing system performs the following steps:

[0148] Acquire echocardiogram images to be detected, and generate a first image based on the echocardiogram images;

[0149] The first image is segmented to obtain segmented image regions, which include: the interventricular septum region, the left ventricle region, and the posterior wall region of the left ventricle.

[0150] The end-diastolic and end-systolic phases in the segmented image region are analyzed by temporal analysis to obtain the analysis region, and the analysis region is measured to obtain the image measurement results.

[0151] Please refer to Figure 4 In this embodiment, the system 100 acquires the echocardiogram image to be detected through the acquisition module 110, and then the system 100 generates a first image based on the echocardiogram image through the image generation module 120. The system 100 performs image region segmentation on the first image according to the image segmentation module 130 to obtain the segmented image region.

[0152] Finally, the system 100 analyzes the end-diastolic and end-systolic phases in the segmented image region through the phase analysis module 140 to select the end-diastolic and end-systolic regions from the segmented image region to obtain the analysis region. The measurement module 150 measures the interventricular septal thickness, left ventricular diameter, and left ventricular posterior wall thickness parameters in the analysis region during the end-diastolic and end-systolic phases to obtain the image measurement results.

[0153] The measurement results can be used to evaluate the overall cardiac function, provide reference data for the diagnosis and analysis of cardiac diseases, avoid errors caused by manual measurement of echocardiograms, solve the influence of subjective factors caused by existing manual measurements, and improve the efficiency and accuracy of measurement results.

[0154] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the following steps:

[0155] Acquire echocardiogram images to be detected, and generate a first image based on the echocardiogram images;

[0156] The first image is segmented to obtain segmented image regions, which include: the interventricular septum region, the left ventricle region, and the posterior wall region of the left ventricle.

[0157] The end-diastolic and end-systolic phases in the segmented image region are analyzed by temporal analysis to obtain the analysis region, and the analysis region is measured to obtain the image measurement results.

[0158] In this embodiment, the system acquires the echocardiogram image to be detected through the acquisition module, then generates a first image based on the echocardiogram image, and performs image region segmentation on the first image according to the image segmentation module to obtain the segmented image region.

[0159] Finally, the system analyzes the end-diastolic and end-systolic phases in the segmented image region through temporal analysis to select the end-diastolic and end-systolic regions from the segmented image region, obtain the analysis region, and measure the interventricular septal thickness, left ventricular diameter, and left ventricular posterior wall thickness parameters in the analysis region during the end-diastolic and end-systolic phases to obtain the image measurement results.

[0160] The measurement results can be used to evaluate the overall cardiac function, provide reference data for the diagnosis and analysis of cardiac diseases, avoid errors caused by manual measurement of echocardiograms, solve the influence of subjective factors caused by existing manual measurements, and improve the efficiency and accuracy of measurement results.

[0161] Figure 5 An internal structural diagram of a computer device in one embodiment is shown. This computer device can specifically be a terminal or a server. Figure 5 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and may also store a computer program that, when executed by the processor, enables the processor to implement an echocardiogram image processing method. The internal memory may also store a computer program that, when executed by the processor, enables the processor to perform an echocardiogram image processing method. Those skilled in the art will understand that… Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0162] In one embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, causes the processor to perform the following steps:

[0163] Acquire echocardiogram images to be detected, and generate a first image based on the echocardiogram images;

[0164] The first image is segmented to obtain segmented image regions, which include: the interventricular septum region, the left ventricle region, and the posterior wall region of the left ventricle.

[0165] The end-diastolic and end-systolic phases in the segmented image region are analyzed by temporal analysis to obtain the analysis region, and the analysis region is measured to obtain the image measurement results.

[0166] In this embodiment, the system acquires the echocardiogram image to be detected through the acquisition module, then generates a first image based on the echocardiogram image, and performs image region segmentation on the first image according to the image segmentation module to obtain the segmented image region.

[0167] Finally, the system analyzes the end-diastolic and end-systolic phases in the segmented image region through temporal analysis to select the end-diastolic and end-systolic regions from the segmented image region, obtain the analysis region, and measure the interventricular septal thickness, left ventricular diameter, and left ventricular posterior wall thickness parameters in the analysis region during the end-diastolic and end-systolic phases to obtain the image measurement results.

[0168] The measurement results can be used to evaluate the overall cardiac function, provide reference data for the diagnosis and analysis of cardiac diseases, avoid errors caused by manual measurement of echocardiograms, solve the influence of subjective factors caused by existing manual measurements, and improve the efficiency and accuracy of measurement results.

[0169] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0170] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0171] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method based on echocardiographic image processing, characterized in that, The method includes: Acquire echocardiogram images to be detected, and generate a first image based on the echocardiogram images; The first image is segmented to obtain segmented image regions, which include: the interventricular septum region, the left ventricle region, and the posterior wall region of the left ventricle. The end-diastolic and end-systolic phases in the segmented image region are analyzed by temporal analysis to obtain the analysis region, and the analysis region is measured to obtain the image measurement results.

2. The method based on echocardiographic image processing according to claim 1, characterized in that, The specific steps of acquiring the echocardiogram image to be detected and generating a first image based on the echocardiogram image further include: The echocardiogram image is marked with M-shaped sampling lines to generate the first image; The first image is segmented to obtain segmented image regions; Temporal analysis is performed on the segmented image region to obtain the analysis region corresponding to the maximum and minimum time of the left ventricular intra-aortic diameter, and the analysis region is measured to obtain image measurement results.

3. The method based on echocardiographic image processing according to claim 2, characterized in that, The method further includes: The first image is filtered and denoised to obtain the second image; The second image is segmented to obtain segmented image regions.

4. The method based on echocardiographic image processing according to claim 3, characterized in that, The specific steps for filtering and denoising the first image to obtain the second image include: A two-dimensional Log-Gabor filter is used to perform a first filtering process on the first image to extract the ultrasonic phase features of the first image; The first image was denoised using speckle suppression anisotropic diffusion filtering to obtain a denoised echocardiogram. The denoised echocardiogram image is multiplied by the ultrasound phase feature to obtain a filtered and denoised image. The filtered and denoised image is processed using bilateral filtering to distinguish the left ventricular region from the interventricular septum region and the left ventricular posterior wall region, thereby obtaining the second image after enhanced processing and differentiation.

5. The method based on echocardiographic image processing according to claim 4, characterized in that, The specific steps for denoising the first image using speckle suppression anisotropic diffusion filtering to obtain a denoised echocardiogram include: Obtain the gradient and local variance of the first image; The diffusion coefficient of the speckle suppression anisotropic diffusion filter is determined based on the gradient and local variance of the first image.

6. The method based on echocardiographic image processing according to claim 3, characterized in that, The step of performing image segmentation on the second image to obtain segmented image regions includes: The second image is segmented using adaptive thresholding to initially segment the interventricular septum region, the left ventricular region, and the posterior wall region of the left ventricle.

7. The method based on echocardiographic image processing according to claim 6, characterized in that, The method further includes: The preliminary segmented region is refined by using distance regularization level set evolution to obtain a precisely segmented image region.

8. A system based on echocardiographic image processing, characterized in that, The system includes: The acquisition module is used to acquire the echocardiogram images to be detected. The image generation module generates a first image based on the echocardiogram image; The image segmentation module segments the first image to obtain segmented image regions; The temporal analysis module analyzes the segmented image regions to obtain the analysis region; The measurement module measures the analysis area to obtain image measurement results.

9. A computer-readable storage medium, characterized in that, The device stores a computer program that, when executed by a processor, causes the processor to perform the steps of the method as described in any one of claims 1 to 7.

10. A computer device, characterized in that, It includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the method as described in any one of claims 1 to 7.

Images