Cardiac ultrasound image processing method, ultrasound equipment and storage medium

Abstract

The embodiment of the invention provides a cardiac ultrasound image processing method, ultrasound equipment and a storage medium. The cardiac ultrasound image processing method comprises the following steps: acquiring a cardiac ultrasound image; determining the section type of the cardiac ultrasound image by using the image classification model; according to the determined section type, image segmentation is carried out on the cardiac ultrasound image to determine position information of a tissue structure contained in the cardiac ultrasound image, and the tissue structure comprises at least one target structure; and determining a target area in the cardiac ultrasound image for the at least one target structure according to the position information of the tissue structure. According to the method, the cardiac ultrasound images are segmented according to the image characteristics of the tissue structures in the cardiac ultrasound images of different section types, so that the image segmentation precision is effectively ensured, and the accuracy of the determined target area is further ensured. The scheme is suitable for heart ultrasonic images of various section types, the universality is high, and a user can conveniently and rapidly carry out comprehensive ultrasonic detection on the heart.

Description

Technical Field

[0001] This invention relates to the field of image processing technology. More specifically, it relates to a method for processing cardiac ultrasound images, an ultrasound device, a storage medium, and a computer program product. Background Technology

[0002] With the development of ultrasound technology, ultrasound examination has been widely used in medicine. In cardiac ultrasound examinations, users typically focus on key target areas in the ultrasound image, depending on the different tissue locations within the heart. This ensures precise definition of the analysis range, filters effective signals, and guarantees the specificity, accuracy, and relevance of blood flow or functional parameter measurements.

[0003] For example, color Doppler and pulsed Doppler are commonly used operating modes in ultrasound examination. Color Doppler images can be used to show the blood flow distribution in a target area. Specifically, it uses color coding to display the direction, velocity, and distribution of blood flow in the blood vessels inside the human body in real time, providing hemodynamic information for disease diagnosis. Thus, for color Doppler, the spatial range of color blood flow imaging is defined by the target area. Pulsed Doppler images can be used to show the blood flow spectrum in a target area. Utilizing the Doppler effect, the blood flow velocity-time curve within a sampling volume at a specific depth can be obtained.

[0004] During ultrasound Doppler imaging, users typically manually identify the target area in the ultrasound image. However, manually identifying the target area is complex, time-consuming, and relies on human experience. Summary of the Invention

[0005] The present invention was proposed in view of the above-mentioned problems.

[0006] According to one aspect of the present invention, a method for processing cardiac ultrasound images is provided. The method for processing cardiac ultrasound images includes:

[0007] Acquire cardiac ultrasound images;

[0008] Using an image classification model, the section type of cardiac ultrasound images is determined;

[0009] Based on the determined section type, the cardiac ultrasound image is segmented to determine the location information of the tissue structures contained in the cardiac ultrasound image, wherein the tissue structures include at least one target structure.

[0010] Based on the location information of the tissue structure, a target region is determined in the cardiac ultrasound image for at least one target structure.

[0011] For example, the location information includes multiple positioning points of the tissue structure and location information of the contour of at least one tissue structure, and the target region includes a color Doppler sampling region. Based on the location information of the tissue structure, determining the target region in the echocardiogram for at least one target structure includes: determining the boundary position of the region of interest of each target structure in the echocardiogram based on the imaging origin of the echocardiogram and the location information of multiple positioning points of the tissue structure and the contour of at least one tissue structure contained in the echocardiogram; and determining the color Doppler sampling region of each target structure in the at least one target structure based on the boundary position of the region of interest of that target structure.

[0012] For example, based on the imaging origin of a cardiac ultrasound image, and according to the positional information of multiple positioning points of tissue structures included in the cardiac ultrasound image and the contour of at least one tissue structure, the boundary position of the region of interest (ROI) of each target structure in the cardiac ultrasound image is determined, including: for each ROI of the at least one target structure, calculating the positional information of multiple boundary reference lines of the ROI based on the imaging origin and the positional information of multiple positioning points of tissue structures included in the cardiac ultrasound image and the contour of at least one tissue structure, wherein the multiple boundary reference lines include a left boundary reference line, a right boundary reference line, an upper boundary reference line, and a lower boundary reference line, the left boundary reference line and the right boundary reference line are rays passing through the imaging origin respectively, and the upper boundary reference line and the lower boundary reference line are arcs centered at the imaging origin; and determining the boundary position of the ROI based on the positional information of the multiple boundary reference lines.

[0013] For example, the calculation of the position information of multiple boundary reference lines of the region of interest based on the position information of multiple positioning points of the tissue structure contained in the imaging origin and the echocardiogram image and the contour of at least one tissue structure includes: calculating the position information of the boundary reference points of the region of interest based on the position information of multiple positioning points of the tissue structure contained in the imaging origin and the echocardiogram image and the contour of at least one tissue structure; and for each boundary reference point, determining the position information of the boundary reference line corresponding to the region of interest based on the position information of the boundary reference point.

[0014] For example, based on the location information of multiple boundary reference lines of the region of interest, including the imaging origin and multiple location points of the tissue structure contained in the cardiac ultrasound image and the location information of the contour of at least one tissue structure, the location information of multiple boundary reference lines of the region of interest is calculated, including: determining the location information of the tangent line that passes through the imaging origin and is tangent to the target contour in the contour of at least one tissue structure, and using the location information of the tangent line as the location information of the boundary reference lines.

[0015] For example, the cardiac ultrasound image includes multiple target structures, the target region includes a color Doppler sampling region, and the method further includes: merging the color Doppler sampling regions of at least two of the multiple target structures to obtain a color Doppler sampling region surrounding the at least two target structures.

[0016] For example, the echocardiogram images include N frames, where N is an integer greater than 1, and the target region includes a color Doppler sampling region. After determining the target region in each frame of the echocardiogram for at least one target structure, the method further includes: for each of the at least one target structure, determining the color Doppler sampling region of the target structure in each subsequently obtained echocardiogram image based on the color Doppler sampling regions of the target structure in at least a portion of the frames in the N frames, such that the color Doppler sampling region in each subsequently obtained echocardiogram image includes the region corresponding to the position of the i-th region in the frame of the echocardiogram image, where i includes all positive integers less than N+1, and the i-th region is the color Doppler sampling region of the target structure in the i-th frame of the N frames.

[0017] For example, the cardiac ultrasound image includes multiple target structures, and the method further includes: in response to a user's selection operation, displaying the target region of the target structure corresponding to the selection operation among the multiple target structures.

[0018] For example, the location information includes multiple localization points of the tissue structure, and the target region includes a pulse Doppler sampling gate region. Based on the determined section type, the echocardiogram image is segmented to determine the location information of the tissue structure contained in the echocardiogram image. This includes: determining multiple localization points of the tissue structure in the echocardiogram image and the confidence value of the target point among the multiple localization points using an image segmentation model based on the section type; and determining a target region in the echocardiogram image for at least one target structure based on the location information of the tissue structure. This includes: for each of the at least one target structures, determining a target region for that target structure based on its localization points and the confidence value of the target point.

[0019] For example, determining a target region for the target structure based on the location points and confidence values ​​of the target points includes: if the confidence value of the target points of the target structure is greater than a first confidence threshold, then determining a pulse Doppler sampling gate region for the target structure based on the location information of the target points; if the confidence value of the target points of the target structure is less than or equal to the first confidence threshold and greater than or equal to a second confidence threshold, then determining a pulse Doppler sampling gate region for the target structure based on multiple location points of the target structure; if the confidence value of the target points of the target structure is less than the second confidence threshold, then determining a pulse Doppler sampling gate region for the target structure based on location points other than the target points among the multiple location points of the target structure.

[0020] For example, the method further includes: adjusting the size of the target region and / or its position in the cardiac ultrasound image in response to a user's adjustment operation on the target region.

[0021] According to another aspect of the present invention, an ultrasound device is provided, comprising: a processor and a memory, wherein the memory stores computer program instructions, which, when executed by the processor, are used to perform the cardiac ultrasound image processing method as described above.

[0022] According to another aspect of the present invention, a storage medium is provided on which program instructions are stored, which, when executed, are used to perform the cardiac ultrasound image processing method as described above.

[0023] According to another aspect of the present invention, a computer program product is provided, comprising computer program instructions which, when executed, are used to perform the cardiac ultrasound image processing method as described above.

[0024] In the above technical solution, firstly, the section type of the cardiac ultrasound image is determined; then, based on the determined section type, the location information of the tissue structures contained in the cardiac ultrasound image is determined, and a target region is then identified in the cardiac ultrasound image for at least one target structure. This allows for automatic identification of the target region, enabling targeted ultrasound examination of that region without requiring manual identification by the user, reducing operation time and lowering the professional requirements for the user. In this technical solution, image segmentation is performed on the cardiac ultrasound image based on the image characteristics of tissue structures in cardiac ultrasound images of different section types, effectively ensuring image segmentation accuracy and thus guaranteeing the accuracy of the identified target region. Furthermore, this solution is applicable to cardiac ultrasound images of various section types, possessing strong versatility and facilitating rapid and comprehensive ultrasound examination of the heart.

[0025] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0026] The above and other objects, features, and advantages of the present invention will become more apparent from the more detailed description of the embodiments of the invention in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0027] Figure 1A A schematic diagram of a cardiac ultrasound image obtained by color Doppler ultrasound imaging according to an embodiment of the present invention is shown;

[0028] Figure 1B A schematic diagram of a pulsed Doppler image and a cardiac ultrasound image according to an embodiment of the present invention is shown;

[0029] Figure 2 A schematic flowchart of a method for processing cardiac ultrasound images according to an embodiment of the present invention is shown;

[0030] Figure 3 A schematic diagram of an image classification model according to an embodiment of the present invention is shown;

[0031] Figure 4 A schematic diagram of an image segmentation model according to an embodiment of the present invention is shown;

[0032] Figure 5A A schematic diagram of an echocardiogram with an apical four-chamber view as shown in an embodiment of the present invention is illustrated.

[0033] Figure 5B A schematic diagram of the color Doppler sampling region of a cardiac ultrasound image with an apical four-chamber view as an embodiment of the present invention is shown.

[0034] Figure 5C A schematic diagram of the color Doppler sampling region of a cardiac ultrasound image with an apical four-chamber view as described in another embodiment of the present invention is shown.

[0035] Figure 6 A schematic diagram of a cardiac ultrasound image with a parasternal left ventricular long-axis section as the section type is shown according to an embodiment of the present invention;

[0036] Figure 7 A schematic diagram of an echocardiogram with an apical four-chamber view as shown in another embodiment of the present invention is illustrated.

[0037] Figure 8 A schematic diagram of a cardiac ultrasound image with two target structures is shown according to another embodiment of the present invention;

[0038] Figure 9 A schematic block diagram of an ultrasonic device according to an embodiment of the present invention is shown. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of the present invention more apparent, exemplary embodiments according to the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are merely a subset of embodiments of the present invention.

[0040] To at least partially solve the above-mentioned technical problems, the present invention provides a method for processing cardiac ultrasound images.

[0041] During an echocardiogram, the position of the ultrasound probe can be adjusted to obtain different sectional views of the heart, depending on the needs. These sectional views can include: apical four-chamber view, apical two-chamber view, apical three-chamber view, apical five-chamber view, focused right ventricular four-chamber view, parasternal left ventricular long-axis view, focused aorta parasternal left ventricular long-axis view, parasternal aortic short-axis view, parasternal pulmonary artery long-axis view, subxiphoid apical four-chamber view, and subxiphoid two-atrial view. Different sectional views allow observation of the morphology of various cardiac structures, providing a comprehensive examination of the heart. Because different sectional views correspond to different sections of the heart, the included tissue structures and characteristics vary.

[0042] Considering the above factors, in the cardiac ultrasound image processing method according to the embodiments of this application, the section type of the cardiac ultrasound image is first determined, and then the location information of the tissue structures contained in the cardiac ultrasound image is determined according to the determined section type. Then, a target region is determined in the cardiac ultrasound image for the target structure. The target region is used to update the current cardiac ultrasound image or generate another cardiac ultrasound image. For example, the target region can be a sampling region; that is, ultrasound data is sampled based on the target region, and measurement, rendering, and other processing are performed based on the sampled ultrasound data. The corresponding processing results are then displayed on the cardiac ultrasound image (which can be directly displayed on the target region) or outside the ultrasound image. In some embodiments, in the updated cardiac ultrasound image, the imaging method within the target region can be different from the imaging method of other regions in the cardiac ultrasound image. For example, the cardiac ultrasound image can be conventional grayscale ultrasound imaging, specifically B-mode ultrasound, where color Doppler imaging of the target region can be performed, and the color Doppler image of the target region is superimposed and displayed on the target region of the cardiac ultrasound image. In these embodiments, the target area may also be referred to as the color Doppler sampling area, which may also be referred to as the C-sampling frame. Figure 1A A schematic diagram of a cardiac ultrasound image obtained by color Doppler ultrasound imaging according to an embodiment of the present invention is shown. Figure 1A As shown in the figure, the area enclosed by the blue box can be the target region. Color Doppler ultrasound imaging is performed within the target region, and the color Doppler image can be superimposed on the cardiac ultrasound image for display. Alternatively, this target region can be used to generate another cardiac ultrasound image, defining the cardiac imaging area targeted by this additional cardiac ultrasound image. For example, the target region can be the pulse Doppler sampling gate region, where the pulse Doppler sampling gate can be called the PW sampling gate. Based on this target region, the spatial target point for pulse Doppler flow measurement can be precisely located to obtain a pulse Doppler image of the blood flow spectrum within the target region. Figure 1B A schematic diagram of a pulsed Doppler image and a cardiac ultrasound image according to an embodiment of the present invention is shown. Figure 1BAs shown, the upper part of the image is a cardiac ultrasound image, and the lower part is a pulsed Doppler image of the blood flow spectrum within the target area. It can be understood that the cardiac ultrasound image display interface can simultaneously display both color Doppler and pulsed Doppler images. The aforementioned cardiac ultrasound image processing method can automatically and accurately determine the target area within cardiac ultrasound images of different slice types. This cardiac ultrasound image processing method can be applied to any electronic device, i.e., executed by any electronic device. Specifically, the cardiac ultrasound image processing method can be applied to the processor of any electronic device, i.e., executed by the processor of any electronic device. The electronic device can be the ultrasound equipment itself used to acquire cardiac ultrasound images, or it can be an electronic device different from the ultrasound equipment but capable of communicating with it. The communication connection described herein can be implemented using any wired and / or wireless connection method.

[0043] For example, Figure 2 A schematic flowchart illustrating a method for processing echocardiogram images according to an embodiment of the present invention is shown. Optionally, this echocardiogram image processing method can be executed in real time during the acquisition of echocardiogram images, thereby enabling real-time imaging based on the target region determined by the processing method. Alternatively, the processing method can also be executed after the acquisition of echocardiogram images. Figure 2 As shown, the method for processing the cardiac ultrasound image includes steps S1100, S1200, S1300 and S1400.

[0044] In step S1100, a cardiac ultrasound image is acquired.

[0045] Echocardiograms can be acquired using ultrasound equipment. For example, a conventional surface ultrasound probe can be used to acquire echocardiogram images. Alternatively, other invasive probes, such as those inserted through the esophagus or via a catheter, can be used to acquire echocardiogram images. The former will be described below as an example. An ultrasound probe can be placed at a specific location on the body surface of the target subject to emit ultrasound waves towards the heart and receive the echo signals. The echo signals are then processed to obtain echocardiogram images. The ultrasound probe can be placed at different locations on the body surface of the target subject to scan the heart, thereby obtaining echocardiogram images of different cross-sectional types. For example, placing the ultrasound probe in the acoustic window region corresponding to the anatomical apex of the left ventricle, with the probe indicator pointing to the 3 o'clock position, the scanning direction pointing towards the right scapula, and the midline of the scanning plane passing through the cardiac cruciate structure, will acquire an echocardiogram image of the apical four-chamber view.

[0046] Optionally, in response to a user's triggering operation on a preset control, the cardiac ultrasound image currently acquired by the ultrasound device can be used as the acquired cardiac ultrasound image.

[0047] In step S1200, the cross-sectional type of the cardiac ultrasound image is determined using an image classification model.

[0048] Image classification models can be Vision Transformers (ViT), hierarchical visual self-attention models based on moving windows (Swin Transformers), etc. Any existing or future model for image classification can be used. Figure 3 A schematic diagram of an image classification model according to an embodiment of the present invention is shown. Figure 3 As shown, the image classification model includes a location encoding module, an image patch linear transformation module, an encoder module, and a multilayer perceptron module. The echocardiogram image can be divided into multiple image patches and input into the location encoding module to obtain the location encoding result. Multiple image patches of the echocardiogram image can be input into the image patch linear transformation module to obtain the transformation result. Both the location encoding result and the transformation result can be input into the encoder module and the multilayer perceptron module, which are connected sequentially, to obtain the classification result of the echocardiogram image. The image classification model can be trained using a first training image and its corresponding annotation information to obtain a trained image classification model. The first training image is also an echocardiogram image, and the annotation information of the first training image can include information about the cross-sectional type of the first training image. For example, the first training image can include any type of echocardiographic training image, such as: apical four-chamber view, apical two-chamber view, apical three-chamber view, apical five-chamber view, focused right ventricular four-chamber view, parasternal left ventricular long-axis view, focused aorta parasternal left ventricular long-axis view, parasternal aorta short-axis view, parasternal pulmonary artery long-axis view, subxiphoid apical four-chamber view, subxiphoid two-atrial view, etc. The acquired echocardiographic image can be classified using a trained image classification model to determine the section type of the echocardiographic image.

[0049] Optionally, after determining the section type of the echocardiogram image, the name of the section type can be displayed in the echocardiogram image display interface. For example, if the section type of the acquired echocardiogram image is determined to be an apical four-chamber view, its common abbreviation "A4C" can be displayed.

[0050] In step S1300, the cardiac ultrasound image is segmented according to the determined section type to determine the location information of the tissue structures contained in the cardiac ultrasound image. The tissue structures include at least one target structure.

[0051] The tissue structures in a cardiac ultrasound image can be local structures of the heart or other tissue structures outside the heart. It is understood that the tissue structures included in cardiac ultrasound images of different slice types can be the same or different, but the location information of the tissue structures will differ. In one example, the slice type of the acquired cardiac ultrasound image can be an apical four-chamber view, which includes the left ventricle, left atrium, mitral valve, right ventricle, right atrium, and tricuspid valve. The target structure is the tissue structure used to identify the target region of interest to the user in the cardiac ultrasound image. In the example above, the target structure may include the mitral and tricuspid valves. It is understood that the target structures in cardiac ultrasound images of different slice types can be the same or different. Specifically, the target structures in a cardiac ultrasound image may include one or more of the mitral valve, tricuspid valve, aortic valve, pulmonary valve, left ventricular outflow tract, and right ventricular outflow tract.

[0052] Because the presentation of tissue structures in cardiac ultrasound images differs from that of the surrounding background tissue, image segmentation algorithms can be used to segment cardiac ultrasound images to determine the location information of the tissue structures contained within them.

[0053] Image segmentation methods can be traditional, such as thresholding and edge detection. Image segmentation can also be achieved using image segmentation models. Different types of cardiac ultrasound images can be segmented using the same or different image segmentation models. Using image segmentation models to segment cardiac ultrasound images allows for the precise extraction of anatomical structures and functional regions of cardiac tissues, achieving the transformation from pixel-level raw images to structured semantic information, providing core support for subsequent image processing.

[0054] In some embodiments, for echocardiogram images with different slice types, an image segmentation method corresponding to that slice type can be used to segment the echocardiogram image. The image segmentation methods corresponding to different slice types can be completely different, or only differ in their parameters. In some alternative embodiments, when using an image segmentation model for image segmentation, the same image segmentation model can be used to segment all echocardiogram images with different slice types. When inputting the echocardiogram image into the image segmentation model, the slice type information of the echocardiogram image is also input, so that the image segmentation model outputs the location information of the group structures contained in the echocardiogram image.

[0055] Image segmentation models can include U-Net, Residual Neural Network (ResNet), Convolutional Next (ConvNeXt), and others.

[0056] The image segmentation model can be any existing or future image segmentation model. The image segmentation model can be trained using a second training image and its corresponding annotation information to obtain a trained image segmentation model. Figure 4 A schematic diagram of an image segmentation model according to an embodiment of the present invention is shown. Figure 4 As shown, the image segmentation model includes an encoder, a decoder, and a skip connection module. An echocardiogram image can be input into the encoder, and through the skip connection module and decoder, an image segmentation result is obtained to determine the location information of the tissue structures contained in the echocardiogram image. The second training image is also an echocardiogram image, and the annotation information of the second training image can include the location information of the tissue structures in the second training image. When performing this step S1300 using the same image segmentation model, the annotation information of the second training image also includes information about the section type of the second training image. For example, the second training image can include echocardiogram training images of any section type, such as: apical four-chamber view, apical two-chamber view, apical three-chamber view, apical five-chamber view, focused right ventricular four-chamber view, parasternal left ventricular long-axis view, focused aortic parasternal left ventricular long-axis view, parasternal aortic short-axis view, parasternal pulmonary artery long-axis view, subxiphoid apical four-chamber view, subxiphoid biatrial view, etc. For example, for echocardiographic training images of the apical four-chamber view, the outlines of the left ventricle, left atrium, mitral valve root, mitral valve tip, right ventricle, right atrium, tricuspid valve root, and tricuspid valve tip can be labeled; for echocardiographic training images of the parasternal left ventricle long-axis view, the outlines of the left ventricle, interventricular septum, left ventricular posterior wall, and aortic valve root can be labeled. In a specific example, the echocardiographic image obtained in step S1100 and the label of the section type determined in step S1200 can be input into the trained image segmentation model. The trained image segmentation model can determine the location information of the tissue structures contained in the echocardiographic image based on the section type of the echocardiographic image.

[0057] In step S1400, a target region is determined in the cardiac ultrasound image for at least one target structure based on the location information of the tissue structure.

[0058] Based on the location information of tissue structures contained in the determined echocardiogram image, a target region can be determined for each target structure in the echocardiogram image to update the current echocardiogram image or generate another echocardiogram image. The target region may include a color Doppler sampling region and / or a pulsed Doppler sampling gate region for the target structure. Taking the apical four-chamber view as an example, the location information of the left atrium, left ventricle, and mitral valve in the acquired echocardiogram image can be used to determine the color Doppler sampling region and / or pulsed Doppler sampling gate region for the mitral valve. Specifically, the color Doppler sampling region of the mitral valve can be jointly determined based on the location information of the left atrium and left ventricle, so that the color Doppler sampling region of the mitral valve can completely cover the area where the mitral valve is located, in order to observe the flow direction, velocity, and distribution of blood near the mitral valve. Based on the location information of the mitral valve, the pulsed Doppler sampling gate region of the mitral valve can be determined to obtain the blood flow spectrum at the mitral valve location. Similarly, the location information of the right atrium, right ventricle, and tricuspid valve in the tissue structure can be used to determine the color Doppler sampling region and / or pulsed Doppler sampling gate region for the tricuspid valve. In some embodiments, the user can keep the position of the ultrasound probe unchanged, and the subsequently acquired cardiac ultrasound images can be directly imaged based on the target region. It is understood that after the user changes the position of the ultrasound probe, the cross-sectional type of the cardiac ultrasound image may change, and steps S1100 to S1400 above can be executed again to determine a new target region. In other embodiments, the target region can be determined for the target structure in each frame of cardiac ultrasound image. In other words, each frame of cardiac ultrasound image acquired in real time by the ultrasound device can be subjected to steps S1100 to S1400 above to determine the target region.

[0059] For example, when determining the target region for a target structure, the operating parameters of color Doppler and / or pulse Doppler can be automatically determined. For instance, while determining the color Doppler sampling region for the target structure, the operating parameters of color Doppler, such as adjustment gain, velocity scale, baseline, and deflection, can be determined. Similarly, while determining the pulse Doppler sampling gate region for the target structure, the operating parameters of pulse Doppler, such as sampling volume, can be determined.

[0060] In the above technical solution, firstly, the section type of the cardiac ultrasound image is determined; then, based on the determined section type, the location information of the tissue structures contained in the cardiac ultrasound image is determined, and a target region is then identified in the cardiac ultrasound image for at least one target structure. This allows for automatic identification of the target region, enabling targeted ultrasound examination of that region without requiring manual identification by the user, reducing operation time and lowering the professional requirements for the user. In this technical solution, image segmentation is performed on the cardiac ultrasound image based on the image characteristics of tissue structures in cardiac ultrasound images of different section types, effectively ensuring image segmentation accuracy and thus guaranteeing the accuracy of the identified target region. Furthermore, this solution is applicable to cardiac ultrasound images of various section types, possessing strong versatility and facilitating rapid and comprehensive ultrasound examination of the heart.

[0061] For example, the location information of the tissue structure determined in step S1200 includes multiple location points of the tissue structure and location information of the outline of at least one tissue structure. Taking a cardiac ultrasound image with an apical four-chamber view as an example, the determined location information of the tissue structure may include the outline of the left ventricle, the outline of the left atrium, the location point of the mitral valve root, the location point of the mitral valve tip, the outline of the right ventricle, the outline of the right atrium, the location point of the tricuspid valve root, and the location point of the tricuspid valve tip. Tissue structures with outlines, such as the left ventricle, left atrium, right ventricle, and right atrium, may have location points of types such as apex and center point. The target area may include a color Doppler sampling area. In the color Doppler sampling area, the blood flow status of the target area can be displayed in color, and the display color can be different for blood flow with different directions and velocities.

[0062] Step S1400 determines the target region in the cardiac ultrasound image for at least one target structure based on the location information of the tissue structure, including steps S1410 and S1420.

[0063] In step S1410, based on the imaging origin of the cardiac ultrasound image, and according to the location information of multiple positioning points of the tissue structures contained in the cardiac ultrasound image and the contour of at least one tissue structure, the boundary position of the region of interest of each target structure in the cardiac ultrasound image is determined.

[0064] Cardiac ultrasound images can be sector-shaped, with the imaging origin being the center of the sector. Within the target body, the relative positional relationships between various tissue structures are generally fixed. The boundary of the region of interest (ROI) containing the target structure can be determined based on the positional information of multiple locating points and / or contours of the surrounding tissue structures. The ROI containing the target structure can include the target structure and its surrounding tissue structures. For each target structure, the tissue structures included in its ROI are fixed. Based on clinical knowledge, for example: the mitral valve ROI includes the left atrium, left ventricle, and mitral valve region; the tricuspid valve ROI includes the right atrium, right ventricle, and tricuspid valve region; the aortic valve ROI includes the aorta, left ventricle, and aortic valve region; the pulmonary valve ROI includes the pulmonary artery, right ventricle, and pulmonary valve region; the right ventricular outflow tract ROI includes the right ventricle, right ventricular outflow tract, and pulmonary valve region; and the left ventricular outflow tract ROI includes the left ventricle, left atrium, aortic valve, and aorta region. It is understandable that, based on the anatomical structure of the heart, if the target structure is the mitral valve, the boundary of the region of interest for the mitral valve can be determined at least based on the location information of multiple locating points and contours of the left atrium, left ventricle, and mitral valve; if the target structure is the tricuspid valve, the boundary of the region of interest for the tricuspid valve can be determined at least based on the location information of multiple locating points and contours of the right atrium, right ventricle, and tricuspid valve; if the target structure is the aortic valve, the boundary of the region of interest for the aortic valve can be determined at least based on the location information of multiple locating points and contours of the aorta, left ventricle, and aortic valve; and if the target structure is the pulmonary valve, the boundary of the region of interest for the pulmonary valve can be determined at least based on the location information of multiple locating points and contours of the pulmonary artery, right ventricle, and pulmonary valve. If the target structure is the right ventricular outflow tract, the boundary position of the region of interest of the right ventricular outflow tract can be determined based on the location information of multiple locating points and contours of the right ventricle, the right ventricular outflow tract, and the pulmonary valve. If the target structure is the left ventricular outflow tract, the boundary position of the region of interest of the left ventricular outflow tract can be determined based on the location information of multiple locating points and contours of the left ventricle, the left atrium, the aortic valve, and the aorta.

[0065] Taking an echocardiogram with an apical four-chamber view as the target structure, the boundary of the region of interest (ROI) of the mitral valve can be determined based on the contours of the left atrium, left ventricle, and mitral valve root. For example, the color Doppler sampling area can be rectangular. The center point of the mitral valve can be determined based on the locations of the left and right mitral valve root. The central axis of the rectangular area can be determined by connecting the center point of the mitral valve to the imaging origin. The lower boundary can be determined by the intersection of the central axis and the left atrium contour; the upper boundary can be determined by the intersection of the central axis and the left ventricle contour; the left boundary can be determined by the location of the left mitral valve root; and the right boundary can be determined by the location of the right mitral valve root. Therefore, the upper, lower, left, and right boundaries of the rectangular color Doppler sampling area can be determined. Figure 5A A schematic diagram of an echocardiogram with an apical four-chamber view, according to an embodiment of the present invention, is shown. Figure 5A As shown in the diagram, the blue outlines correspond to the outlines of the left ventricle and left atrium. White dot 1 is the imaging origin, white dot 2 is the location point of the left mitral valve root, white dot 3 is the location point of the right mitral valve root, and white dot 6 is the center point of the mitral valve calculated based on the location points of the left and right mitral valve roots. The line connecting white dot 6 and white dot 1 is the central axis. White dot 4 is the intersection of the central axis and the left atrial outline, and white dot 5 is the intersection of the central axis and the left ventricular outline. Therefore, the lines passing through white dots 2, 3, 4, and 5 can be defined as the boundary positions of the region of interest (ROI) of the mitral valve. It can be understood that the boundary position of the ROI containing the target structure can be used as internal calculation data, rather than being directly displayed to the user.

[0066] In step S1420, for each target structure in at least one target structure, the color Doppler sampling region of the target structure is determined according to the boundary position of the region of interest where the target structure is located.

[0067] Taking the apical four-chamber view of an echocardiogram as an example, with the mitral valve as the target structure, after determining the boundary of the region of interest (ROI) of the mitral valve, the area enclosed by this boundary can be defined as the color Doppler sampling region. Exemplarily, the electronic device used to perform the echocardiogram image processing method may include an output device or be communicatively connected to an output device. The output device may include a display, etc. The echocardiogram image can be displayed on the display screen. After determining the color Doppler sampling region of the target structure, it can be superimposed on the echocardiogram image display screen. Figure 5B A schematic diagram of the color Doppler sampling region of a cardiac ultrasound image with an apical four-chamber view, according to an embodiment of the present invention, is shown. Figure 5BAs shown in the figure, the area within the white dashed box is the color Doppler sampling area of ​​the target structure (mitral lobe).

[0068] In the above technical solution, based on the imaging origin of the cardiac ultrasound image, and according to the positional information of multiple positioning points of the tissue structures contained in the cardiac ultrasound image and the contour of at least one tissue structure, the boundary position of the region of interest where the target structure is located is determined, and then the color Doppler sampling area of ​​the target structure is determined. Thus, the target structure can be surrounded by the color Doppler sampling area, providing a reasonable color display range and ensuring the display effect of color Doppler.

[0069] For example, step S1410 includes steps S1411 and S1412.

[0070] In step S1411, for each region of interest in at least one target structure, the position information of multiple boundary reference lines of the region of interest is calculated based on the imaging origin and multiple positioning points of the tissue structure contained in the cardiac ultrasound image and the position information of the contour of at least one tissue structure.

[0071] Multiple boundary reference lines can include a left boundary reference line, a right boundary reference line, an upper boundary reference line, and a lower boundary reference line. The left and right boundary reference lines are rays passing through the imaging origin, respectively. The upper and lower boundary reference lines are arcs centered at the imaging origin. It can be understood that the region enclosed by the left, right, upper, and lower boundary reference lines can be shaped like a fan ring, meaning the final determined color Doppler sampling area of ​​the target structure is fan-shaped. Taking a cardiac ultrasound image with an apical four-chamber view as the cutaway, and the mitral valve as the target structure, multiple boundary reference lines can be determined based on the imaging origin, the outline of the left atrium, the outline of the left ventricle, and the location points of the mitral valve root. For example, the center point of the mitral valve can be determined based on the location points of the left and right mitral valve root. The central axis of the sector region can be determined by connecting the center point of the mitral valve with the imaging origin. An arc can be determined based on the intersection of the central axis and the left atrial contour, with the imaging origin as the center and the distance between the intersection and the imaging origin as the radius; this arc is the lower boundary reference line. An arc can be determined based on the intersection of the central axis and the left ventricular contour, with the imaging origin as the center and the distance between the intersection and the imaging origin as the radius; this arc is the upper boundary reference line. The ray connecting the imaging origin and the location point of the left mitral valve root can be determined as the left boundary reference line. The ray connecting the imaging origin and the location point of the right mitral valve root can be determined as the right boundary reference line.

[0072] In step S1412, for each region of interest in at least one target structure, the boundary position of the region of interest is determined based on the position information of its multiple boundary reference lines.

[0073] For example, the upper boundary reference line intersects with both the left and right boundary reference lines. The position of the curve between these two intersection points can determine the upper boundary location. Similarly, the lower boundary reference line intersects with both the left and right boundary reference lines, and the position of the curve between these two intersection points can determine the lower boundary location. The left boundary location can be determined based on the intersection points of the left boundary reference line with both the upper and lower boundary reference lines. The right boundary location can be determined based on the intersection points of the right boundary reference line with both the upper and lower boundary reference lines. The color Doppler sampling area can be determined based on these boundary locations. Figure 5C A schematic diagram of the color Doppler sampling region of a cardiac ultrasound image with an apical four-chamber view, according to another embodiment of the present invention, is shown. Figure 5C As shown in the figure, the fan-shaped area within the blue dashed line is the color Doppler sampling area of ​​the target structure (mitral lobe).

[0074] In the above technical solution, based on the positional information of the imaging origin and multiple positioning points and contours, the positional information of multiple boundary reference lines of the target structure is calculated. The left and right boundary reference lines are rays passing through the imaging origin, respectively, while the upper and lower boundary reference lines are arcs centered at the imaging origin. Since the cardiac ultrasound image itself is fan-shaped, defining the color Doppler sampling area as a concentric fan ring allows the color Doppler sampling area to naturally match the geometric characteristics of the cardiac ultrasound image, thereby better focusing on the target structure and its surrounding area, reducing interference from irrelevant areas, and improving the display effect.

[0075] For example, step S1411 calculates the location information of the boundary reference points of the region of interest, including performing steps S1411A and S1411B for each region of interest of at least one target structure.

[0076] In step S1411A, based on the imaging origin and the location information of multiple positioning points of the tissue structures contained in the echocardiogram image and the contour of at least one tissue structure, the location information of the boundary reference points of the region of interest is calculated. These boundary reference points can be used to determine the corresponding boundary reference lines.

[0077] Taking a parasternal left ventricular long-axis section as the cut type and the mitral valve as the target structure as an example, the location information of the boundary reference points of the region of interest of the mitral valve can be determined based on the location information of the left ventricle, interventricular septum, left ventricular posterior wall, and aortic valve. The centroid of the left ventricle can be calculated based on the location information of the left ventricular contour, and this centroid can be used as the boundary reference point of the left boundary reference line; the upper lateral root point of the aortic valve can be used as the boundary reference point of the right boundary reference line; the upper intersection point of the line connecting the imaging origin and the aforementioned centroid point with the contour of the interventricular septum can be used as the boundary reference point of the upper boundary reference line; and the point on the contour of the left ventricular posterior wall that is farthest from the imaging origin can be used as the boundary reference point of the lower boundary reference line.

[0078] In step S1411B, for each boundary reference point among the boundary reference points, the position information of the boundary reference line corresponding to the boundary reference point is determined based on the position information of the boundary reference point.

[0079] The left and right boundary reference lines can be rays passing through the imaging origin, while the upper and lower boundary reference lines can be arcs centered at the imaging origin. Boundary reference lines passing through boundary reference points can be determined based on the imaging origin; these are the boundary reference lines corresponding to the boundary reference points. For the boundary reference points corresponding to the left and right boundary reference lines, the line connecting the boundary reference point and the imaging origin can be used as the left or right boundary reference line. For the boundary reference points corresponding to the upper and lower boundary reference lines, an arc centered at the imaging origin with a radius equal to the distance between the imaging origin and the boundary reference point can be used as the upper or lower boundary reference line. Figure 6 A schematic diagram of a cardiac ultrasound image with a parasternal left ventricular long-axis section as described in an embodiment of the present invention is shown. Figure 6 As shown in the diagram, the red outline at the top represents the interventricular septum, the purple outline in the middle represents the left ventricle, and the yellow outline at the bottom represents the posterior wall of the left ventricle. White dot 1 is the imaging origin, white dot 2 is the superior aortic valve root, white dot 3 is the inferior aortic valve root, white dot 4 is the centroid of the left ventricle, white dot 5 is the upper intersection of the line connecting the centroid and the imaging origin with the outline of the interventricular septum, and white dot 6 is the point on the outline of the posterior wall of the left ventricle furthest from the imaging origin. It can be understood that white dot 2 can be considered the boundary reference point of the right boundary reference line, white dot 4 can be considered the boundary reference point of the left boundary reference line, white dot 5 can be considered the boundary reference point of the upper boundary reference line, and white dot 6 can be considered the boundary reference point of the upper boundary reference line. The positions of the boundary reference lines can be determined based on white dots 2, 4, 5, and 6, thereby determining the color Doppler sampling area. Figure 6 The area enclosed by the white lines is the determined color Doppler sampling area.

[0080] In the above technical solution, the positional information of the boundary reference points of the target structure is calculated based on the positional information of the imaging origin and multiple positioning points and contours of the tissue structure; then, based on the calculated positional information of the boundary reference points, the positional information of the boundary reference line corresponding to the boundary reference points is determined. Thus, the boundary reference line can be accurately determined based on the boundary reference points, ensuring a reasonable range for the color Doppler sampling area. Moreover, the computational cost is low and the computation speed is fast.

[0081] For example, step S1411, which calculates the position information of the boundary reference point of the region of interest, may include step S1411C. In step S1411C, the position information of the tangent line passing through the imaging origin and tangent to the target contour in the contour of at least one tissue structure is determined, and the position information of the tangent line is used as the position information of the boundary reference line.

[0082] The target contour may include the ventricular contour, atrial contour, septal contour, etc. Taking a cardiac ultrasound image with an apical four-chamber view as the section type and the tricuspid valve as the target structure as an example, the boundary position can be determined based on the location points of the right atrial contour, right ventricular contour, and tricuspid valve root. In some embodiments, based on two tangent lines passing through the imaging origin and tangent to the right atrial contour, the position information of the tangent line located on the right side of the right atrium can be used as the position information of the right boundary reference line, and the position information of the tangent line located on the left side of the right atrium can be used as the position information of the left boundary reference line. In other embodiments, based on two tangent lines passing through the imaging origin and tangent to the right ventricular contour, the position information of the tangent line located on the right side of the right ventricle can be used as the position information of the right boundary reference line, and the position information of the tangent line located on the left side of the right ventricle can be used as the position information of the left boundary reference line.

[0083] For example, a method for processing cardiac ultrasound images according to an embodiment of the present invention may include the above-described steps S1411A, S1411B and S1411C. In other words, these steps may be performed together in one embodiment to determine the location information of the boundary reference line of the region of interest. Figure 7 A schematic diagram of an echocardiogram with an apical four-chamber view, according to another embodiment of the present invention, is shown. Figure 7As shown, the target structure is the tricuspid valve. The upper blue outline represents the right ventricle, and the lower blue outline represents the right atrium. White dot 1 is the imaging origin, white dot 2 is the location point of the left tricuspid valve root, and white dot 3 is the location point of the right tricuspid valve root. Based on the location points of the left and right tricuspid valve roots, the center point can be determined, and white dot 4 is the center point. The point furthest from this center point on the right ventricular outline, i.e., the right ventricular apex, can be determined, and white dot 5 is the right ventricular apex. The line connecting the right ventricular apex and the center point can be used as the right ventricular midline. The two intersection points of the perpendicular bisector of the right ventricular midline and the right ventricular outline can be determined, and white dot 6 is the intersection point on the left side of the right ventricular midline, and white dot 8 is the intersection point on the right side of the right ventricular midline. White dot 7 is the point on the right atrial outline furthest from the imaging origin. White dot 5 can be the boundary reference point of the upper boundary reference line, white dot 6 can be the boundary reference point of the left boundary reference line, and white dot 7 can be the boundary reference point of the lower boundary reference point. Figure 7 The red dashed line can be considered as a tangent line passing through the imaging origin 1 and tangent to the outline of the right atrium; this red dashed line can serve as a reference line for the right boundary. Therefore, based on the positional information of the white dots 5, 6, and 7 and the red dashed line, the color Doppler sampling area of ​​the target structure can be determined. Figure 7 The area enclosed by the yellow lines is the color Doppler sampling area.

[0084] In the above technical solution, the position information of the tangent line passing through the imaging origin and tangent to the target contour is determined, and the position information of the tangent line is used as the position information of the boundary reference line. Therefore, the position information of the left boundary reference line and / or the right boundary reference line can be determined quickly and accurately, and it is ensured that the tissue structure of interest is surrounded by the color Doppler sampling area, providing the user with a reasonable color Doppler sampling area.

[0085] For example, the cardiac ultrasound image includes multiple target structures, and the target region includes a color Doppler sampling region.

[0086] It is understandable that, based on cardiac anatomy, in echocardiographic images with an apical four-chamber view, the target structures may include the mitral and tricuspid valves; in echocardiographic images with a parasternal left ventricular long-axis view, the target structures may include the mitral and aortic valves; in echocardiographic images with a focused aorta-parasternal left ventricular long-axis view, the target structures may include the mitral and aortic valves; in echocardiographic images with a pulmonary artery long-axis view, the target structures may include the pulmonary valve and the right ventricular outflow tract; and in echocardiographic images with an apical five-chamber view... In echocardiogram images with a focused right ventricular four-chamber view, target structures may include the aortic valve and left ventricular outflow tract; in an apical three-chamber view, target structures may include the aortic valve and left ventricular outflow tract; in a parasternal aortic short-axis view, target structures may include the aortic valve and pulmonary valve; and in a subxiphoid apical four-chamber view, target structures may include the mitral valve and tricuspid valve. The target region can be a color Doppler sampling area used to display blood flow in the target structure and surrounding area in color. It is understood that echocardiogram images may include three or more target structures; for simplicity, these will not be listed here.

[0087] The method for processing cardiac ultrasound images may further include step S1500. In step S1500, the color Doppler sampling regions of at least two of the multiple target structures are merged to obtain a color Doppler sampling region surrounding the at least two target structures.

[0088] Multiple target structures can be identified in a cardiac ultrasound image, and a corresponding color Doppler sampling region is determined for each target structure. In some embodiments, the user can perform a merging operation using an input device. For example, the target structures may include the mitral valve and the tricuspid valve, and the merging operation may include the user selecting the mitral valve and the tricuspid valve and clicking the merge control. It is understood that each target structure may have a corresponding color Doppler sampling region, and the color Doppler sampling regions of the user-selected target structures can be merged, resulting in a merged color Doppler sampling region that completely surrounds the selected target structures. For example, based on the boundary reference lines of each target structure among the selected target structures, the boundary reference lines of all the user-selected target structures can be integrated, and the outermost boundary reference line can be taken to form a color Doppler sampling region that simultaneously surrounds all selected target structures. In other embodiments, when the cardiac ultrasound image includes multiple target structures, the color Doppler sampling regions of at least two target structures can be automatically merged. For example, the distance between any two color Doppler sampling regions can be determined, and two color Doppler sampling regions with a distance less than a distance threshold can be merged to obtain a color Doppler sampling region surrounding the two target structures. Optionally, when the cardiac ultrasound image includes multiple target structures, the color Doppler sampling regions of all target structures can be merged directly.

[0089] Figure 8 A schematic diagram of a cardiac ultrasound image with two target structures is shown according to another embodiment of the present invention. Figure 8 As shown, the color Doppler sampling regions of target structure 1 and target structure 2 can be identified separately in the cardiac ultrasound image. The color Doppler sampling regions of target structure 1 and target structure 2 can be merged, and the merged color Doppler sampling region can be... Figure 8 The area enclosed by the red outline.

[0090] In the above technical solution, the color Doppler sampling regions of at least two target structures are merged to obtain a color Doppler sampling region surrounding at least two target structures. Therefore, the merged color Doppler sampling region can simultaneously display the blood flow conditions of at least two target structures, facilitating user comparison and observation of the blood flow conditions of the included target structures.

[0091] For example, a cardiac ultrasound image includes multiple target structures. Different slice types of cardiac ultrasound images may include different multiple target structures, as explained above regarding multiple target structures; for brevity, this will not be repeated here. The cardiac ultrasound image processing method may further include step S1600. In step S1600, in response to a user's selection operation, the target region of the target structure corresponding to the selection operation among the multiple target structures is displayed.

[0092] Multiple target structures can be identified in a cardiac ultrasound image, and a corresponding target region is defined for each target structure. Users can select target structures using an input device. For example, target structures may include the mitral valve and the tricuspid valve. The selection operation may include the user checking the checkbox corresponding to the mitral valve; the mitral valve is the target structure selected. In some embodiments, the target region corresponding to each target structure may be displayed before the selection operation, and only the target region of the target structure selected may be displayed after the selection operation. In other embodiments, the target region may not be displayed before the selection operation, but may be displayed after the selection operation.

[0093] In the above technical solution, in response to the user's selection operation, the target area of ​​the target structure corresponding to the selection operation is displayed among multiple target structures. Only the target area of ​​the user-selected target structure can be displayed, allowing the user to focus on observing the selected target structure and avoiding interference from other areas on the screen.

[0094] For example, the echocardiogram image includes N frames, where N is an integer greater than 1. The N frames may include echocardiogram images acquired at different times during at least one cardiac cycle. The target region includes a color Doppler sampling region. After determining the target region in each frame of the echocardiogram image for at least one target structure in step S1400, the echocardiogram image processing method may further include step S1700. In step S1700, for each target structure in the at least one target structure, based on at least a portion of the color Doppler sampling regions of the target structure in the N frames, the color Doppler sampling region of the target structure in each subsequently acquired echocardiogram image is determined, such that the color Doppler sampling region in each subsequently acquired echocardiogram image includes the region corresponding to the position of the i-th region in that frame of the echocardiogram image. Here, i includes all positive integers less than N+1. The i-th region is the color Doppler sampling region of the target structure in the i-th frame of the N frames.

[0095] It is understood that the same target structure can be included in all N frames of images. Due to the different acquisition times of the images and the different times the heart is in the cardiac cycle, the position of the target structure may vary slightly due to motion, and consequently, the position and size of the determined color Doppler sampling region may vary slightly. For the same target structure, a merged region of the color Doppler sampling regions of the target structure in at least a portion of the N frames can be determined, and this merged region includes the color Doppler sampling regions in at least a portion of the frames. The accuracy of the color Doppler sampling region in each frame of the N frames can be determined. At least a portion of the frames may include accurately determined color Doppler sampling regions, and at least a portion of the frames may not include inaccurate color Doppler sampling regions. The color Doppler sampling region of the target structure in each subsequent frame of echocardiography is the aforementioned merged region. For example, the currently acquired echocardiography images may include sequentially acquired images 1, 2, and 3. The color Doppler sampling region of target structure 1 is determined in each frame. The color Doppler sampling regions of target structure 1 in images 1 and 3 can be merged to obtain the merged color Doppler sampling region, i.e., the merged region mentioned above. For subsequent images 4, 5, etc., the color Doppler sampling region of target structure 1 can be directly determined as the merged region mentioned above. In other words, after obtaining cardiac ultrasound images after N frames, the color Doppler sampling region can be determined without performing the above steps S1100 to S1400.

[0096] In the above technical solution, for each target structure in at least one target structure, the color Doppler sampling region of the target structure in each subsequent frame of echocardiogram image is determined directly based on the color Doppler sampling regions of the target structure in at least a portion of the N frames of images. This integrates the color Doppler sampling regions of the target structure across multiple frames, avoiding the target structure exceeding the ultrasound color Doppler sampling region due to heartbeat. The merged region is then directly used as the color Doppler sampling region in the subsequent echocardiogram images, avoiding the need to calculate the color Doppler sampling region for each target structure in every echocardiogram image, thus reducing computational load and saving computational resources.

[0097] For example, the location information includes multiple location points of the tissue structure. The target region includes a pulsed Doppler sampling gate region. The pulsed Doppler sampling gate region may include a sampling gate and a sampling line. The pulsed Doppler sampling gate region can be used to indicate the emission location of the pulse signal to obtain the blood flow spectrum of the pulsed Doppler sampling gate region. If the target structure is the mitral valve, the sampling gate may be located at the tip of the mitral valve, and the sampling line may be aligned with the direction of blood flow into the mitral valve; if the target structure is the tricuspid valve, the sampling gate may be located at the tip of the tricuspid valve, and the sampling line may be aligned with the direction of blood flow into the tricuspid valve; if the target structure is the aortic valve, the sampling gate may be located at the aortic valve orifice; if the target structure is the pulmonary valve, the sampling gate may be located at the pulmonary valve orifice; if the target structure is the left ventricular outflow tract, the sampling gate may be located in the central area of ​​the left ventricular outflow tract; if the target structure is the right ventricular outflow tract, the sampling gate may be located in the central area of ​​the right ventricular outflow tract.

[0098] Step S1300 involves image segmentation of the cardiac ultrasound image based on the determined section type, including step S1310. In step S1310, based on the section type, an image segmentation model is used to determine multiple localization points of tissue structures in the cardiac ultrasound image and the confidence values ​​of target points among these localization points.

[0099] As mentioned earlier, image segmentation models can be used to determine the location information of tissue structures contained in echocardiogram images. Location information can include multiple localization points of the tissue structure. Furthermore, the image segmentation model can be used to determine the confidence value of target points among these multiple localization points. Target points can include at least the localization points of the mitral valve tip, tricuspid valve tip, aortic valve orifice, pulmonary valve orifice, the central region of the left ventricular outflow tract, and the central region of the right ventricular outflow tract. It is understood that due to the beating of the heart, echocardiogram images may be unclear, and the location information of target points may not be accurately determined. Confidence values ​​can be used to measure the accuracy of the determination of the target point's location information. The confidence value can range from [0, 1]. The closer the confidence value is to 1, the higher the accuracy of the target point's location information; the closer it is to 0, the lower the accuracy of the target point's location information.

[0100] Step S1400, which determines the target region in the cardiac ultrasound image based on the location information of the tissue structure, may include step S1430.

[0101] In step S1430, for each target structure in at least one target structure, a target region is determined for that target structure based on the location point of that target structure and the confidence value of the target point.

[0102] For example, if the confidence value of the target point is greater than a preset value, the pulse Doppler sampling gate region can be directly determined based on the target point's location information. For instance, a sampling gate for the pulse Doppler sampling gate region can be placed at the target point's location, and sampling lines can be set. If the confidence value of the target point is less than or equal to the preset value, the pulse Doppler sampling gate region can be determined based on the target structure's location points. For instance, the target point's position can be corrected based on the target structure's location points, and a sampling gate for the pulse Doppler sampling gate region can be placed at the corrected target point location, with sampling lines set. Image segmentation models can also be used to determine the location information of the contour of at least one tissue structure within the cardiac ultrasound image. The target point's position can be corrected jointly based on the target structure's location points and the location information of the contour of at least one tissue structure.

[0103] In the above technical solution, the target region includes the pulse Doppler sampling gate region. For the target structure, the target region is determined based on the location points of the target structure and the confidence values ​​of the target points. This reduces the impact of the target structure's movement on the target point's location, improves the accuracy of the pulse Doppler sampling gate region determination, and ensures the accuracy and reliability of the acquired blood flow spectrum.

[0104] For example, step S1430 includes steps S1431, S1432 and S1433.

[0105] It is understood that one step can be selected from steps S1431, S1432, and S1433 based on the confidence value of the target point. Taking the mitral valve as the target structure, the location points of the mitral valve can include the location points of the left mitral valve root, the right mitral valve root, and the mitral valve tip. The location point of the mitral valve tip can be the target point. Taking the left ventricular outflow tract as the target structure, the left ventricular outflow tract includes the location points of the central region of the left ventricular outflow tract and the outline of the left ventricular outflow tract. The target point can be the location point of the central region of the left ventricular outflow tract. For example, the confidence value can range from [0, 1]. It is understood that the first confidence threshold can be greater than the second confidence threshold, and the user can set the size of the first and second confidence thresholds as needed.

[0106] In step S1431, if the confidence value of the target point of the target structure is greater than the first confidence threshold, then the pulse Doppler sampling gate region of the target structure is determined based on the location information of the target point. The first confidence threshold can be 0.8. If the confidence value of the target point of the target structure is greater than the first confidence threshold, then the sampling gate of the pulse Doppler sampling gate region can be placed directly at the location of the target point based on the location information of the target point, and sampling lines can be set to determine the pulse Doppler sampling gate region of the target structure.

[0107] In step S1432, if the confidence value of the target point of the target structure is less than or equal to a first confidence threshold and greater than or equal to a second confidence threshold, then the pulse Doppler sampling gate region of the target structure is determined based on multiple positioning points of the target structure. The first confidence threshold can be 0.8, and the second confidence threshold can be 0.5. Taking the mitral valve as an example, if the confidence value of the target point is less than or equal to the first confidence threshold and greater than or equal to the second confidence threshold, the position of the positioning point can be corrected based on multiple positioning points of the target structure. For example, the line connecting the valve roots can be determined based on the positioning points of the left and right mitral valve roots, and the perpendicular bisector of the line connecting the valve roots can be determined. The projection point of the target point (the positioning point of the mitral valve tip) on the perpendicular bisector is taken as the corrected target point. The sampling gate of the pulse Doppler sampling gate region can be placed at the position of the corrected target point, and sampling lines can be set to determine the pulse Doppler sampling gate region of the target structure.

[0108] In step S1433, if the confidence value of the target point of the target structure is less than the second confidence threshold, then the pulse Doppler sampling gate region of the target structure is determined based on the positioning points other than the target point among the multiple positioning points of the target structure. The second confidence threshold can be 0.5. Taking the mitral valve as an example, if the confidence value of the target point of the target structure is less than the second confidence threshold, the pulse Doppler sampling gate region of the target structure can be determined based on the positioning points of the left and right mitral valve roots. For example, the perpendicular bisector of the line connecting the positioning points of the left and right mitral valve roots can be determined based on their positions. In the direction of the mitral tip along the vertical bisector, taking the midpoint of the line connecting the positioning point of the left mitral valve root and the positioning point of the right mitral valve root as the starting point, determine a position equal to a preset length, place the sampling gate of the pulse Doppler sampling gate region at this position, and set the sampling line to determine the pulse Doppler sampling gate region of the target structure.

[0109] In the above technical solution, if the confidence value of the target point is greater than the first confidence threshold, the pulse Doppler sampling gate region of the target structure is directly determined based on the location information of the target point. If the confidence value of the target point is less than or equal to the first confidence threshold and greater than or equal to the second confidence threshold, the pulse Doppler sampling gate region is determined based on multiple positioning points of the target structure. If the confidence value of the target point is less than the second confidence threshold, the pulse Doppler sampling gate region is determined based on positioning points other than the target point. Therefore, based on the confidence value of the target point, even if the location information of the target point is inaccurate, the pulse Doppler sampling gate region can be reasonably determined, ensuring the accuracy and reliability of the acquired blood flow spectrum.

[0110] For example, the method for processing echocardiogram images further includes step S1800. In step S1800, in response to a user's adjustment operation on a target region, the size and / or position of the target region in the echocardiogram image are adjusted. After the target region is determined for a target structure, the echocardiogram image and the corresponding target region can be displayed on the display interface of the output device. The user can perform the adjustment operation using an input device. For example, by selecting the target region with a mouse, the user can scale, rotate, drag, etc., the target region to adjust its size and / or position in the echocardiogram image. The adjusted target region can be applied to all subsequently acquired echocardiogram images. Echocardiogram images acquired after the adjustment operation can be processed based on the adjusted target region until a new target region is determined or the target region is adjusted again.

[0111] In the above technical solution, the size and / or position of the target area in the cardiac ultrasound image are adjusted in response to the user's adjustment operation. This allows users to adjust the automatically determined target area, improving the user experience.

[0112] For example, the target area may include a color Doppler sampling area and a pulse Doppler sampling gate area. Users can turn the color Doppler function on or off; users can turn the pulse Doppler function on or off. In other words, users can achieve different observation purposes by independently switching these two functions on and off: simultaneously turning on both functions allows observation of both color flow imaging and blood flow spectrum; turning on only the color Doppler function allows observation of only color flow imaging; turning on only the pulse Doppler function allows observation of only the blood flow spectrum.

[0113] By way of example, according to another aspect of the present invention, an apparatus for processing cardiac ultrasound images is also provided. The apparatus includes a processor. The processor is used to execute the cardiac ultrasound image processing method of any embodiment of the present application.

[0114] By way of example, according to another aspect of the present invention, an ultrasonic device is also provided. Figure 9 A schematic block diagram of an ultrasound device 900 according to an embodiment of the present invention is shown. The ultrasound device 900 includes a processor 910 and a memory 920. The memory 920 stores computer program instructions, which, when executed by the processor 910, are used to perform the cardiac ultrasound image processing method as described above.

[0115] For example, the aforementioned ultrasound equipment can be various ultrasound imaging devices, such as ultrasound diagnostic instruments or ultrasound imaging workstations. Ultrasound diagnostic instruments can be used to perform ultrasound imaging examinations. Specifically, ultrasound diagnostic instruments can have different probe types to examine different areas, such as the heart, abdomen, gynecological organs, and breasts. Ultrasound imaging workstations can be devices integrating functional modules such as patient registration, image acquisition, diagnostic editing, report printing, image post-processing, medical record retrieval, and statistical analysis. Ultrasound imaging workstations can be communicatively connected to ultrasound diagnostic instruments, for example, through any wired or wireless communication method. Ultrasound diagnostic instruments can transmit the acquired ultrasound echo signals to the ultrasound imaging workstation for processing, analysis, and storage.

[0116] By way of example, according to another aspect of the present invention, a storage medium is also provided, on which program instructions are stored, which, when executed, are used to perform the cardiac ultrasound image processing method described above. The storage medium may, for example, include an erasable programmable read-only memory (EPROM), a portable read-only memory (CD-ROM), a USB memory, or any combination of the above storage media. The storage medium may be any combination of one or more computer-readable storage media.

[0117] By way of example, according to another aspect of the present invention, a computer program product is also provided, including computer program instructions, which, when executed, are used to perform the method for processing cardiac ultrasound images as described above.

[0118] Those skilled in the art can understand the specific implementation scheme and beneficial effects of the above-mentioned cardiac ultrasound image processing device, ultrasound equipment, storage medium and computer program product by reading the relevant description of the above-mentioned cardiac ultrasound image processing method. For the sake of brevity, they will not be described in detail here.

[0119] Although exemplary embodiments have been described herein with reference to the accompanying drawings, it should be understood that the above exemplary embodiments are merely illustrative and are not intended to limit the scope of this application. Various changes and modifications can be made therein by those skilled in the art without departing from the scope and spirit of this application. All such changes and modifications are intended to be included within the scope of this application as claimed in the appended claims.

[0120] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0121] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed.

[0122] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of this application may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0123] Similarly, it should be understood that, in order to streamline this application and aid in understanding one or more of the various inventive aspects, features of this application may sometimes be grouped together in a single embodiment, figure, or description thereof in the description of exemplary embodiments of this application. However, this approach should not be construed as reflecting an intention that the claimed application requires more features than are expressly recited in each claim. Rather, as reflected in the corresponding claims, its inventive point lies in solving the corresponding technical problem with features fewer than all features of a single disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of this application.

[0124] Those skilled in the art will understand that, apart from the mutual exclusion of features, all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or apparatus so disclosed can be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.

[0125] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features but not others included in other embodiments, combinations of features from different embodiments are intended to be within the scope of this application and form different embodiments. For example, in the claims, any one of the claimed embodiments can be used in any combination.

[0126] The various component embodiments of this application can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some modules in the cardiac ultrasound image processing apparatus according to embodiments of this application. This application can also be implemented as an apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such an implementation of this application can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

[0127] It should be noted that the above embodiments are illustrative of this application and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. This application can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.

[0128] The above description is merely a specific embodiment or illustration of the embodiments of this application. The scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. The scope of protection of this application shall be determined by the scope of the claims.

Claims

1. A method for processing cardiac ultrasound images, characterized in that, include: Acquire cardiac ultrasound images; The cross-sectional type of the cardiac ultrasound image is determined using an image classification model; Based on the determined section type, the cardiac ultrasound image is segmented to determine the location information of the tissue structures contained in the cardiac ultrasound image, wherein the tissue structures include at least one target structure; Based on the location information of the tissue structure, a target region is determined in the cardiac ultrasound image for the at least one target structure.

2. The method for processing cardiac ultrasound images according to claim 1, characterized in that, The location information includes multiple positioning points of the tissue structure and location information of the outline of at least one tissue structure; the target area includes a color Doppler sampling area. The step of determining a target region in the cardiac ultrasound image for at least one target structure based on the location information of the tissue structure includes: Based on the imaging origin of the cardiac ultrasound image, and according to the location information of multiple positioning points of the tissue structures contained in the cardiac ultrasound image and the contour of at least one tissue structure, the boundary position of the region of interest of each target structure in the at least one target structure is determined in the cardiac ultrasound image. For each of the at least one target structures, the color Doppler sampling region of the target structure is determined based on the boundary position of the region of interest where the target structure is located.

3. The method for processing cardiac ultrasound images according to claim 2, characterized in that, The step of determining the boundary position of the region of interest for each of the at least one target structures in the cardiac ultrasound image, based on the imaging origin of the cardiac ultrasound image and according to the positional information of multiple positioning points of the tissue structures included in the cardiac ultrasound image and the contour of at least one tissue structure, includes: For each region of interest in the at least one target structure, Based on the imaging origin and the positional information of multiple positioning points of the tissue structures contained in the cardiac ultrasound image and the contour of at least one tissue structure, the positional information of multiple boundary reference lines of the region of interest is calculated. The multiple boundary reference lines include a left boundary reference line, a right boundary reference line, an upper boundary reference line, and a lower boundary reference line. The left boundary reference line and the right boundary reference line are rays passing through the imaging origin, respectively. The upper boundary reference line and the lower boundary reference line are arcs centered at the imaging origin. The boundary position of the region of interest is determined based on the position information of the multiple boundary reference lines.

4. The method for processing cardiac ultrasound images according to claim 3, characterized in that, The step of calculating the position information of multiple boundary reference lines of the region of interest based on the imaging origin and the position information of multiple positioning points of the tissue structures contained in the cardiac ultrasound image and the contour of at least one tissue structure includes: Based on the imaging origin and the location information of multiple positioning points of the tissue structures contained in the cardiac ultrasound image and the contour of at least one tissue structure, the location information of the boundary reference point of the region of interest is calculated. For each of the boundary reference points, the position information of the boundary reference line corresponding to the region of interest is determined based on the position information of the boundary reference point.

5. The method for processing cardiac ultrasound images according to claim 3, characterized in that, The step of calculating the position information of multiple boundary reference lines of the region of interest based on the imaging origin and the position information of multiple positioning points of the tissue structures contained in the cardiac ultrasound image and the contour of at least one tissue structure includes: Determine the position information of the tangent line that passes through the imaging origin and is tangent to the target contour in the contour of the at least one tissue structure, and use the position information of the tangent line as the position information of the boundary reference line.

6. The method for processing cardiac ultrasound images according to claim 1, characterized in that, The cardiac ultrasound image includes multiple target structures, the target region includes a color Doppler sampling region, and the method further includes: The color Doppler sampling regions of at least two of the multiple target structures are merged to obtain a color Doppler sampling region surrounding the at least two target structures.

7. The method for processing cardiac ultrasound images according to claim 1, characterized in that, The cardiac ultrasound images comprise N frames, where N is an integer greater than 1, and the target region includes a color Doppler sampling region. After determining the target region in each frame of echocardiography for the at least one target structure, the method further includes: For each of the at least one target structures, based on the color Doppler sampling regions of the target structure in at least a portion of the N frames of images, the color Doppler sampling region of the target structure in each subsequent frame of echocardiography is determined, such that the color Doppler sampling region in each subsequent frame of echocardiography includes the region corresponding to the position of the i-th region in the frame of echocardiography, where i includes all positive integers less than N+1, and the i-th region is the color Doppler sampling region of the target structure in the i-th frame of the N frames of images.

8. The method for processing cardiac ultrasound images according to claim 1, characterized in that, The cardiac ultrasound image includes multiple target structures, and the method further includes: In response to the user's selection operation, the target region of the target structure corresponding to the selection operation is displayed among the plurality of target structures.

9. The method for processing cardiac ultrasound images according to claim 1, characterized in that, The location information includes multiple positioning points of the tissue structure, and the target area includes the pulse Doppler sampling gate region. The step of segmenting the cardiac ultrasound image according to the determined section type to determine the location information of the tissue structures contained in the cardiac ultrasound image includes: Based on the section type, an image segmentation model is used to determine the confidence values ​​of the plurality of localization points of the tissue structure in the cardiac ultrasound image and the target points among the plurality of localization points; The step of determining a target region in the cardiac ultrasound image for at least one target structure based on the location information of the tissue structure includes: For each of the at least one target structures, the target region is determined based on the location point of the target structure and the confidence value of the target point.

10. The method for processing cardiac ultrasound images according to claim 9, characterized in that, The step of determining the target region for the target structure based on the location points of the target structure and the confidence values ​​of the target points includes: If the confidence value of the target point of the target structure is greater than the first confidence threshold, then the pulse Doppler sampling gate region of the target structure is determined based on the location information of the target point of the target structure. If the confidence value of the target point of the target structure is less than or equal to the first confidence threshold and greater than or equal to the second confidence threshold, then the pulse Doppler sampling gate region of the target structure is determined based on the plurality of positioning points of the target structure. If the confidence value of the target point of the target structure is less than the second confidence threshold, then the pulse Doppler sampling gate region of the target structure is determined based on the positioning points other than the target point among the plurality of positioning points of the target structure.

11. The method for processing cardiac ultrasound images according to any one of claims 1 to 10, characterized in that, The method further includes: In response to a user's adjustment operation on the target region, the size of the target region and / or its position in the cardiac ultrasound image are adjusted.

12. An ultrasonic device, comprising: Processor and memory, characterized in that, The memory stores computer program instructions, which, when executed by the processor, are used to perform the method for processing cardiac ultrasound images as described in any one of claims 1 to 11.

13. A storage medium on which program instructions are stored, characterized in that, The program instructions, when executed, are used to perform the method for processing cardiac ultrasound images as described in any one of claims 1 to 11.

14. A computer program product comprising computer program instructions, characterized in that, The computer program instructions, when executed, are used to perform the method for processing cardiac ultrasound images as described in any one of claims 1 to 11.

Images