Coronary angiography image recognition method, device and system based on segment division

Abstract

The present invention provides a method, device and system for coronary angiography image recognition based on segmentation, firstly obtaining a coronary angiography image; then segmenting the coronary angiography image to obtain a segmentation result of the coronary angiography image; then identifying the coronary angiography image to determine the main vessel obstruction information; then inputting the main vessel obstruction information and the segmentation result into a pre-established semantic segmentation model to obtain a segmentation result. By first identifying whether the main vessel is blocked, the divided segments are adaptively changed according to the blockage of the main vessel, thereby avoiding identifying the segments on other blood vessels as the segments of the main vessel when the main vessel is blocked, thereby ensuring the effectiveness of the recognition.

Description

Technical Field

[0001] The present invention belongs to the technical field of image processing, and in particular relates to a method, device and system for coronary angiography image recognition based on segmentation. Background Art

[0002] Segment identification of coronary angiography images is an important technique in cardiovascular imaging that allows physicians to accurately identify and analyze different segments of the coronary arteries. During a coronary angiography, a special dye or contrast agent is typically used to clearly visualize the blood vessels of the heart. These images are then used to assess the presence, extent, and location of coronary artery disease.

[0003] Segment recognition technology is one of the key technologies in angiographic image processing. Identifying these segments is crucial for diagnosing coronary artery stenosis, plaques or other abnormalities. Doctors can refer to the recognition results of coronary angiography images to more accurately plan treatment procedures by identifying and marking segments. For example, as a reference for percutaneous coronary intervention or coronary artery bypass surgery. In the prior art, angiographic images are usually processed by semantic segmentation, and its accuracy is affected by data quality and depends to a large extent on the quality and representativeness of training data. Different people have different coronary conditions, so the data used by conventional machine learning is difficult to represent all coronary conditions, resulting in inaccurate segment recognition in the image. Summary of the invention

[0004] In view of this, the present invention provides a method, device and system for coronary angiography image recognition based on segment division, aiming to solve the problem of inaccurate segment recognition in images in the prior art.

[0005] A first aspect of an embodiment of the present invention provides a coronary angiography image recognition method based on segmentation, comprising:

[0006] Obtain coronary angiographic images;

[0007] Segmenting the coronary angiography image to obtain a segmentation result of the coronary angiography image;

[0008] Identify coronary angiography images and determine the main blood vessel blockage information;

[0009] Input the main vessel obstruction information and segmentation results into the pre-established semantic segmentation model to obtain the segmentation results;

[0010] According to the segment division results, the lesions in the coronary angiography images are determined.

[0011] In a possible implementation, the main vessel obstruction information and the segmentation result are input into a pre-established semantic segmentation model to obtain a segmentation result, including:

[0012] The main vessel obstruction information and the segmentation results are input into a pre-established semantic segmentation model, and the segmented blood vessels in the segmentation results are divided into first-category segments, second-category segments, and third-category segments according to their importance;

[0013] Among them, the importance of the first category segments is greater than that of the second category segments; the importance of the second category segments is greater than that of the third category segments.

[0014] In a possible implementation, the main vessel obstruction information and the segmentation result are input into a pre-established semantic segmentation model, and the segmented blood vessels in the segmentation result are divided into first-category segments, second-category segments, and third-category segments according to their importance, including:

[0015] When the main vessel obstruction information is unobstructed, the first type of segments are divided from the segmentation results according to the semantic segmentation model;

[0016] According to the semantic segmentation model, the second and third category segments connected to the first category segments are divided.

[0017] In a possible implementation, the main vessel obstruction information and the segmentation result are input into a pre-established semantic segmentation model, and the segmented blood vessels in the segmentation result are divided into first-category segments, second-category segments, and third-category segments according to their importance, including:

[0018] When the main vessel occlusion information is occlusion, the first type of segments except the main vessel are divided from the segmentation results according to the semantic segmentation model;

[0019] According to the semantic segmentation model, the second and third category segments connected to the first category segments except the main blood vessels are divided.

[0020] In a possible implementation, the lesion type and classification in the coronary angiography image are determined according to the segment division result, including:

[0021] Stenosis identification is performed on coronary angiography images to obtain multiple stenosis areas;

[0022] A comprehensive score is performed based on the locations of multiple stenosis areas and the segments in which the stenosis areas are located to determine the lesion classification in the coronary angiography image.

[0023] In a possible implementation, a comprehensive score is performed based on the locations of multiple stenosis areas and the segments in which the stenosis areas are located to determine the lesion type in the coronary angiography image, including:

[0024] Determine a bifurcation node according to the first-category segment and the second-category segment and the third-category segment connected to the first-category segment;

[0025] The scores of the three segments connected by the bifurcation node are multiplied by the characteristic value of the narrow area on each segment to obtain the total score of the bifurcation node;

[0026] According to the total score of each bifurcation node, it is determined whether each bifurcation node has a bifurcation lesion.

[0027] In a possible implementation, the lesion in the coronary angiography image is determined based on the scores of multiple stenosis areas and each segment, including:

[0028] Determine a trifurcated node according to the first-category segment and the second-category segment and the third-category segment connected to the first-category segment;

[0029] The scores of the four segments connected by the trifurcated node are multiplied by the characteristic value of the narrow area on each segment to obtain the total score of the trifurcated node;

[0030] According to the total score of each trifurcated node, it is determined whether each trifurcated node has a trifurcated lesion.

[0031] A second aspect of an embodiment of the present invention provides a coronary angiography image recognition device based on segment division, comprising:

[0032] An acquisition module, used for acquiring coronary angiography images;

[0033] A segmentation module is used to segment the coronary angiography image to obtain a segmentation result of the coronary angiography image;

[0034] A determination module, used to identify the coronary angiography image and determine the main blood vessel obstruction information;

[0035] A segmentation module, used for inputting the main vessel obstruction information and the segmentation result into a pre-established semantic segmentation model to obtain a segmentation result;

[0036] The recognition module is used to determine the lesions in the coronary angiography image according to the segment division result.

[0037] A third aspect of an embodiment of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the segment-based coronary angiography image recognition method of the first aspect are implemented.

[0038] A fourth aspect of an embodiment of the present invention provides an angiography image processing system, comprising a vascular angiography device and the electronic device according to the third aspect above.

[0039] The fifth aspect of an embodiment of the present invention provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the steps of the segment-based coronary angiography image recognition method of the first aspect as described above.

[0040] The segment-based coronary angiography image recognition method, device and system provided by the embodiment of the present invention first obtain a coronary angiography image; then segment the coronary angiography image to obtain a segmentation result of the coronary angiography image; then recognize the coronary angiography image to determine the main vessel blockage information; then input the main vessel blockage information and the segmentation result into a pre-established semantic segmentation model to obtain a segmentation result. By first identifying whether the main vessel is blocked, the divided segments are adaptively changed according to the blockage of the main vessel, thereby avoiding identifying the segments on other blood vessels as the segments of the main vessel when the main vessel is blocked, thereby ensuring the effectiveness of recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0042] Figure 1 This is an application scenario diagram of coronary angiography image recognition based on segment division provided by an embodiment of the present invention;

[0043] Figure 2 is a flow chart of an implementation of a coronary angiography image recognition method based on segment division provided by an embodiment of the present invention;

[0044] Figure 3 is a schematic structural diagram of a coronary angiography image recognition device based on segment division provided by an embodiment of the present invention;

[0045] Figure 4 It is a schematic diagram of the structure of a terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0047] Figure 1: is an application scenario diagram of coronary angiography image recognition based on segment division provided by an embodiment of the present invention. Figure 1 As shown, in some embodiments, the segment-based coronary angiography image recognition provided by the embodiment of the present invention can be applied to, but not limited to, this application scenario. The system may include: an angiography device 11 and a processing terminal 12.

[0048] First, the catheter needs to be inserted through the patient's femoral artery or other peripheral artery and sent to the ascending aorta. Then, the left or right coronary artery orifice is found and inserted. Then, contrast agent is injected into the coronary artery. At this time, the angiography device 11 captures the developed angiography image and sends it to the electronic device 12. The electronic device 12 segments and processes the angiography image to obtain a clear cardiovascular segmentation map.

[0049] Among them, the electronic device 12 can be a terminal or a server. The terminal can be an inspection terminal equipped on a medical X-ray inspection device, a doctor's office terminal, etc. The server can be a management server of a hospital information management system of a hospital, or a cloud server, which is not limited here.

[0050] Usually, when a patient has symptoms of coronary discomfort, the patient needs to undergo static and dynamic coronary examinations, wherein the static examination includes CT, MR, etc., and the dynamic examination is specifically DSA (Digital subtraction angiography) coronary angiography technology. After the examination, the corresponding lesion score is performed to provide data reference for the doctor's diagnosis and treatment. The present invention improves the lesion score after dynamic examination.

[0051] Figure 2 FIG. 1 is a flow chart of the implementation of the coronary angiography image recognition method based on segment division provided by an embodiment of the present invention. Figure 2 As shown, the coronary angiography image recognition method based on segment division includes:

[0052] S210, obtaining a coronary angiography image.

[0053] In an embodiment of the present invention, the initial angiographic image may be Figure 1 The data may be taken in real time by the angiography device in the hospital, or may be obtained from the hospital's information management system, which is not limited here.

[0054] S220, segmenting the coronary angiography image to obtain a segmentation result of the coronary angiography image.

[0055] In the embodiment of the present invention, due to reasons such as X-ray resolution, 2D grayscale artifacts, dynamic changes in contrast agent filling level, etc., there is usually a large amount of noise in the contrast image. Therefore, it is necessary to preprocess the contrast image and segment the blood vessels, and then identify the segments and lesions based on the segmented coronary vascular images.

[0056] S230, identifying the coronary angiography image to determine the main blood vessel obstruction information.

[0057] In the embodiment of the present invention, the segmented blood vessels may specifically include the left main trunk, the anterior descending branch, the circumflex branch, the right coronary artery, etc., which are not limited here. The main blood vessel is usually the clearest blood vessel branch in the image, so a conventional recognition model can be used to identify whether the main blood vessel is blocked. The recognition model can be a neural network, a deep learning model, an adversarial network, etc., which are not limited here.

[0058] S240, inputting the main vessel obstruction information and the segmentation result into a pre-established semantic segmentation model to obtain a segmentation result.

[0059] In the embodiment of the present invention, the semantic segmentation model can be a supervised learning model, a semi-supervised learning model, etc., which is not limited here. The semantic segmentation model has two decoders, corresponding to the main blood vessel obstruction information as blocked and unblocked.

[0060] The coronary artery tree has multiple bifurcations and bends on the surface of the heart. These complex structures make it difficult to track blood vessels and identify segments. And due to individual differences, the anatomical structure of each person's coronary artery is different, which requires that in the process of image recognition, especially segment recognition, the model can recognize such individual differences. In addition, the patient's breathing, movement, and the above-mentioned image interference and other reasons further increase the complexity of segment recognition, making it difficult to effectively locate the lesion. The present invention provides a coarse-grained scoring method, which aims to simplify the segment recognition process and improve the efficiency of segment recognition.

[0061] In some embodiments, the main blood vessel obstruction information and the segmentation result are input into a pre-established semantic segmentation model to obtain a segmentation result, including: inputting the main blood vessel obstruction information and the segmentation result into a pre-established semantic segmentation model, and dividing the blood vessels segmented in the segmentation result into first-category segments, second-category segments and third-category segments according to their importance; wherein the importance of the first-category segments is greater than that of the second-category segments; and the importance of the second-category segments is greater than that of the third-category segments.

[0062] In an embodiment of the present invention, particularly important segments in the coronary artery can be defined as first-category segments, such as the left main trunk, the anterior descending branch of the main vessel, the circumflex branch of the main vessel, etc. A portion of the more important segments are defined as second-category segments, such as the right coronary main vessel, the vascular branches on the anterior descending branch close to the left main trunk, and the vascular branches on the circumflex branch close to the left main trunk (the vascular branches that are a preset number of closest to the left main trunk or the vascular branches within the design length of the left main trunk can be defined as close to the left main trunk). Generally important segments are defined as third-category segments, such as the right coronary non-main vessel, the vascular branches on the anterior descending branch far away from the left main trunk, and the vascular branches on the circumflex branch far away from the left main trunk. Other small blood vessels that have no effect on the patient's physical condition are of low importance and can be ignored during segment identification.

[0063] The present invention divides all blood vessels into three categories for segment identification based on the importance of the blood vessels. There is no need to accurately identify the type of each segment, thus greatly reducing the complexity of segment identification.

[0064] In some embodiments, the main blood vessel obstruction information and the segmentation results are input into a pre-established semantic segmentation model, and the blood vessels segmented in the segmentation results are divided into first-category segments, second-category segments, and third-category segments according to their importance, including: when the main blood vessel obstruction information is not blocked, dividing the first-category segments from the segmentation results according to the semantic segmentation model; dividing the second-category segments and the third-category segments connected to the first-category segments according to the semantic segmentation model.

[0065] In some embodiments, the main vessel obstruction information and the segmentation results are input into a pre-established semantic segmentation model, and the blood vessels segmented in the segmentation results are divided into first-category segments, second-category segments, and third-category segments according to their importance, including: when the main vessel obstruction information is obstruction, dividing the first-category segments other than the main vessels from the segmentation results according to the semantic segmentation model; dividing the second-category segments and the third-category segments connected to the first-category segments other than the main vessels according to the semantic segmentation model.

[0066] In the embodiment of the present invention, when the main vessel is not blocked, one of the decoders in the semantic segmentation model works to determine the type of each segment according to a conventional recognition scheme. When the main vessel is blocked, another decoder in the semantic segmentation model works to no longer search for the main vessel and the segments of the branches on the main vessel in the image, and only recognizes and classifies other segments. This can avoid misidentifying other segments as main vessel segments, and ensure recognition accuracy.

[0067] After completing the segment division, it is necessary to further judge the type, degree, and classification of coronary heart disease lesions. In some embodiments, the lesions in the coronary angiography image can be determined based on the segment division results. Specifically, this can be achieved through the following steps: stenosis identification is performed on the coronary angiography image to obtain multiple stenosis areas; comprehensive scoring is performed based on the locations of the multiple stenosis areas and the segments in which the stenosis areas are located to determine the lesion classification in the coronary angiography image.

[0068] In the embodiment of the present invention, conventional recognition algorithms can be used for stenosis recognition, which is not limited here. The higher the importance of a segment, the greater the impact of its stenosis, and the higher the lesion score; the more stenosis areas in a segment, the higher the lesion score.

[0069] Specifically, a special scoring table can be formulated for the first category segments, and different first category segments have different scores, which can be looked up in the table when used. For example, the score of the segment on the left main trunk is 5, the score of the segment on the anterior descending branch is 1.5, and the score of the segment on the circumflex branch is 3.5 or 2.5. The score of the second category segment is uniformly defined as 1, and the score of the third category segment is uniformly defined as 0.5. The above score is only an example of the present invention and is not intended to be limiting. In actual use, the doctor can adaptively modify the score value at any time.

[0070] The present invention provides the following two embodiments to illustrate the lesion scoring process, but is not intended to be limiting.

[0071] In some embodiments, a comprehensive score is performed based on multiple stenosis areas and the locations of each segment in which the stenosis areas are located to determine the type of lesions in the coronary angiography image, including: determining a bifurcation node based on a first category segment and a second category segment and a third category segment connected to the first category segment; multiplying the scores of the three segments connected to the bifurcation node by the characteristic values ​​of the stenosis areas on each segment to obtain a total score of the bifurcation node; and determining whether each bifurcation node has a bifurcation lesion based on the total score of each bifurcation node.

[0072] In the embodiment of the present invention, there are usually 7 bifurcation nodes in the coronary artery image, and the scores of the 7 bifurcation nodes are different from each other. (Except the right coronary artery) All are associated with a special important segment, and the special important segments are classified separately, and the right coronary artery is also processed separately, so it is conditionally inferred whether the bifurcation is scored. For example, the intersection of the left main trunk, the anterior descending branch, and the circumflex branch is a bifurcation node. Any two adjacent segments on the anterior descending branch, and the second-category segment or the third-category segment connected to one of the two segments, is a bifurcation node.

[0073] In the embodiment of the present invention, the characteristic value of the stenosis area depends on the detection accuracy of the stenosis detection algorithm. If only the presence of stenosis can be detected, the characteristic value depends on the number of stenosis areas on the segment. For example, the characteristic value of one stenosis area can be defined as 1. If the stenosis detection algorithm can detect the degree of stenosis of each stenosis area in addition to the presence of stenosis, the characteristic value depends on the degree of each stenosis area on the segment. For example, the characteristic value is defined as the maximum value of the stenosis degree on the segment, or as the sum of the stenosis degrees of each stenosis area.

[0074] In the embodiment of the present invention, the characteristic value of the segment is multiplied by the score to obtain the score of the segment. The above calculation is performed on the three segments connected to the bifurcation node, and then the sum is calculated to obtain the total score. If the total score is greater than the threshold set by the doctor, it indicates that there is a lesion in the bifurcation node.

[0075] In some embodiments, lesions in a coronary angiography image are determined based on scores of multiple stenosis areas and each segment, including: determining a trifurcated node based on a first category segment and a second category segment and a third category segment connected to the first category segment; multiplying the scores of the four segments connected to the trifurcated node by the characteristic values ​​of the stenosis area on each segment to obtain a total score of the trifurcated node; and determining whether each trifurcated node has a trifurcated lesion based on the total score of each trifurcated node.

[0076] In the embodiment of the present invention, there are usually 5 trifurcated nodes in the coronary artery image, and the scores of the 5 trifurcated nodes are different from each other, and (except for the right coronary artery) they are all associated with a special important segment. For example, the intersection of the four segments of the left main trunk, the anterior descending branch, the circumflex branch and the small branch on the left main trunk is a trifurcated node. Any two adjacent segments on the anterior descending branch, and the two segments connected by one of the two segments (these two segments can be second-class segments or third-class segments) are a trifurcated node.

[0077] In the embodiment of the present invention, the characteristic value of the stenosis area depends on the detection accuracy of the stenosis detection algorithm. If only the presence of stenosis can be detected, the characteristic value depends on the number of stenosis areas on the segment. For example, the characteristic value of one stenosis area can be defined as 1. If the stenosis detection algorithm can detect the degree of stenosis of each stenosis area in addition to the presence of stenosis, the characteristic value depends on the degree of each stenosis area on the segment. For example, the characteristic value is defined as the maximum value of the stenosis degree on the segment, or as the sum of the stenosis degrees of each stenosis area.

[0078] In the embodiment of the present invention, the characteristic value of the segment is multiplied by the score to obtain the score of the segment. The four segments connected by the trifurcated node are all subjected to the above calculation, and then the sum is taken to obtain the total score. If the total score is greater than the threshold set by the doctor, it indicates that there is a lesion in the trifurcated node.

[0079] In summary, the beneficial effects of the present invention are specifically as follows:

[0080] 1. By first identifying whether the main blood vessel is blocked, the divided segments are adaptively changed according to the blockage of the main blood vessel, thereby avoiding identifying the segments on other blood vessels as the segments of the main blood vessel when the main blood vessel is blocked, thereby ensuring the effectiveness of the identification.

[0081] 2. A coarse-grained segment identification and scoring method is provided, aiming to simplify the segment identification process, simplify the scoring system for lesion identification, and improve the efficiency of segment identification.

[0082] 3. It greatly reduces the processing process of angiographic image segment recognition, has stronger real-time characteristics than existing technologies, and can quickly provide doctors with relevant information to assist them in real-time diagnosis and decision-making.

[0083] 4. A scoring method is provided to further determine the type, degree and classification of coronary heart disease lesions with the help of the identified segments.

[0084] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0085] Figure 3 FIG. 1 is a schematic diagram of the structure of a coronary angiography image recognition device based on segment division provided by an embodiment of the present invention. Figure 3 As shown, in some embodiments, the segment-based coronary angiography image recognition device 3 includes:

[0086] An acquisition module 310 is used to acquire a coronary angiography image;

[0087] The segmentation module 320 is used to segment the coronary angiography image to obtain a segmentation result of the coronary angiography image;

[0088] A determination module 330 is used to identify the coronary angiography image and determine the main blood vessel obstruction information;

[0089] The segmentation module 340 is used to input the main vessel obstruction information and the segmentation result into a pre-established semantic segmentation model to obtain a segmentation result.

[0090] Optionally, the segmentation module 340 is used to: input the main vessel obstruction information and the segmentation results into a pre-established semantic segmentation model, and divide the blood vessels segmented in the segmentation results into first-category segments, second-category segments and third-category segments according to their importance; wherein the importance of the first-category segments is greater than that of the second-category segments; and the importance of the second-category segments is greater than that of the third-category segments.

[0091] Optionally, the segmentation module 340 is used to: when the main blood vessel obstruction information is unobstructed, divide the first category of segments from the segmentation results according to the semantic segmentation model; and divide the second category of segments and the third category of segments connected to the first category of segments according to the semantic segmentation model.

[0092] Optionally, the segmentation module 340 is used to: when the main blood vessel obstruction information is obstruction, divide the first category segments other than the main blood vessels from the segmentation results according to the semantic segmentation model; and divide the second category segments and the third category segments connected to the first category segments other than the main blood vessels according to the semantic segmentation model.

[0093] Optionally, the device also includes an identification module, which is used to: identify stenosis in the coronary angiography image to obtain multiple stenosis areas; perform comprehensive scoring based on the multiple stenosis areas and the locations of each segment in which the stenosis areas are located to determine the degree or type of lesions in the coronary angiography image.

[0094] Optionally, an identification module is used to: determine a bifurcation node based on a first-category segment and a second-category segment and a third-category segment connected to the first-category segment; multiply the scores of the three segments connected to the bifurcation node by the characteristic values ​​of the stenosis area on each segment to obtain a total score of the bifurcation node; and determine whether each bifurcation node has a bifurcation lesion based on the total score of each bifurcation node.

[0095] Optionally, an identification module is used to: determine a trifurcated node based on a first-category segment and a second-category segment and a third-category segment connected to the first-category segment; multiply the scores of the four segments connected to the trifurcated node by the characteristic values ​​of the stenosis area on each segment to obtain a total score of the trifurcated node; and determine whether each trifurcated node has a trifurcated lesion based on the total score of each trifurcated node.

[0096] The segment division-based coronary angiography image recognition device provided in this embodiment can be used to execute the above method embodiments. Its implementation principle and technical effects are similar and will not be described in detail in this embodiment.

[0097] Figure 4 Schematic diagram of the structure of the terminal provided by the embodiment of the present invention. Figure 4As shown, an embodiment of the present invention provides a terminal 4, which includes: a processor 40, a memory 41, and a computer program 42 stored in the memory 41 and executable on the processor 40. When the processor 40 executes the computer program 42, the steps in the above-mentioned coronary angiography image recognition method based on segmentation are implemented, for example Figure 2 Alternatively, when the processor 40 executes the computer program 42, the functions of each module / unit in the above-mentioned system embodiments are realized, for example Figure 3 The functions of each module are shown.

[0098] Exemplarily, the computer program 42 may be divided into one or more modules / units, one or more modules / units are stored in the memory 41, and are executed by the processor 40 to implement the present invention. One or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program 42 in the terminal 4.

[0099] The terminal 4 may be a terminal or a server. The terminal may be a mobile phone, MCU, ECU, industrial computer, etc., which are not limited here. The server may be a physical server, a cloud server, etc., which are not limited here. The terminal 4 may include, but is not limited to, a processor 40 and a memory 41. Those skilled in the art will understand that Figure 4 It is only an example of terminal 4 and does not constitute a limitation on terminal 4. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal may also include input and output devices, network access devices, buses, etc.

[0100] The processor 40 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0101] The memory 41 may be an internal storage unit of the terminal 4, such as a hard disk or memory of the terminal 4. The memory 41 may also be an external storage device of the terminal 4, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal 4. Further, the memory 41 may also include both an internal storage unit and an external storage device of the terminal 4. The memory 41 is used to store computer programs and other programs and data required by the terminal. The memory 41 may also be used to temporarily store data that has been output or is to be output.

[0102] An embodiment of the present invention provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps in the above-mentioned embodiment of the coronary angiography image recognition method based on segmentation are implemented.

[0103] The computer readable storage medium stores a computer program 42, which includes program instructions. When the program instructions are executed by the processor 40, all or part of the processes in the above-mentioned embodiment method are implemented. The computer program 42 can also be used to instruct related hardware to complete. The computer program 42 can be stored in a computer readable storage medium. When the computer program 42 is executed by the processor 40, the steps of each of the above-mentioned method embodiments can be implemented. Among them, the computer program 42 includes computer program code, which can be in source code form, object code form, executable file or some intermediate form. The computer readable medium can include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium.

[0104] The computer-readable storage medium may be an internal storage unit of the terminal of any of the foregoing embodiments, such as a hard disk or memory of the terminal. The computer-readable storage medium may also be an external storage device of the terminal, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the terminal. Further, the computer-readable storage medium may also include both an internal storage unit of the terminal and an external storage device. The computer-readable storage medium is used to store computer programs and other programs and data required by the terminal. The computer-readable storage medium may also be used to temporarily store data that has been output or is to be output.

[0105] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0106] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0107] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0108] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0109] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0110] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0111] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0112] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. Computer-readable media can include: any entity or device that can carry computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0113] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A coronary angiography image recognition method based on segmentation, It is characterized in that include: Obtain coronary angiographic images; Segmenting the coronary angiography image to obtain a segmentation result of the coronary angiography image; Identifying the coronary angiography image to determine main blood vessel obstruction information; Inputting the main blood vessel obstruction information and the segmentation result into a pre-established semantic segmentation model, and dividing the blood vessels segmented in the segmentation result into first-category segments, second-category segments, and third-category segments according to their importance; The importance of the first type of segments is greater than that of the second type of segments; the importance of the second type of segments is greater than that of the third type of segments; The main vessel obstruction information and the segmentation result are input into a pre-established semantic segmentation model, and the segmented blood vessels in the segmentation result are divided into first-category segments, second-category segments, and third-category segments according to their importance, including: When the main blood vessel obstruction information indicates that the main blood vessel is not obstructed, dividing the first type of segments from the segmentation result according to a semantic segmentation model; According to the semantic segmentation model, the second and third category segments connected to the first category segments are divided; When the main blood vessel obstruction information is obstruction, dividing the first type of segments except the main blood vessel from the segmentation result according to the semantic segmentation model; According to the semantic segmentation model, the second and third category segments connected to the first category segments except the main blood vessels are divided.

2. The coronary angiography image recognition method based on segmentation according to claim 1, It is characterized in that The method further comprises: Performing stenosis identification on the coronary angiography image to obtain multiple stenosis areas; A comprehensive score is performed based on the multiple stenosis areas and the positions of the various segments in which the stenosis areas are located to determine the lesion classification in the coronary angiography image.

3. The coronary angiography image recognition method based on segmentation according to claim 2, It is characterized in that A comprehensive score is performed according to the multiple stenosis areas and the locations of the various segments where the stenosis areas are located to determine the lesion type in the coronary angiography image, including: Determine a bifurcation node according to a first-category segment and a second-category segment and a third-category segment connected to the first-category segment; The scores of the three segments connected by the bifurcation node are multiplied by the characteristic values ​​of the narrow areas on each segment connected by the bifurcation node, and then the sum is calculated to obtain the total score of the bifurcation node. According to the total score of each bifurcation node, it is determined whether each bifurcation node has a bifurcation lesion.

4. The coronary angiography image recognition method based on segmentation according to claim 2, It is characterized in that A comprehensive score is performed according to the multiple stenosis areas and the locations of the various segments where the stenosis areas are located to determine the lesion type in the coronary angiography image, including: Determine a trifurcated node according to a first-category segment and a second-category segment and a third-category segment connected to the first-category segment; The scores of the four segments connected by the trifurcation node are multiplied by the characteristic values ​​of the narrow areas on each segment connected by the trifurcation node, and then the sum is calculated to obtain the total score of the trifurcation node. According to the total score of each trifurcated node, it is determined whether each trifurcated node has a trifurcated lesion.

5. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the steps of the coronary angiography image recognition method based on segmentation as described in any one of claims 1 to 4 are implemented.

6. A radiographic image processing system, It is characterized in that The device comprises an angiography device and the electronic device as claimed in claim 5 above.

7. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the coronary angiography image recognition method based on segmentation as described in any one of claims 1 to 4 above are implemented.

Images