Intravascular ultrasonic image processing method and device and ultrasonic system

Abstract

The invention provides an intravascular ultrasonic image processing method and device and an ultrasonic system. The processing method comprises the following steps: acquiring an intravascular ultrasonic image sequence of a target object; screening a target ultrasonic image in the intravascular ultrasonic image sequence at least based on the cardiac information corresponding to each intravascular ultrasonic image to obtain a target ultrasonic image sequence; for each preset axis in the at least two preset axes, converting the target ultrasonic image sequence into a blood vessel long-axis image corresponding to the preset axis; for each blood vessel long-axis image, determining a long-axis lumen boundary of the blood vessel in the blood vessel long-axis image; and for each target ultrasonic image, determining and marking the cross section lumen boundary of the blood vessel in the target ultrasonic image based on the respective long-axis lumen boundary in each blood vessel long-axis image. The accuracy of the lumen boundary of the cross section can be improved, and a user can be better assisted in obtaining the actual condition of the blood vessel of the target object.

Description

Technical Field

[0001] The present invention relates to the technical field of medical image processing, and more particularly to a method for processing intravascular ultrasound images, a device for processing intravascular ultrasound images, an ultrasound system, a storage medium, and a computer program product. Background Art

[0002] Intravascular Ultrasound (IVUS) is an intravascular imaging technology that can provide cross-sectional and longitudinal views of blood vessels and lumens. Furthermore, clinical measurement parameters such as the diameter and area of blood vessels and lumens, plaque burden, and eccentricity can be determined.

[0003] For example, intravascular ultrasound can send an intravascular ultrasound catheter into the distal end of a target blood vessel through a contrast catheter and then start retracting. During the retraction process, ultrasonic signals are sent and received in real time, and signal-related processing is performed to obtain the image information of the cross-section of the blood vessel and lumen. After the retraction is completed, multiple intravascular ultrasound images can be obtained. Then, by processing these ultrasound images, the lumen boundary in the ultrasound image can be known, and thus the user can be assisted in knowing the actual situation of the target blood vessel.

[0004] In related technologies, the accuracy of the lumen boundary in the generated intravascular ultrasound images is relatively low. Therefore, how to better generate the lumen boundary in ultrasound images to better assist users in obtaining relevant blood vessel information is a technical problem that those skilled in the art urgently need to solve. Summary of the Invention

[0005] In view of the above problems, the present invention is proposed. The present invention provides a method for processing intravascular ultrasound images, a device for processing intravascular ultrasound images, an ultrasound system, a storage medium, and a computer program product.

[0006] According to one aspect of the present invention, there is provided a method for processing intravascular ultrasound images, the processing method comprising: obtaining a sequence of intravascular ultrasound images of a target object, wherein the sequence of intravascular ultrasound images comprises a plurality of intravascular ultrasound images arranged in the acquisition order; screening the target ultrasound images in the sequence of intravascular ultrasound images based at least on the respective cardiac information corresponding to each intravascular ultrasound image, so as to obtain a sequence of target ultrasound images, wherein the cardiac information corresponding to each target ultrasound image is within a preset cardiac range, and the target ultrasound images in the sequence of target ultrasound images are arranged in the acquisition order; for each of at least two preset axes, converting the sequence of target ultrasound images into a vascular long-axis image corresponding to the preset axis, wherein the pixels located on the preset axis in the target ultrasound image correspond to a column of pixels in the vascular long-axis image; for each vascular long-axis image, determining the long-axis lumen boundary of the blood vessel in the vascular long-axis image; for each target ultrasound image, determining and marking the cross-sectional lumen boundary of the blood vessel in the target ultrasound image based on the long-axis lumen boundary in each vascular long-axis image.

[0007] Exemplarily, the processing method further comprises: determining the cross-sectional lumen boundary of the blood vessel in the non-selected ultrasound images in the sequence of intravascular ultrasound images based on the cross-sectional lumen boundary in each target ultrasound image, wherein the non-selected ultrasound images are other images in the sequence of intravascular ultrasound images that are different from the target ultrasound images.

[0008] Exemplarily, determining the cross-sectional lumen boundary of the blood vessel in the non-selected ultrasound images in the sequence of intravascular ultrasound images based on the cross-sectional lumen boundary in each target ultrasound image comprises:

[0009] For a first preset number of non-selected ultrasound images after the first ultrasound image in the sequence of intravascular ultrasound images, determining the cross-sectional lumen boundary of the non-selected ultrasound image based on the cross-sectional lumen boundary in the previous first reference image of the non-selected ultrasound image; and for a second preset number of non-selected ultrasound images before the second ultrasound image in the sequence of intravascular ultrasound images, determining the cross-sectional lumen boundary of the non-selected ultrasound image based on the cross-sectional lumen boundary in the next second reference image of the non-selected ultrasound image.

[0010] Wherein, the first ultrasound image and the second ultrasound image are two adjacent target ultrasound images in the sequence of target ultrasound images, the sum of the first preset number and the second preset number is equal to the total number of non-selected ultrasound images between the first ultrasound image and the second ultrasound image in the sequence of intravascular ultrasound images, and the first reference image and the second reference image are both intravascular ultrasound images in the sequence of intravascular ultrasound images for which the cross-sectional lumen boundary has been determined.

[0011] Exemplarily, based on the cross-sectional lumen boundaries in each of the respective target ultrasound images, determining the cross-sectional lumen boundaries of the blood vessels in the non-selected ultrasound images in the intravascular ultrasound image sequence includes:

[0012] For the adjacent third ultrasound image and fourth ultrasound image in the intravascular ultrasound image sequence, perform the following steps until the cross-sectional lumen boundary of the blood vessel in the fourth ultrasound image is determined, where the third ultrasound image is an intravascular ultrasound image for which the cross-sectional lumen boundary has been determined, and the fourth ultrasound image is a non-selected ultrasound image for which the cross-sectional lumen boundary is to be determined:

[0013] Based on a preset angular interval, determine a plurality of preset angles in a preset angular range, and for each preset angle, determine the image similarity between the third ultrasound image and the fourth ultrasound image rotated by that preset angle as the image similarity corresponding to that preset angle;

[0014] In the case where the maximum image similarity is greater than the similarity threshold, based on the cross-sectional lumen boundary in the third ultrasound image and the preset angle corresponding to the maximum image similarity, determine the cross-sectional lumen boundary in the fourth ultrasound image, and in the case where the maximum image similarity is less than or equal to the similarity threshold, reset the preset angular interval and / or the preset angular range based on preset conditions.

[0015] Exemplarily, the processing method further includes:

[0016] Determine a first region of interest in the third ultrasound image and a second region of interest in the fourth ultrasound image;

[0017] Determining the image similarity between the third ultrasound image and the fourth ultrasound image rotated by that preset angle as the image similarity corresponding to that preset angle includes:

[0018] Based on the pixels in the first region of interest and the pixels in the second region of interest rotated by that preset angle, determine the image similarity corresponding to that preset angle.

[0019] Exemplarily, screening the target ultrasound images in the intravascular ultrasound image sequence based at least on the respective cardiac information corresponding to each intravascular ultrasound image to obtain a target ultrasound image sequence includes:

[0020] Based on the intravascular ultrasound images in the intravascular ultrasound image sequence, detect and screen out the intravascular ultrasound images including contrast catheters in the intravascular ultrasound image sequence to obtain a first image sequence;

[0021] Based on the respective cardiac information corresponding to each intravascular ultrasound image in the first image sequence, screen the target ultrasound images in the first image sequence to obtain a target ultrasound image sequence.

[0022] Exemplarily, for each vascular long-axis image, determining the long-axis lumen boundary of the blood vessel in the vascular long-axis image includes:

[0023] For each vascular long-axis image, inputting the vascular long-axis image into a trained boundary detection model to obtain the long-axis lumen boundary in the vascular long-axis image, where the trained boundary detection model is trained based on vascular long-axis training images and lumen training boundaries corresponding to the vascular long-axis training images.

[0024] Exemplarily, for each vascular long-axis image, inputting the vascular long-axis image into a trained boundary detection model to obtain the long-axis lumen boundary in the vascular long-axis image includes:

[0025] For each vascular long-axis image, if the vascular long-axis image does not conform to a preset standard size, performing a scaling process on the vascular long-axis image based on the preset standard size to obtain a standardized vascular long-axis image, inputting the standardized vascular long-axis image into the trained boundary detection model, and performing an inverse process of the scaling process on the lumen boundary output by the trained boundary detection model to obtain the long-axis lumen boundary in the vascular long-axis image.

[0026] According to another aspect of the present invention, there is also provided a processing device for intravascular ultrasound images, the processing device including: an image acquisition module, a target ultrasound image screening module, a long-axis image generation module, a long-axis lumen boundary determination module, and a cross-sectional lumen boundary determination module.

[0027] The image acquisition module is configured to acquire a sequence of intravascular ultrasound images of a target object, where the sequence of intravascular ultrasound images includes a plurality of intravascular ultrasound images arranged in the acquisition order;

[0028] The target ultrasound image screening module is configured to screen the target ultrasound images in the sequence of intravascular ultrasound images based at least on the respective cardiac motion information corresponding to each intravascular ultrasound image to obtain a sequence of target ultrasound images, where the cardiac motion information corresponding to each target ultrasound image is within a preset cardiac motion range, and the target ultrasound images in the sequence of target ultrasound images are arranged in the acquisition order;

[0029] The long-axis image generation module is configured to, for each of at least two preset axes, convert the sequence of target ultrasound images into a vascular long-axis image corresponding to the preset axis, where the pixels in the target ultrasound image located on the preset axis correspond to a column of pixels in the vascular long-axis image;

[0030] The long-axis lumen boundary determination module is configured to, for each vascular long-axis image, determine the long-axis lumen boundary of the blood vessel in the vascular long-axis image;

[0031] A cross-sectional lumen boundary determination module is configured to determine and mark the cross-sectional lumen boundary of a blood vessel in each target ultrasound image based on the long-axis lumen boundary in each respective blood vessel long-axis image for each target ultrasound image.

[0032] According to another aspect of the present invention, there is also provided an ultrasound system including a memory and a processor, wherein: the memory is used to store a computer program; the processor is used to execute the computer program to implement the above-mentioned method for processing intravascular ultrasound images.

[0033] According to yet another aspect of the present invention, there is also provided a storage medium storing a computer program / instructions, and the computer program / instructions are used to execute the above-mentioned method for processing intravascular ultrasound images when running.

[0034] According to another aspect of the present invention, there is also provided a computer program product including computer program instructions, and the computer program instructions are used to execute the above-mentioned method for processing intravascular ultrasound images when run by a processor.

[0035] According to the above solution of the embodiment of the present invention, intravascular ultrasound images can be screened based on cardiac information to obtain target ultrasound images. Since the cardiac information of the target ultrasound images is all within a preset cardiac range, the long-axis lumen boundary in the obtained blood vessel long-axis image is less affected by cardiac changes, and its long-axis lumen boundary will also be smoother, more in line with the actual extension of the blood vessel, which is beneficial to improving the accuracy of the subsequent cross-sectional lumen boundary and can better assist the user in knowing the actual situation of the blood vessel of the target object. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] By describing the embodiments of the present invention in more detail in conjunction with the drawings, the above and other objects, features, and advantages of the present invention will become more obvious. The drawings are used to provide a further understanding of the embodiments of the present invention, and constitute a part of the specification, and are used to explain the present invention together with the embodiments of the present invention, and do not constitute a limitation to the present invention. In the drawings, the same reference numerals generally represent the same components or steps.

[0037] Figure 1 Shows a schematic flowchart of a method for processing intravascular ultrasound images according to an embodiment of the present invention;

[0038] Figure 2 Shows a schematic diagram of a cardiac signal according to an embodiment of the present invention;

[0039] Figure 3 Shows a schematic diagram of a converted blood vessel long-axis image according to an embodiment of the present invention;

[0040] Figure 4 Shows a schematic diagram of a long-axis lumen boundary according to an embodiment of the present invention;

[0041] Figure 5 Shows a schematic diagram of generating a cross-sectional lumen boundary according to an embodiment of the present invention;

[0042] Figure 6 Shows a schematic diagram of a first ultrasound image, a second ultrasound image, and a non-selected ultrasound image according to an embodiment of the present invention;

[0043] Figure 7 Shows a schematic block diagram of a processing device for intravascular ultrasound images according to an embodiment of the present invention; and

[0044] Figure 8 Shows a schematic block diagram of an ultrasound system according to an embodiment of the present invention. Detailed Description of the Invention

[0045] In order 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 only a part of the embodiments of the present invention, rather than all of the embodiments of the present invention. It should be understood that the present invention is not limited by the exemplary embodiments described herein. Based on the embodiments of the present invention described herein, all other embodiments obtained by those skilled in the art without creative efforts shall fall within the protection scope of the present invention.

[0046] Due to the influence of the movement of the heart, blood vessels, catheters, etc., there may be a lot of interference information in the intravascular ultrasound image sequence, and the accuracy of the blood vessel contour in the intravascular ultrasound image generated by the related technology is relatively low.

[0047] In order to at least partially solve the above problems, an embodiment of the present invention provides a method for processing intravascular ultrasound images. Figure 1 Shows a schematic flowchart of a method for processing intravascular ultrasound images according to an embodiment of the present invention. As Figure 1 shown, the method may include the following steps S110 to S150.

[0048] In step S110, an intravascular ultrasound image sequence of a target object is acquired.

[0049] The above target object can be a subject for whom intravascular ultrasound images need to be collected. For example, a contrast catheter can be punctured into the femoral artery or radial artery of the subject and extended to the coronary ostium of the heart. Then, a guiding catheter is extended through the contrast catheter to the target blood vessel to assist in determining the orientation of the ultrasound catheter (also known as an IVUS catheter, where IVUS stands for Intravascular Ultrasound). The ultrasound catheter can be sent along the guiding catheter to the target blood vessel and continuously or at preset time intervals emit ultrasonic waves and continuously receive or at preset time intervals receive the echoes of the ultrasonic waves during the process of retracting and rotating along the guiding catheter. Based on the echoes received at different times, an ultrasonic image corresponding to each time can be generated through an image reconstruction algorithm in related technologies as the intravascular ultrasound image. The intravascular ultrasound image sequence can include a plurality of intravascular ultrasound images arranged in the acquisition order. Specifically, the above acquisition order can be the acquisition order of the echoes for generating the intravascular ultrasound images. It should be understood that for different target blood vessels, the puncture position and extension direction of the contrast catheter can be different, which can be determined according to the actual needs of the user, and the embodiments of the present invention do not limit this here.

[0050] Exemplarily, since the ultrasound catheter collects images sequentially along the blood vessel axis, the positional morphologies of the plurality of intravascular ultrasound images arranged in the acquisition order basically correspond to the positional morphology of the blood vessel in the long axis direction. Therefore, the blood vessel state of the target object can be evaluated based on the intravascular ultrasound image sequence of the target object. For example, subsequent processing can be performed to obtain the lumen boundary of the blood vessel, etc.

[0051] In step S120, at least based on the cardiac information corresponding to each intravascular ultrasound image, the target ultrasound images in the intravascular ultrasound image sequence are screened to obtain a target ultrasound image sequence.

[0052] The above cardiac information can be used to represent the cardiac state of the target object, and different times or time periods during a cardiac cycle can be used as one of the cardiac information. The above cardiac cycle can be used to represent the time period during which the heart undergoes one diastole and systole.

[0053] In some embodiments, cardiac signals can be obtained through a heart rate monitor. The device for collecting intravascular ultrasound images can be synchronized with the time of the heart rate monitor so as to continuously obtain cardiac signals while continuously collecting intravascular ultrasound images. The cardiac states at different moments can be represented by the cardiac signals at different moments. Since the device for collecting intravascular ultrasound images has been synchronized with the time of the heart rate monitor, the cardiac state corresponding to the acquisition moment of each intravascular ultrasound image can be determined. Specifically, for example, the cardiac period can include the period corresponding to the P wave (the waveform corresponding to atrial contraction), the period corresponding to the QRS wave (the waveform corresponding to ventricular contraction), and the period corresponding to the T wave (the waveform corresponding to ventricular relaxation) in the cardiac signal (depending on different definitions, it may also include the flat waveforms between the above waveforms). The preset cardiac range can be any moment or period in the cardiac period.

[0054] In still other embodiments, the cardiac information corresponding to the intravascular ultrasound images can be determined through the cardiac recognition algorithms or models in the related art. Specifically, for example, the sequence of intravascular ultrasound images can be input into the cardiac recognition algorithms or models to establish the relationship between each intravascular ultrasound image in the sequence of intravascular ultrasound images and the cardiac information.

[0055] It should be understood that the above cardiac information can be used to represent the cardiac state at a specific moment or period, or can also be used to represent the generally described cardiac state. For example, the cardiac information can be used to represent the cardiac state at the start moment of the generation of the P wave, and the cardiac state within 0.3 seconds from the start moment of the generation of the T wave. For another example, the cardiac information can be used to represent the cardiac state during atrial systole, the cardiac state during ventricular systole, the cardiac state during ventricular diastole, etc.

[0056] The cardiac information corresponding to each target ultrasound image can all be within the preset cardiac range, and the target ultrasound images in the target ultrasound image sequence are arranged in the acquisition order.

[0057] It should be understood that the intravascular ultrasound images can also be screened in combination with other preset conditions. The above preset conditions can include: the imaging clarity is higher than the clarity threshold, the acquisition moment is within the time period threshold, etc. The specific values of the above clarity threshold and time period threshold can be set by the user according to the actual situation, and the embodiments of the present invention do not limit this here.

[0058] Refer to Figure 2 , Figure 2 shows a schematic diagram of a cardiac signal according to an embodiment of the present invention. In combination with Figure 2 , for each cardiac period, it can include a moment or period that meets the preset cardiac range, and the intravascular ultrasound image collected at this moment or period can be used as the above target ultrasound image. The preset cardiac range can be set according to the actual needs of the user, and the embodiments of the present invention do not limit this here.

[0059] In step S130, for each of at least two preset axes, the target ultrasound image sequence is converted into a vessel long-axis image corresponding to the preset axis.

[0060] Refer to Figure 3 , Figure 3 , which shows a schematic diagram of converting a vessel long-axis image according to an embodiment of the present invention. Combining Figure 3 , an ultrasound catheter (IVUS catheter can be referred to) can be inserted into the distal end of the target vessel and then withdrawn into the angiography catheter (the catheter is inserted into the distal end and withdrawn to the starting point can be referred to). During the withdrawal process, multiple intravascular ultrasound images can be obtained as the intravascular ultrasound image sequence.

[0061] Due to the periodic movement of the heart and the relative movement between the vessel and the ultrasound catheter, there will be repeated intravascular ultrasound images in the intravascular ultrasound image sequence. Specifically, these repeated intravascular ultrasound images are generated based on the echoes of ultrasonic waves reflected from the same position. Therefore, there is redundant information in the vessel long-axis image obtained from the intravascular ultrasound image sequence, which is not conducive to improving the processing efficiency of intravascular ultrasound images. In addition, the vessel contour presented in the vessel long-axis image will also show a serrated shape, which obviously does not conform to the actual extension of the vessel. In the embodiments of the present invention, based on the cardiac information, the intravascular ultrasound images are screened to obtain the target ultrasound images. The cardiac information of the target ultrasound images is all within the preset cardiac range. Therefore, the long-axis lumen boundary in the vessel long-axis image obtained accordingly is less affected by cardiac changes, and its long-axis lumen boundary will be smoother and more in line with the actual extension of the vessel.

[0062] The preset axis can be any axis in the plane of the intravascular ultrasound image. The preset axis can intersect with the catheter withdrawal direction. This axis can pass through the center of the intravascular ultrasound image. In one example, the preset axis can be 4, such as the axes of 0 degrees, 45 degrees, 90 degrees, and 135 degrees. The measurement standard of the above angles can be the angle between the preset axis and the horizontal axis in the intravascular ultrasound image. It should be understood that the specific angles, positions, etc. of the preset axis can be flexibly set by the user according to actual needs, and the embodiments of the present invention do not limit this here.

[0063] The pixels located on the preset axis in the target ultrasound image correspond to a column of pixels in the vessel long-axis image. For easy understanding, take Figure 3 the first preset axis and the second preset axis in as an example. The first preset axis is the vertical axis in this example, and the second preset axis is the horizontal axis in this example. The pixels located on the first preset axis can correspond to a column of pixels in the vessel long-axis image. The pixels located on the second preset axis can correspond to a column of pixels in another vessel long-axis image.

[0064] The correspondence between the pixels on the preset axis in the target ultrasound image and a column of pixels in the blood vessel long-axis image can be determined based on the acquisition order of the target ultrasound image. For example, the number of columns is positively or negatively correlated with the acquisition order. Specifically, for example, the pixels on the preset axis in the first target ultrasound image can correspond to the first column of pixels in the blood vessel long-axis image corresponding to the preset axis. The pixels on the preset axis in the last target ultrasound image can correspond to the last column of pixels in the blood vessel long-axis image corresponding to the preset axis. Therefore, based on the acquisition order of the target ultrasound image, the pixels on the preset axis in each target ultrasound image can be mapped to the blood vessel long-axis image. For each preset axis, a blood vessel long-axis image corresponding to the preset axis can be generated. In one example, the pixel values of the pixels in the blood vessel long-axis image can be the same as or similar to the pixel values of the corresponding pixels in the target ultrasound image, which is not limited in the embodiments of the present invention.

[0065] In step S140, for each blood vessel long-axis image, determine the long-axis lumen boundary of the blood vessel in the blood vessel long-axis image.

[0066] The above-mentioned method for determining the long-axis lumen boundary can be determined by an edge extraction algorithm or an edge extraction model in related technologies, which is not limited in the embodiments of the present invention. The above-mentioned long-axis lumen boundary can be used to represent the boundary of the inner wall of the blood vessel in the blood vessel long-axis image. Since the blood vessel wall has a certain thickness, the long-axis lumen boundary can also be used to represent the boundary of the outer wall of the blood vessel in the blood vessel long-axis image.

[0067] Refer to Figure 4 , Figure 4 shows a schematic diagram of the long-axis lumen boundary according to an embodiment of the present invention. Combining Figure 4 , different blood vessel long-axis images corresponding to different preset axes (4 preset axes in the example of Figure 4 ) can be generated based on the target ultrasound image sequence. Then, determine the long-axis lumen boundary in each blood vessel long-axis image corresponding to each preset axis (such as the long-axis lumen boundary tracing in Figure 4 ).

[0068] In step S150, for each target ultrasound image, based on the long-axis lumen boundary in each blood vessel long-axis image, determine and mark the cross-sectional lumen boundary of the blood vessel in the target ultrasound image.

[0069] Since there is a correspondence between the pixels on the preset axis in the target ultrasound image and the pixels in the blood vessel long-axis image, the pixel points on the long-axis lumen boundary can be mapped to the target ultrasound image to obtain the pixel points on the cross-sectional lumen boundary.

[0070] Refer to Figure 5 , Figure 5Shows a schematic diagram of generating a cross-sectional lumen boundary provided by an embodiment of the present invention. In combination with Figure 5 , a column of pixels in the long-axis image of the blood vessel corresponding to the preset axis can correspond to the pixels located on the preset axis in the target ultrasound image. Therefore, the pixels on the long-axis lumen boundary in the long-axis images of the blood vessel corresponding to the four preset axes (i.e., cross-sections 1 to 4) can be mapped to the target ultrasound image. Specifically, for each cross-section, there are two intersecting pixels between a column of pixels and the long-axis lumen boundary, and these two pixels are the lumen boundaries of the blood vessel. These two pixels can be mapped to the target ultrasound image to obtain two pixels located on the cross-sectional lumen boundary. The more preset axes there are, the more pixels on the cross-sectional lumen boundary can be mapped, so the obtained cross-sectional lumen boundary can be more accurate. In this example, each preset axis can determine two pixels located on the cross-sectional lumen boundary. Therefore, the long-axis images of the blood vessel corresponding to the four preset axes can determine 8 pixels on the cross-sectional lumen boundary in a target ultrasound image. By performing curve fitting on the obtained pixels on the cross-sectional lumen boundary through an interpolation algorithm, the closed cross-sectional lumen boundary in the target ultrasound image can be obtained. In this example, only one target ultrasound image of the distal end of the blood vessel, one target ultrasound image of the middle end of the blood vessel, and one target ultrasound image of the proximal end of the blood vessel are shown. However, it should be understood that each column of pixels in the long-axis image of the blood vessel can determine a cross-sectional lumen boundary in a target ultrasound image. Preferably, the cubic spline interpolation method can be used to generate a complete cross-sectional lumen boundary based on the position information of the obtained pixels on the cross-sectional lumen boundary (such as the 8 pixels on the cross-sectional lumen boundary mentioned above) in the target ultrasound image. The cubic spline interpolation method can require only less computational amount and storage amount to obtain an accurate curve.

[0071] According to the above solution of the embodiment of the present invention, the intravascular ultrasound images can be screened based on the cardiac motion information to obtain the target ultrasound images. Since the cardiac motion information of the target ultrasound images is all within the preset cardiac motion range, the long-axis lumen boundary in the obtained long-axis image of the blood vessel is less affected by cardiac motion changes, and its long-axis lumen boundary will also be smoother and more in line with the actual extension of the blood vessel, which is beneficial to improving the accuracy of the subsequent cross-sectional lumen boundary and can better assist the user in knowing the actual situation of the blood vessel of the target object.

[0072] Exemplarily, step S120, screening the target ultrasound images in the intravascular ultrasound image sequence based at least on the cardiac motion information corresponding to each intravascular ultrasound image to obtain a target ultrasound image sequence, may include: step S121 and step S122.

[0073] In step S121, based on the intravascular ultrasound images in the intravascular ultrasound image sequence, detect and screen out the intravascular ultrasound images including the contrast catheter in the intravascular ultrasound image sequence to obtain a first image sequence.

[0074] In combination with the actual situation, during the retraction process of the ultrasound catheter, ultrasonic waves are continuously emitted and echoes are received. Eventually, the ultrasound catheter will retract into the contrast catheter. Therefore, there is a part of intravascular ultrasound images in the intravascular ultrasound image sequence that users do not pay attention to. These intravascular ultrasound images that are not of concern may include ultrasound images within the contrast catheter. Based on the image recognition algorithms or models in related technologies, it can be determined whether an intravascular ultrasound image is an intravascular ultrasound image including the contrast catheter. For example, if the number of pixels with pixel values close to pure black (e.g., close to (0, 0, 0), and the specific range can be flexibly set by the user) in the intravascular ultrasound image exceeds the pixel quantity threshold, then this intravascular ultrasound image can be regarded as an image generated after the ultrasound catheter has retracted into the contrast catheter. The contour in this intravascular ultrasound image is not the cavity contour of the blood vessel, but the cavity contour of the contrast catheter. In some embodiments, if one or at least a continuous first threshold number of intravascular ultrasound images are detected as intravascular ultrasound images including the contrast catheter, then all the subsequent intravascular ultrasound images can be regarded as intravascular ultrasound images including the contrast catheter and these images can be screened out. The specific values of the above-mentioned pixel quantity threshold and the first threshold can be determined according to the actual needs of the user, and the embodiments of the present invention do not limit them here.

[0075] In step S122, based on the respective cardiac motion information corresponding to each intravascular ultrasound image in the first image sequence, the target ultrasound images in the first image sequence are screened to obtain a target ultrasound image sequence.

[0076] The screening process of the intravascular ultrasound images in the first image sequence based on the cardiac motion information can refer to the relevant steps in step S120, and the embodiments of the present invention will not elaborate here.

[0077] According to the above solution of the embodiments of the present invention, based on the intravascular ultrasound images in the intravascular ultrasound image sequence, the intravascular ultrasound images including the contrast catheter in the intravascular ultrasound image sequence can be detected and screened out to obtain a first image sequence. Then, based on the respective cardiac motion information corresponding to each intravascular ultrasound image in the first image sequence, the target ultrasound images in the first image sequence are screened to obtain a target ultrasound image sequence. The above solution takes into account the actual scenario where the ultrasound catheter may retract into the contrast catheter. Compared with the related technology of directly using the generated intravascular ultrasound image sequence to obtain the lumen boundary, the data volume of the first image sequence in the embodiments of the present invention is smaller, which is beneficial to improving the generation efficiency of the cross-sectional lumen boundary and more meets the actual needs of doctors.

[0078] Exemplarily, in step S140, for each vascular long-axis image, determining the long-axis lumen boundary of the blood vessel in the vascular long-axis image may include: for each vascular long-axis image, inputting the vascular long-axis image into the trained boundary detection model to obtain the long-axis lumen boundary in the vascular long-axis image.

[0079] The trained boundary detection model is trained based on the vascular long-axis training images and the lumen training boundaries corresponding to the vascular long-axis training images. The network architecture of the above-mentioned boundary detection model is not limited in the embodiments of the present invention. For example, a convolutional neural network architecture, a U-Net architecture, etc. can be used. Specifically, for example, the training method of the boundary detection model can be as follows: Obtain vascular long-axis training images. In one example, the vascular long-axis training images can be generated based on an intravascular ultrasound training image sequence. The above-mentioned intravascular ultrasound training images can be intravascular ultrasound images collected for the blood vessels of a training object (which can be the same as or different from the target object). Arranging them in the acquisition order of the intravascular ultrasound training images can obtain an intravascular ultrasound training image sequence. If the intravascular ultrasound images of the target object have undergone the processing of step S121, the relevant steps of step S121 in the above text can also be referred to for the intravascular ultrasound training image sequence, and the intravascular ultrasound training images including the contrast catheter in the intravascular ultrasound training image sequence are detected and screened out to obtain a second image sequence. The target ultrasound training images can be screened out based on the cardiac motion information of each intravascular ultrasound training image in the intravascular ultrasound training image sequence or the second image sequence in the manner of step S120 in the above text to obtain a third image sequence. The intravascular ultrasound training images in the third image sequence can be converted into the vascular long-axis training images corresponding to the preset axis based on the preset axis in the manner of step S130 in the above text. The lumen training boundaries in the vascular long-axis training images can be labeled manually or by relevant algorithms. The untrained boundary detection model can generate lumen prediction boundaries based on the input vascular long-axis training images, and at least use the difference between the lumen training boundaries and the lumen prediction boundaries as one of the loss values. After training with multiple vascular long-axis training images (which can be generated for different training objects respectively or corresponding to different preset axes), the trained boundary detection model can be obtained.

[0080] In one example, the long-axis training image of the blood vessel can be standardized, and then the long-axis training image of the blood vessel adjusted to a preset standard size can be used for training. It should be understood that before generating the long-axis lumen boundary of the blood vessel long-axis image, the above standardization can also be performed on the blood vessel long-axis image, and then the long-axis image of the blood vessel adjusted to the above preset standard size can be used to generate the long-axis lumen boundary. Among them, the preset standard size can be a square size, for example: 256×256. Since the blood vessel long-axis image is an image corresponding to the blood vessel morphology, generally a long-shaped image, this is not convenient for the model to extract features. Through the above method, it is beneficial to reduce the irrelevant influence on the model to generate the long-axis lumen boundary due to the image size. Exemplarily, for each blood vessel long-axis image, the blood vessel long-axis image is input into the trained boundary detection model to obtain the long-axis lumen boundary in the blood vessel long-axis image, including: for each blood vessel long-axis image, if the blood vessel long-axis image does not conform to the preset standard size, then based on the preset standard size, the blood vessel long-axis image is scaled to obtain a standardized blood vessel long-axis image, the standardized blood vessel long-axis image is input into the trained boundary detection model, and the inverse process of the scaling process is performed on the lumen boundary output by the trained boundary detection model to obtain the long-axis lumen boundary in the blood vessel long-axis image. Of course, if the blood vessel long-axis image conforms to the preset standard size, then the blood vessel long-axis image is directly input into the trained boundary detection model, and the lumen boundary output by the trained boundary detection model is determined as the long-axis lumen boundary in the blood vessel long-axis image.

[0081] According to the above solution of the embodiment of the present invention, for each blood vessel long-axis image, the blood vessel long-axis image can be input into the trained boundary detection model to obtain the long-axis lumen boundary in the blood vessel long-axis image, so as to automatically and accurately generate the long-axis lumen boundary, which is beneficial to improving the accuracy of the cross-sectional lumen boundary generated based on the long-axis lumen boundary.

[0082] Exemplarily, the above processing method may further include step S210.

[0083] In step S210, based on the cross-sectional lumen boundary in each respective target ultrasound image, the cross-sectional lumen boundary of the blood vessel in the non-selected ultrasound images in the intravascular ultrasound image sequence is determined.

[0084] The unselected ultrasound images are other images in the intravascular ultrasound image sequence that are different from the target ultrasound image. In one example, for any two consecutive target ultrasound images in the intravascular ultrasound image sequence, the unselected ultrasound images after the first target ultrasound image and before the last target ultrasound image among the two target ultrasound images can have their cross-sectional lumen boundaries determined based on the cross-sectional lumen boundary in the first target ultrasound image in sequence (in the acquisition order). For example, based on the similarity between the unselected ultrasound image and the intravascular ultrasound image A (which can be the first target ultrasound image in the above example or an unselected ultrasound image whose cross-sectional lumen boundary has been determined) before the unselected ultrasound image, the rotation angle of the unselected ultrasound image relative to the intravascular ultrasound image A can be determined. Then, based on this rotation angle, the pixels on the cross-sectional lumen boundary in the intravascular ultrasound image A are rotationally mapped to serve as the pixels on the cross-sectional lumen boundary in the unselected ultrasound image. It is also possible to only rotationally map the preset pixels in the intravascular ultrasound image A in the above text, and then perform curve fitting through the interpolation algorithm in the above text to obtain the cross-sectional lumen boundary in the unselected ultrasound image. The above preset pixels can be any pixels on the cross-sectional lumen boundary. Further, the above preset pixels can be the pixels on the intravascular ultrasound image A located on the preset axis, and the pixels on this preset axis are directly obtained by mapping the pixels on the long-axis lumen boundary. Therefore, the accuracy of the cross-sectional lumen boundary in the unselected ultrasound image obtained by performing curve fitting based on this preset pixel is higher.

[0085] In another example, the unselected ultrasound images after the first target ultrasound image and before the last target ultrasound image among the two target ultrasound images can also have their cross-sectional lumen boundaries determined based on the cross-sectional lumen boundary in the last target ultrasound image in sequence (in the reverse order of the acquisition order). The specific process can refer to the above example, and the embodiments of the present invention will not elaborate here.

[0086] It should be understood that the intravascular ultrasound images including the contrast catheter in the intravascular ultrasound image sequence can also be screened with reference to step S121 in the above text. In other words, the above unselected ultrasound images can also be other images in the first image sequence that are different from the target ultrasound image.

[0087] According to the above solution of the embodiments of the present invention, the cross-sectional lumen boundaries of the blood vessels in the unselected ultrasound images in the intravascular ultrasound image sequence can be determined based on the cross-sectional lumen boundaries in each target ultrasound image respectively. The above solution can provide the user with more intravascular ultrasound images with cross-sectional lumen boundaries, and can better assist the user in understanding the actual situation of the blood vessels of the target object.

[0088] Exemplarily, step S210 of determining the cross-sectional lumen boundary of the blood vessel in the non-selected ultrasound images in the intravascular ultrasound image sequence based on the cross-sectional lumen boundary in each respective target ultrasound image may include: for a first preset number of non-selected ultrasound images after the first ultrasound image in the intravascular ultrasound image sequence, determining the cross-sectional lumen boundary of the non-selected ultrasound image based on the cross-sectional lumen boundary in the previous first reference image of the non-selected ultrasound image; and for a second preset number of non-selected ultrasound images before the second ultrasound image in the intravascular ultrasound image sequence, determining the cross-sectional lumen boundary of the non-selected ultrasound image based on the cross-sectional lumen boundary in the next second reference image of the non-selected ultrasound image.

[0089] The first reference image may be an intravascular ultrasound image in the intravascular ultrasound image sequence for which the cross-sectional lumen boundary has been determined. For example, the first reference image may be a non-selected ultrasound image after the first ultrasound image for which the cross-sectional lumen boundary has been determined or the first ultrasound image. In some examples, if the intravascular ultrasound images including the contrast catheter in the intravascular ultrasound image sequence are screened out with reference to step S121 above, the non-selected ultrasound images after the first ultrasound image may also be the non-selected ultrasound images after the first ultrasound image in the first image sequence, and the first reference image may be an intravascular ultrasound image in the first image sequence for which the cross-sectional lumen boundary has been determined.

[0090] The second reference image may be an intravascular ultrasound image in the intravascular ultrasound image sequence for which the cross-sectional lumen boundary has been determined. For example, the second reference image may be a non-selected ultrasound image before the second ultrasound image for which the cross-sectional lumen boundary has been determined or the second ultrasound image. In some examples, if the intravascular ultrasound images including the contrast catheter in the intravascular ultrasound image sequence are screened out with reference to step S121 above, the non-selected ultrasound images before the second ultrasound image may also be the non-selected ultrasound images before the second ultrasound image in the first image sequence, and the second reference image may be an intravascular ultrasound image in the first image sequence for which the cross-sectional lumen boundary has been determined.

[0091] The first ultrasound image and the second ultrasound image may be two adjacent target ultrasound images in the target ultrasound image sequence. For example, the acquisition order of the first ultrasound image may be before that of the second ultrasound image.

[0092] The sum of the first preset quantity and the second preset quantity is equal to the total number of non - selected ultrasound images between the first ultrasound image and the second ultrasound image in the intravascular ultrasound image sequence. In some examples, if the intravascular ultrasound images including the contrast catheter in the intravascular ultrasound image sequence are screened according to step S121 in the above text, the sum of the first preset quantity and the second preset quantity can also be equal to the total number of non - selected ultrasound images between the first ultrasound image and the second ultrasound image in the first image sequence.

[0093] Refer to Figure 6 , Figure 6 which shows a schematic diagram of the first ultrasound image, the second ultrasound image, and the non - selected ultrasound images according to an embodiment of the present invention. Combining Figure 6 , the first ultrasound image, the non - selected ultrasound images after the first ultrasound image in the first group, and the non - selected ultrasound images before the second ultrasound image in the second group can be located within the same cardiac cycle. The second ultrasound image, Figure 6 the other non - selected ultrasound images in Figure 6 can be located within another cardiac cycle, and the intravascular ultrasound images after the other non - selected ultrasound images (refer to the last ellipse) can be located within another cardiac cycle. In this example, both the first preset quantity and the second preset quantity are 2. It should be understood that in some examples, the specific values of the first preset quantity and the second preset quantity can be flexibly set by developers. In addition, during the actual acquisition process, the total number of non - selected ultrasound images between the first ultrasound image and the second ultrasound image can also be evenly divided directly to quickly determine the first preset quantity and the second preset quantity. The first reference image of the first non - selected ultrasound image after the first ultrasound image in the first group can be the first ultrasound image, and the first reference image of the second non - selected ultrasound image after the first ultrasound image can be the first non - selected ultrasound image after the first ultrasound image. The second reference image of the first non - selected ultrasound image before the second ultrasound image in the second group can be the second ultrasound image, and the second reference image of the second non - selected ultrasound image before the second ultrasound image can be the first non - selected ultrasound image before the second ultrasound image.

[0094] Combined with the actual scenario, if the cardiac motion information of multiple target ultrasound images in the same cardiac motion period meets the preset cardiac motion range, the first ultrasound image and the second ultrasound image can be two adjacent target ultrasound images in the target ultrasound image sequence, and there is at least one non-selected ultrasound image between these two target ultrasound images in the intravascular ultrasound image sequence or the first image sequence. Specifically, for example, the first ultrasound image can be the last target ultrasound image among the continuously arranged target ultrasound images in the intravascular ultrasound image sequence or the first image sequence, and the second ultrasound image can be the first target ultrasound image among the continuously arranged target ultrasound images in the intravascular ultrasound image sequence or the first image sequence, and the first ultrasound image and the second ultrasound image are adjacent in the target ultrasound image sequence.

[0095] According to the above solution of the embodiment of the present invention, for the first preset number of non-selected ultrasound images after the first ultrasound image in the intravascular ultrasound image sequence, based on the cross-sectional lumen boundary in the previous first reference image of the non-selected ultrasound image, the cross-sectional lumen boundary of the non-selected ultrasound image can be determined. And for the second preset number of non-selected ultrasound images before the second ultrasound image in the intravascular ultrasound image sequence, based on the cross-sectional lumen boundary in the next second reference image of the non-selected ultrasound image, the cross-sectional lumen boundary of the non-selected ultrasound image can be determined. The above solution can determine the cross-sectional lumen boundary based on the first reference image or the second reference image adjacent to the non-selected ultrasound image, so the distribution of the cross-sectional lumen boundary conforms to the actual blood vessel extension of the target object (for example, if a blood vessel long-axis image is established based on this, the long-axis lumen boundary therein will be smoother), which is beneficial to improving the accuracy of the cross-sectional lumen boundary.

[0096] Exemplarily, step S210, determining the cross-sectional lumen boundary of the blood vessel in the non-selected ultrasound images in the intravascular ultrasound image sequence based on the cross-sectional lumen boundary in each target ultrasound image respectively, may include: for the adjacent third ultrasound image and fourth ultrasound image in the intravascular ultrasound image sequence, perform the following steps S211 and S212 until the cross-sectional lumen boundary of the blood vessel in the fourth ultrasound image is determined.

[0097] The third ultrasound image is an intravascular ultrasound image for which the cross-sectional lumen boundary has been determined. For example, the third ultrasound image can be the first reference image or the second reference image in the above text. The fourth ultrasound image can be a non-selected ultrasound image for which the cross-sectional lumen boundary is to be determined. It should be understood that the third ultrasound image and the fourth ultrasound image can also be adjacent ultrasound images in the first image sequence in the above text.

[0098] In step S211, based on a preset angular interval, within a preset angular range, multiple preset angles are determined, and for each preset angle, the image similarity between the third ultrasonic image and the fourth ultrasonic image after rotating the preset angle is determined as the image similarity corresponding to the preset angle.

[0099] The above-mentioned preset angular interval and preset angular range can be set by the user according to the actual situation, and the embodiments of the present invention do not limit this here. For example, if the preset angular range is from -45 degrees to 45 degrees and the preset angular interval is 20 degrees, then the following multiple preset angles can be determined: -45 degrees, -25 (-45 + 20) degrees, -5 (-25 + 20) degrees, 15 (-5 + 20) degrees, 35 (15 + 20) degrees.

[0100] Based on the preset angle and the fourth ultrasonic image, the rotated fourth ultrasonic image can be determined. For example, the abscissa and ordinate of the pixels on the fourth ultrasonic image can be multiplied by the cosine and sine of the preset angle respectively to obtain the rotated fourth ultrasonic image. In another example, a part of the image in the fourth ultrasonic image can also be rotated to reduce the computational amount and improve the generation efficiency of the cross-sectional lumen boundary.

[0101] The above-mentioned image similarity can be determined through similarity generation algorithms and similarity generation models in related technologies, and the embodiments of the present invention do not elaborate on this here.

[0102] In step S212, when the maximum image similarity is greater than the similarity threshold, based on the cross-sectional lumen boundary in the third ultrasonic image and the preset angle corresponding to the maximum image similarity, the cross-sectional lumen boundary in the fourth ultrasonic image is determined, and when the maximum image similarity is less than or equal to the similarity threshold, the preset angular interval and / or preset angular range are reset based on preset conditions.

[0103] Based on the preset angle corresponding to the maximum image similarity, the pixels on the cross-sectional lumen boundary in the third ultrasonic image can be rotationally mapped as the pixels on the cross-sectional lumen boundary in the fourth ultrasonic image. It is also possible to only rotationally map the preset pixels in the third ultrasonic image mentioned above, and then perform curve fitting through the interpolation algorithm mentioned above to obtain the cross-sectional lumen boundary in the fourth ultrasonic image. The above-mentioned preset pixels can be any pixels located on the cross-sectional lumen boundary in the third ultrasonic image. Further, the above-mentioned preset pixels can be the pixels located on the preset axis in the third ultrasonic image, and the pixels on this preset axis are directly mapped based on the pixels on the long-axis lumen boundary, so the accuracy of the cross-sectional lumen boundary in the fourth ultrasonic image obtained by curve fitting based on this preset pixel is higher.

[0104] In some examples, it is also possible to determine whether to reset the preset angle interval and / or the preset angle range based on the number of executions of step S211. If the number of executions of step S211 is greater than the number threshold, the cross-sectional lumen boundary in the fourth ultrasound image can be determined based on the cross-sectional lumen boundary in the third ultrasound image and the preset angle corresponding to the maximum image similarity. The specific values of the above similarity threshold, number threshold, and preset conditions can be determined by the developer according to actual needs, and are not limited in the embodiments of the present invention.

[0105] In yet another example, the preset conditions for resetting the preset angle interval and / or the preset angle range based on preset conditions may include using a larger preset angle range and / or using a smaller preset angle interval, etc.

[0106] According to the above solution of the embodiments of the present invention, multiple preset angles can be determined in the preset angle range based on the preset angle interval, and for each preset angle, the image similarity between the third ultrasound image and the fourth ultrasound image rotated by the preset angle can be determined as the image similarity corresponding to the preset angle. Then, when the maximum image similarity is greater than the similarity threshold, the cross-sectional lumen boundary in the fourth ultrasound image is determined based on the cross-sectional lumen boundary in the third ultrasound image and the preset angle corresponding to the maximum image similarity, and when the maximum image similarity is less than or equal to the similarity threshold, the preset angle interval and / or the preset angle range are reset based on preset conditions. The above solution can determine the cross-sectional lumen boundary in the fourth ultrasound image by determining the preset angle corresponding to the maximum image similarity. Combining with the actual scenario, the blood vessel or the ultrasound catheter may rotate at different positions. By determining the above preset angle, the rotation of the blood vessel and the ultrasound catheter can be effectively shown, thereby improving the accuracy of the cross-sectional lumen boundary.

[0107] Exemplarily, the above processing method may further include: determining a first region of interest in the third ultrasound image and a second region of interest in the fourth ultrasound image.

[0108] The above first region of interest may at least include the region within the cross-sectional lumen boundary in the third ultrasound image, and the above second region of interest may at least include the region within the cross-sectional lumen boundary in the fourth ultrasound image. In one example, the first region of interest and the second region of interest can be obtained through the region segmentation algorithm or region segmentation model in related technologies, which will not be elaborated in the embodiments of the present invention. In yet another example, the first region of interest in the third ultrasound image can also be determined first, and then the first region of interest is mapped to the fourth ultrasound image through the mapping parameter threshold to obtain the second region of interest. The specific value of the above mapping parameter threshold can be set by the user according to the actual situation, and is not limited in the embodiments of the present invention.

[0109] In yet another example, the maximum distance between each pixel located on the cross-sectional lumen boundary in the third ultrasound image and the center of the third ultrasound image can be determined, and the maximum distance is multiplied by 2 and then added with a bias distance threshold to obtain the side length of a rectangle, and the center of the rectangle can be the center of the third ultrasound image. The area of the rectangle can be used as the first region of interest mentioned above. Then, through the mapping parameter threshold in the above text, the second region of interest can be obtained. By using the bias distance threshold, it is possible to make the first region of interest include the cross-sectional lumen boundary as much as possible, which is beneficial to improving the accuracy of the cross-sectional lumen boundary in the fourth ultrasound image. The specific value of the above bias distance threshold can be set by the user according to the actual situation, and the embodiments of the present invention do not limit this here.

[0110] Further, determining the image similarity between the third ultrasound image and the fourth ultrasound image after rotating the preset angle as the image similarity corresponding to the preset angle may include: determining the image similarity corresponding to the preset angle based on the pixels in the first region of interest and the pixels in the second region of interest after rotating the preset angle.

[0111] The above image similarity can be determined through the similarity generation algorithm and similarity generation model in related technologies, and the embodiments of the present invention will not elaborate on this here.

[0112] According to the above solution of the embodiments of the present invention, the image similarity can be determined by using the first region of interest and the second region of interest. On the one hand, the amount of calculation can be reduced, which is beneficial to improving the generation efficiency of the cross-sectional lumen boundary; on the other hand, the influence of non-region of interest on the similarity calculation can be reduced, and thus the representativeness of the image similarity can be improved, which is beneficial to improving the accuracy of the cross-sectional lumen boundary.

[0113] The embodiments of the present invention also provide a processing device for intravascular ultrasound images. Figure 7 FIG. shows a schematic block diagram of a processing device for intravascular ultrasound images according to an embodiment of the present invention. Combining Figure 7 , the processing device 300 may include: an image acquisition module 310, a target ultrasound image screening module 320, a long-axis image generation module 330, a long-axis lumen boundary determination module 340, and a cross-sectional lumen boundary determination module 350.

[0114] The image acquisition module 310 is used to acquire an intravascular ultrasound image sequence of a target object. Among them, the intravascular ultrasound image sequence includes a plurality of intravascular ultrasound images arranged in the acquisition order. The target ultrasound image screening module 320 is used to screen the target ultrasound images in the intravascular ultrasound image sequence based on at least the cardiac motion information corresponding to each intravascular ultrasound image, so as to obtain a target ultrasound image sequence. Among them, the cardiac motion information corresponding to each target ultrasound image is within a preset cardiac motion range, and the target ultrasound images in the target ultrasound image sequence are arranged in the acquisition order. The long-axis image generation module is used to convert the target ultrasound image sequence into a vascular long-axis image corresponding to each of at least two preset axes. Among them, the pixels located on the preset axis in the target ultrasound image correspond to a column of pixels in the vascular long-axis image. The long-axis lumen boundary determination module 340 is used to determine the long-axis lumen boundary of the blood vessel in each vascular long-axis image. The cross-sectional lumen boundary determination module 350 is used to determine and mark the cross-sectional lumen boundary of the blood vessel in each target ultrasound image based on the long-axis lumen boundary in each vascular long-axis image.

[0115] Exemplarily, the processing device 300 further includes a non-selected ultrasound image boundary determination module.

[0116] The non-selected ultrasound image boundary determination module is used to determine the cross-sectional lumen boundary of the blood vessel in the non-selected ultrasound images in the intravascular ultrasound image sequence based on the cross-sectional lumen boundary in each target ultrasound image. Among them, the non-selected ultrasound images are other images in the intravascular ultrasound image sequence that are different from the target ultrasound images.

[0117] Exemplarily, the non-selected ultrasound image boundary determination module may include a boundary determination sub-module.

[0118] The boundary determination sub-module is used to determine the cross-sectional lumen boundary of the non-selected ultrasound image based on the cross-sectional lumen boundary in the previous first reference image of the non-selected ultrasound image for the first preset number of non-selected ultrasound images after the first ultrasound image in the intravascular ultrasound image sequence; and for the second preset number of non-selected ultrasound images before the second ultrasound image in the intravascular ultrasound image sequence, determine the cross-sectional lumen boundary of the non-selected ultrasound image based on the cross-sectional lumen boundary in the next second reference image of the non-selected ultrasound image. Among them, the first ultrasound image and the second ultrasound image are two adjacent target ultrasound images in the target ultrasound image sequence, the sum of the first preset number and the second preset number is equal to the total number of non-selected ultrasound images between the first ultrasound image and the second ultrasound image in the intravascular ultrasound image sequence, and the first reference image and the second reference image are both intravascular ultrasound images in the intravascular ultrasound image sequence for which the cross-sectional lumen boundary has been determined.

[0119] Exemplarily, the unselected ultrasound image boundary determination module includes a fourth ultrasound image boundary determination module.

[0120] The fourth ultrasound image boundary determination module is configured to perform the following steps for the adjacent third ultrasound image and fourth ultrasound image in the intravascular ultrasound image sequence until the cross-sectional lumen boundary of the blood vessel in the fourth ultrasound image is determined, where the third ultrasound image is an intravascular ultrasound image for which the cross-sectional lumen boundary has been determined, and the fourth ultrasound image is an unselected ultrasound image for which the cross-sectional lumen boundary is to be determined: Based on a preset angular interval, determine a plurality of preset angles in a preset angular range, and for each preset angle, determine the image similarity between the third ultrasound image and the fourth ultrasound image rotated by the preset angle as the image similarity corresponding to the preset angle; in the case where the maximum image similarity is greater than the similarity threshold, determine the cross-sectional lumen boundary in the fourth ultrasound image based on the cross-sectional lumen boundary in the third ultrasound image and the preset angle corresponding to the maximum image similarity, and in the case where the maximum image similarity is less than or equal to the similarity threshold, reset the preset angular interval and / or the preset angular range based on a preset condition.

[0121] Exemplarily, the processing device 300 further includes a region of interest determination module.

[0122] The region of interest determination module is configured to determine a first region of interest in the third ultrasound image and a second region of interest in the fourth ultrasound image;

[0123] The fourth ultrasound image boundary determination module includes an image similarity determination module.

[0124] The image similarity determination module is configured to determine the image similarity corresponding to the preset angle based on the pixels in the first region of interest and the pixels in the second region of interest rotated by the preset angle.

[0125] Exemplarily, the target ultrasound image screening module 320 includes: a first image sequence generation module, a target ultrasound image sequence generation module.

[0126] The first image sequence generation module is configured to detect and screen out the intravascular ultrasound images including the contrast catheter in the intravascular ultrasound image sequence based on the intravascular ultrasound images in the intravascular ultrasound image sequence to obtain a first image sequence. The target ultrasound image sequence generation module is configured to screen the target ultrasound images in the first image sequence based on the respective cardiac information corresponding to each intravascular ultrasound image in the first image sequence to obtain a target ultrasound image sequence.

[0127] Exemplarily, the long-axis lumen boundary determination module 340 includes a boundary output module.

[0128] The boundary output module is configured to input each vascular long-axis image into a trained boundary detection model to obtain the long-axis lumen boundary in the vascular long-axis image, where the trained boundary detection model is trained based on vascular long-axis training images and the corresponding lumen training boundaries of the vascular long-axis training images.

[0129] Exemplarily, the boundary output module is further configured to, for each vascular long-axis image, if the vascular long-axis image does not meet the preset standard size, perform scaling processing on the vascular long-axis image based on the preset standard size to obtain a standardized vascular long-axis image, input the standardized vascular long-axis image into the trained boundary detection model, and perform the inverse processing of the scaling processing on the lumen boundary output by the trained boundary detection model to obtain the long-axis lumen boundary in the vascular long-axis image.

[0130] According to another aspect of the present invention, an ultrasound system is further provided. Figure 8 A schematic block diagram of an ultrasound system 400 according to an embodiment of the present invention is shown. As Figure 8 shown, the ultrasound system 400 includes a processor 410 and a memory 420. A computer program is stored in the memory 420, and when the computer program instructions are run by the processor 410, they are used to execute the above-mentioned method for processing intravascular ultrasound images.

[0131] In addition, according to yet another aspect of the present invention, a storage medium is further provided. Program instructions are stored on the storage medium, and when the program instructions are run by a computer or a processor, the computer or the processor is caused to execute the corresponding steps of the above-mentioned method for processing intravascular ultrasound images according to the embodiments of the present invention, and to implement the corresponding modules in the above-mentioned processing device for intravascular ultrasound images according to the embodiments of the present invention or the corresponding modules in the above-mentioned ultrasound system. The storage medium may include, for example, a memory card of a smart phone, a storage component of a tablet computer, a hard disk of a personal computer, a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a portable compact disc read-only memory (CD-ROM), a USB memory, or any combination of the above storage media. The computer-readable storage medium may be any combination of one or more computer-readable storage media. According to still another aspect of the present invention, a computer program product is further provided, including computer program instructions, and when the computer program instructions are run by a computer or a processor, the computer or the processor is caused to execute the corresponding steps of the above-mentioned method for processing intravascular ultrasound images.

[0132] Those of ordinary skill in the art can understand the specific implementation solutions of the above-mentioned ultrasound system and storage medium by reading the relevant descriptions of the above-mentioned method for processing intravascular ultrasound images. For the sake of brevity, they are not described herein again.

[0133] Although example embodiments have been described herein with reference to the accompanying drawings, it should be understood that the above example embodiments are merely exemplary and are not intended to limit the scope of the present invention thereto. Those of ordinary skill in the art can make various changes and modifications therein without departing from the scope and spirit of the present invention. All such changes and modifications are intended to be included within the scope of the present invention as claimed in the appended claims.

[0134] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0135] In several embodiments provided by the present invention, 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 example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed.

[0136] In the specification provided herein, a large number of specific details are described. However, it can be understood that the embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures, and technologies are not shown in detail so as not to obscure the understanding of this specification.

[0137] Similarly, it should be understood that, in order to streamline the present invention and help understand one or more of the various aspects of the invention, in the description of the exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, the method of the present invention should not be construed as reflecting the intention that the claimed present invention requires more features than those expressly recited in each claim. Rather, as reflected in the corresponding claims, the inventive point lies in that the corresponding technical problems can be solved with features less than all the features of a single disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into the detailed description, where each claim itself serves as a separate embodiment of the present invention.

[0138] Those skilled in the art will appreciate that, except where features are mutually exclusive, any combination can be employed of all the features disclosed in this specification (including the accompanying claims, abstract and drawings), and of all the processes or units of any method or device so disclosed. Each feature disclosed in this specification (including the accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.

[0139] In addition, those skilled in the art will understand that, although some embodiments described herein include some features included in other embodiments but not others, combinations of features of different embodiments are meant to be within the scope of the present invention and form different embodiments. For example, in the claims, any one of the claimed embodiments can be used in any combination.

[0140] Each component embodiment of the present invention can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art should understand that a microprocessor or a digital signal processor (DSP) can be used in practice to implement some or all of the functions of some of the modules in the intravascular ultrasound image processing device according to the embodiments of the present invention. The present invention can also be implemented as a device program (e.g., a computer program and a computer program product) for performing part or all of the methods described herein. Such a program implementing the present invention 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, or provided on a carrier signal, or in any other form.

[0141] It should be noted that the above embodiments illustrate rather than limit the present invention, and those skilled in the art can design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in the claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention can be implemented by means of hardware comprising several distinct elements and by means of a suitably programmed computer. In the unit claims listing several devices, several of these devices can be embodied by the same item of hardware. The use of the words first, second, and third, etc. does not denote any order. These words can be interpreted as names.

[0142] As described above, it is only the specific implementation manner of the present invention or the description of the specific implementation manner. The protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should be covered within the protection scope of the present invention. The protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for processing intravascular ultrasound images, characterized in that: The method comprises: Acquire an intravascular ultrasound image sequence of a target object, wherein the intravascular ultrasound image sequence includes a plurality of intravascular ultrasound images arranged in an acquisition order; At least based on the cardiac information corresponding to each intravascular ultrasound image, the target ultrasound images in the intravascular ultrasound image sequence are screened to obtain a target ultrasound image sequence, wherein the cardiac information corresponding to each target ultrasound image is within a preset cardiac range, and the target ultrasound images in the target ultrasound image sequence are arranged according to the acquisition order; For each preset axis of at least two preset axes, converting the target ultrasound image sequence into a blood vessel long axis image corresponding to the preset axis, wherein pixels in the target ultrasound image located on the preset axis correspond to a column of pixels in the blood vessel long axis image; For each blood vessel long-axis image, determining a long-axis lumen boundary of the blood vessel in the blood vessel long-axis image; For each target ultrasound image, based on the long-axis lumen boundary in each of the blood vessel long-axis images, a cross-sectional lumen boundary of the blood vessel in the target ultrasound image is determined and marked.

2. The method according to claim 1, characterized in that The method further comprises: Based on the cross-sectional lumen boundary in each target ultrasound image, the cross-sectional lumen boundary of the blood vessel in the non-selected ultrasound image in the intravascular ultrasound image sequence is determined, wherein the non-selected ultrasound image is other image in the intravascular ultrasound image sequence that is different from the target ultrasound image.

3. The method according to claim 2, characterized in that The step of determining the cross-sectional lumen boundary of the blood vessel in the non-selected ultrasound image in the intravascular ultrasound image sequence based on the cross-sectional lumen boundary in each target ultrasound image comprises: For a first preset number of non-selected ultrasound images after a first ultrasound image in the intravascular ultrasound image sequence, the cross-sectional lumen boundary of the non-selected ultrasound image is determined based on the cross-sectional lumen boundary in a first reference image before the non-selected ultrasound image; and for a second preset number of non-selected ultrasound images before a second ultrasound image in the intravascular ultrasound image sequence, the cross-sectional lumen boundary of the non-selected ultrasound image is determined based on the cross-sectional lumen boundary in a second reference image after the non-selected ultrasound image, The first ultrasonic image and the second ultrasonic image are two adjacent target ultrasonic images in the target ultrasonic image sequence, the sum of the first preset number and the second preset number is equal to the total number of non-selected ultrasonic images between the first ultrasonic image and the second ultrasonic image in the intravascular ultrasonic image sequence, and the first reference image and the second reference image are both intravascular ultrasonic images in which the cross-sectional lumen boundary has been determined in the intravascular ultrasonic image sequence.

4. The method according to claim 2, characterized in that The step of determining the cross-sectional lumen boundary of the blood vessel in the non-selected ultrasound image in the intravascular ultrasound image sequence based on the cross-sectional lumen boundary in each target ultrasound image comprises: For the adjacent third ultrasound image and fourth ultrasound image in the intravascular ultrasound image sequence, the following steps are performed until the cross-sectional lumen boundary of the blood vessel in the fourth ultrasound image is determined, wherein the third ultrasound image is an intravascular ultrasound image whose cross-sectional lumen boundary has been determined, and the fourth ultrasound image is a non-selected ultrasound image whose cross-sectional lumen boundary is to be determined: Based on the preset angle interval, a plurality of preset angles are determined in the preset angle interval, and for each preset angle, an image similarity between the third ultrasonic image and the fourth ultrasonic image rotated by the preset angle is determined as the image similarity corresponding to the preset angle; When the maximum image similarity is greater than a similarity threshold, the cross-sectional lumen boundary in the fourth ultrasound image is determined based on the cross-sectional lumen boundary in the third ultrasound image and a preset angle corresponding to the maximum image similarity, and when the maximum image similarity is less than or equal to the similarity threshold, the preset angle interval and / or the preset angle range is reset based on preset conditions.

5. The method according to claim 4, characterized in that The method further comprises: determining a first region of interest in the third ultrasound image and a second region of interest in the fourth ultrasound image; The determining of the image similarity between the third ultrasonic image and the fourth ultrasonic image rotated by the preset angle as the image similarity corresponding to the preset angle includes: Based on the pixels in the first focus area and the pixels in the second focus area rotated by the preset angle, the image similarity corresponding to the preset angle is determined.

6. The method according to claim 1, characterized in that The step of screening the target ultrasound image in the intravascular ultrasound image sequence based at least on the cardiac information corresponding to each intravascular ultrasound image to obtain the target ultrasound image sequence includes: Based on the intravascular ultrasound images in the intravascular ultrasound image sequence, detecting and screening out the intravascular ultrasound images including the angiography catheter in the intravascular ultrasound image sequence to obtain a first image sequence; Based on the cardiac information corresponding to each intravascular ultrasound image in the first image sequence, the target ultrasound image in the first image sequence is screened to obtain a target ultrasound image sequence.

7. The method according to claim 1, characterized in that The step of determining the long-axis lumen boundary of the blood vessel in each blood vessel long-axis image comprises: For each vascular long-axis image, the vascular long-axis image is input into a trained boundary detection model to obtain the long-axis lumen boundary in the vascular long-axis image, wherein the trained boundary detection model is trained based on the vascular long-axis training image and the lumen training boundary corresponding to the vascular long-axis training image.

8. The method according to claim 7, characterized in that For each vascular long-axis image, the vascular long-axis image is input into a trained boundary detection model to obtain a long-axis lumen boundary in the vascular long-axis image, including: For each vascular long-axis image, if the vascular long-axis image does not conform to a preset standard size, the vascular long-axis image is scaled based on the preset standard size to obtain a standardized vascular long-axis image, and the standardized vascular long-axis image is input into the trained boundary detection model, and the inverse processing of the scaling process is performed on the lumen boundary output by the trained boundary detection model to obtain the long-axis lumen boundary in the vascular long-axis image.

9. A device for processing intravascular ultrasound images, characterized in that: The device comprises: An image acquisition module, used for acquiring an intravascular ultrasound image sequence of a target object, wherein the intravascular ultrasound image sequence includes a plurality of intravascular ultrasound images arranged in an acquisition order; a target ultrasound image screening module, configured to screen target ultrasound images in the intravascular ultrasound image sequence based at least on the cardiac information corresponding to each intravascular ultrasound image, so as to obtain a target ultrasound image sequence, wherein the cardiac information corresponding to each target ultrasound image is within a preset cardiac range, and the target ultrasound images in the target ultrasound image sequence are arranged according to the acquisition order; a long-axis image generation module, configured to convert the target ultrasound image sequence into a blood vessel long-axis image corresponding to each preset axis of at least two preset axes, wherein pixels in the target ultrasound image located on the preset axis correspond to a column of pixels in the blood vessel long-axis image; A long-axis lumen boundary determination module, used for determining the long-axis lumen boundary of the blood vessel in each blood vessel long-axis image; The cross-sectional lumen boundary determination module is used to determine and mark the cross-sectional lumen boundary of the blood vessel in each target ultrasound image based on the long-axis lumen boundary in each blood vessel long-axis image.

10. An ultrasound system, characterized in that: It comprises a memory and a processor, wherein: the memory is used to store a computer program; the processor is used to execute the computer program to implement the method for processing intravascular ultrasound images as described in any one of claims 1-8.

11. A storage medium storing a computer program / instruction, characterized in that: The computer program / instructions are used to execute the method for processing intravascular ultrasound images as described in any one of claims 1 to 8 when running.

12. A computer program product comprising computer program instructions, characterized in that When the computer program instructions are executed by a processor, they are used to execute the method for processing intravascular ultrasound images according to any one of claims 1 to 8.

Images