Methods, devices, and storage media for training a detection model to detect vascular adhesions

Abstract

The application discloses a method and device for training a detection model for detecting vascular adhesion and a storage medium. The method comprises: collecting multi-modal image data, wherein the multi-modal image data comprises digital subtraction angiography data, CT angiography data and magnetic resonance angiography data; performing vascular segmentation based on the multi-modal image data to obtain a vascular graph and vascular information; extracting a binary graph of a single vessel according to the vascular graph and the vascular information, and labeling whether the single vessel passes through an adhesion vessel; constructing training data based on the multi-modal image data, the binary graph of the single vessel and the labeling result; inputting the training data into a detection model for vascular adhesion detection to obtain a detection result, so as to train the detection model for vascular adhesion detection. The scheme disclosed by the application can improve the accurate positioning and efficient detection of vascular adhesion.

Description

Technical Field

[0001] This application generally relates to the field of vascular adhesion detection technology. More specifically, this application relates to a method for training a detection model for detecting vascular adhesion, a method for detecting vascular adhesion, an apparatus, and a computer-readable storage medium. Background Technology

[0002] Digital subtraction angiography (DSA) is the gold standard for diagnosing vascular diseases, and its images show intracranial vessels in a tree-like structure. Current techniques use vessel segmentation and skeletonization algorithms to extract centerlines and construct vascular maps. However, due to the volume effect, vascular adhesions easily form ring structures, leading to reduced efficiency, poor stability, and even program crashes in path tracing algorithms. Traditional methods rely on manual annotation or simple geometric analysis, which cannot efficiently and accurately detect vascular adhesions.

[0003] Therefore, there is an urgent need to provide a scheme for training a detection model for vascular adhesions in order to improve the accurate localization and efficient detection of vascular adhesions. Summary of the Invention

[0004] In order to at least address one or more of the technical problems mentioned above, this application proposes a scheme for training a detection model for detecting vascular adhesions in several aspects.

[0005] In a first aspect, this application provides a method for training a detection model for detecting vascular adhesions, comprising: acquiring multimodal image data, wherein the multimodal image data includes digital subtraction angiography data, CT angiography data, and magnetic resonance angiography data; performing vascular segmentation based on the multimodal image data to obtain vascular maps and vascular information; extracting binary maps of single vessels based on the vascular maps and vascular information, and marking whether they pass through adherent vessels; constructing training data based on the multimodal image data, the binary maps of the single vessels, and the marking results; and inputting the training data into a detection model to perform vascular adhesion detection to obtain detection results, thereby training the detection model for vascular adhesion detection.

[0006] In some embodiments, the vascular information includes at least the position information and radius information of each pixel on the vascular map.

[0007] In other embodiments, the binary image of the single blood vessel is extracted by the following operations: determining a target pixel and a radius at the target pixel on the single blood vessel based on the blood vessel image and the blood vessel information; setting the pixels within the radius at the target pixel as the center of the target pixel as the target value, and setting the remaining pixels as 0, to extract the binary image of the single blood vessel.

[0008] In some other embodiments, the method further includes: determining whether a pixel passes through an adhesive blood vessel based on the blood vessel map and the blood vessel information; and marking whether the pixel passes through an adhesive blood vessel or not based on the determination result to obtain the marking result.

[0009] In some other embodiments, the method further includes: performing maximum bounding box cropping on the single blood vessel to obtain the cropped single blood vessel; and constructing the training data based on the multimodal image data, the binary map of the cropped single blood vessel, and the annotation results.

[0010] In some other embodiments, the detection model includes an encoder and a classifier, and the training data is input into the detection model to perform vascular adhesion detection to obtain detection results, thereby training the vascular adhesion detection model: based on the training data, the encoder is used to perform an encoding operation to obtain an encoding result; the classifier is used to perform a classification operation on the encoding result to obtain the vascular adhesion detection result, thereby training the vascular adhesion detection model.

[0011] In some other embodiments, the detection model includes a machine learning model or a deep learning model.

[0012] In a second aspect, this application provides a method for detecting vascular adhesion, comprising: acquiring multimodal image data, wherein the multimodal image data includes digital subtraction angiography data, CT angiography data, and magnetic resonance angiography data; performing vascular segmentation based on the multimodal image data to obtain vascular maps and vascular information; extracting a binary map of a single vessel based on the vascular maps and vascular information; and inputting the multimodal image data and the binary map of the single vessel into a detection model trained by one or more embodiments of the first aspect to perform vascular adhesion detection, thereby obtaining a detection result.

[0013] In a third aspect, this application provides an apparatus comprising: a processor; and a memory having computer instructions stored thereon for training a detection model for detecting vascular adhesions, wherein when the computer instructions are executed by the processor, the plurality of embodiments of the first aspect are implemented; or having computer instructions for detecting vascular adhesions stored thereon, wherein when the computer instructions are executed by the processor, the embodiment of the second aspect is implemented.

[0014] In a fourth aspect, this application provides a computer-readable storage medium having stored thereon computer program instructions for training a detection model for detecting vascular adhesions, wherein when executed by one or more processors, the computer program instructions cause to implement the plurality of embodiments of the first aspect described above; or having stored computer instructions for detecting vascular adhesions, wherein when executed by a processor, the computer instructions cause to implement one embodiment of the second aspect described above.

[0015] Using the training scheme for the detection model of vascular adhesions provided above, this embodiment of the application provides multi-dimensional feature inputs to the model by fusing different imaging methods with multimodal data, thereby enhancing its adaptability to complex vascular scenes. By extracting binary images of single blood vessels from vascular maps and vascular information, the vascular boundary blurring caused by gray-level overlap or noise interference in the original image is compensated for, providing clear geometric clues for the model. The original multimodal data and binary images are simultaneously input into the detection model for training, enabling the model to process both the gray-level information of the original image and the spatial structure information of the binary image. This enhances the spatial localization accuracy of adhesion locations, and the combined effect of both gives the model stronger cross-modal generalization performance, avoiding overfitting caused by single-modal training. Attached Figure Description

[0016] The above and other objects, features, and advantages of exemplary embodiments of this application will become readily understood by reading the following detailed description with reference to the accompanying drawings. In the drawings, several embodiments of this application are illustrated by way of example and not limitation, and the same or corresponding reference numerals denote the same or corresponding parts, wherein:

[0017] Figure 1 This is an exemplary flowchart illustrating a method for training a detection model for detecting vascular adhesions according to an embodiment of this application;

[0018] Figure 2 This is an exemplary flowchart illustrating the overall training of a detection model for detecting vascular adhesions according to an embodiment of this application;

[0019] Figure 3 This is an exemplary schematic diagram illustrating a portion of a blood vessel segment according to an embodiment of this application;

[0020] Figure 4 This is an exemplary flowchart illustrating a method for detecting vascular adhesions according to an embodiment of this application;

[0021] Figure 5 This is an exemplary structural block diagram illustrating a device according to an embodiment of this application. Detailed Implementation

[0022] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] It should be understood that the terms "comprising" and "including" used in the specification and claims of this application indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0024] It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application. As used in this specification and claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this specification and claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations.

[0025] As used in this specification and claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if [described condition or event] is detected" may be interpreted, depending on the context, as "once determined," "in response to determination," "once [described condition or event] is detected," or "in response to detection of [described condition or event]."

[0026] The specific embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0027] Figure 1 This is an exemplary flowchart illustrating a method 100 for training a detection model for detecting vascular adhesions according to an embodiment of this application. Figure 1 As shown, the method 100 includes: step S101: acquiring multimodal image data, wherein the multimodal image data includes digital subtraction angiography data, CT angiography data, and magnetic resonance angiography data; step S102: performing vessel segmentation based on the multimodal image data to obtain vessel maps and vessel information; step S103: extracting binary images of single vessels based on vessel maps and vessel information, and marking whether they pass through adherent vessels; step S104: constructing training data based on multimodal image data, binary images of single vessels, and marking results; and step S105: inputting the training data into a detection model to perform vessel adhesion detection to obtain detection results, thereby training the detection model for vessel adhesion detection.

[0028] First, in step S101, multimodal image data is acquired. This multimodal image data may include digital subtraction angiography (DSA) data, CT angiography data, and magnetic resonance angiography (MRI) data. DSA data is the gold standard for screening and diagnosing various vascular diseases, providing detailed information about the vascular lumen. CT angiography and MRI data can also be used for vascular disease screening and follow-up, respectively supplementing the spatial relationship between blood vessels and surrounding tissues and displaying hemodynamic characteristics. This embodiment of the application uses three modalities of data to train the detection model, covering vascular structural information based on different imaging principles and improving the model's adaptability to complex scenarios.

[0029] Next, in step S102, blood vessel segmentation is performed based on multimodal image data to obtain a blood vessel map and blood vessel information. In some embodiments, image processing methods or deep learning methods can be used for blood vessel segmentation. The aforementioned image processing methods may include, but are not limited to, image edge detection and image morphology methods. The aforementioned deep learning methods may be based on existing pre-trained segmentation models for blood vessel segmentation, and this application does not impose any limitations on this. Blood vessel information can be obtained from the segmented blood vessels using, for example, the SKNW library. In some embodiments, blood vessel information may include, but is not limited to, the position information and radius information of each pixel on the blood vessel map.

[0030] Based on the obtained vascular map and vascular information, in step S103, a binary image of a single blood vessel is extracted according to the vascular map and vascular information, and whether it passes through adherent blood vessels is marked. In some embodiments, the binary image of a single blood vessel can be extracted by the following operation: determining the target pixel and its radius on the single blood vessel based on the vascular map and the vascular information; setting the pixels within the radius region centered on the target pixel to the target value (e.g., 1), and setting the remaining pixels to 0, thus extracting the binary image of the single blood vessel. This ensures the smoothness and integrity of the blood vessel boundary and effectively suppresses noise interference. The overlay operation enhances the continuity of the main blood vessel, eliminates isolated noise points, and provides clear structural feature input for the model.

[0031] As an example, suppose the target pixel on a single blood vessel is Xi (i = 1 to n). Centered on Xi, the pixels within a radius ri of the target pixel Xi are set to 1, and the remaining pixels are set to 0, in order to extract the binary image X_binary of the single blood vessel.

[0032] In some embodiments, it can be determined whether a target node passes through adherent blood vessels based on the vascular map and vascular information. Based on the determination result, the target node is labeled as having passed through or not passing through adherent blood vessels to obtain the labeling result. That is, two types of labels are formed: one type indicating that the target node has passed through adherent blood vessels, whose value can be set to 1; and the other type indicating that the target node has not passed through adherent blood vessels, whose value can be set to 0. In some implementation scenarios, a threshold can be set based on, for example, the difference in vascular radius or the grayscale value of the vascular map. When the difference in radius between adjacent vascular segments exceeds the threshold, it is determined to be adhesion.

[0033] Further, in step S104, training data is constructed based on the multimodal image data, the binary map of a single vessel, and the annotation results. As an example, suppose the multimodal image data is denoted as X, the binary map of a single vessel is denoted as X_binary, and the annotation result is "passing through adherent vessels," with a value of 1. In this scenario, the training data [[X,X_binary],1] can be obtained. Similarly, if the annotation result is "not passing through adherent vessels," with a value of 0, then the training data [[X,X_binary],0] can be obtained.

[0034] In some implementation scenarios, maximum bounding box cropping can be performed on individual blood vessels to obtain cropped single-vessel images. Training data can then be constructed based on multimodal image data, the binary image of the cropped single-vessel image, and the annotation results. This method of maximum bounding box cropping focuses on high-incidence areas of adhesion (such as branch junctions) and reduces interference from redundant background information. The cropped image patch size is standardized to reduce computational complexity while preserving complete vascular structural features, ensuring the model can quickly locate adhesion positions during inference and improving detection speed.

[0035] It can be understood that the maximum bounding box is the smallest rectangular region containing the target object (a single blood vessel). The boundary is determined by calculating the extreme coordinates (x_min, x_max, y_min, y_max) of all pixels in the blood vessel segment in the vascular image, ensuring complete coverage of the blood vessel segment while minimizing redundant background. As an example, let's denote the binary image of the cropped single blood vessel as X_binary_box, the multimodal image data as X, and the annotation result as passing through adherent blood vessels, with a value of 1. In this scenario, we can obtain training data [[X, X_binary_box], 1]. Alternatively, the annotation result can be "passing through adherent blood vessels," with a value of 0, resulting in training data [[X, X_binary_box], 0].

[0036] Finally, in step S105, the training data is input into the detection model to perform vascular adhesion detection to obtain detection results, thereby training the vascular adhesion detection model. In some embodiments, the detection model may include a machine learning model or a deep learning model. In other embodiments, the detection model may include an encoder and a classifier. Specifically, based on the training data, an encoder is used to perform an encoding operation to obtain an encoding result, and then a classifier is used to perform a classification operation on the encoding result to obtain the vascular adhesion detection result, thereby training the vascular adhesion detection model.

[0037] In an exemplary scenario, if the encoder is denoted as `encoder` and the classifier as `clf`, the encoding result `feature = encoder(X, X_binary)` is first obtained through encoding operations using the encoder. For a binary image of a cropped single blood vessel, `feature = encoder(X, X_binary_box)`. Next, the encoding result is classified using the classifier, `pred = clf(feature)`, and a cross-entropy loss function, such as `Loss = CrossEntroy(label, pred)`, is calculated for backpropagation to update the parameters of the encoder `encoder` and the classifier `clf`, thereby training the detection model for blood vessel adhesion detection. Here, `label` represents the label value.

[0038] As described above, the multimodal data in this application provides multi-dimensional feature inputs to the model by fusing different imaging methods, enhancing adaptability to complex vascular scenes and reducing the risk of misjudgment caused by imaging artifacts in a single modality. It provides a structured input basis for adhesion detection through vessel segmentation and vessel map extraction, ensuring accurate representation of vessel morphology and spatial location. By extracting binary images of single vessels and superimposing spherical regions to generate discretized structural features (such as connected regions and diameter variations), it compensates for the blurred vessel boundaries caused by grayscale overlap or noise interference in the original images, providing clear geometric clues for the model. Furthermore, by cropping single vessels using maximum bounding boxes, it avoids focusing on areas with high adhesion rates, reduces redundant background information interference, standardizes input dimensions, and improves detection efficiency.

[0039] Furthermore, this embodiment of the application also achieves joint feature extraction through a cascaded architecture of encoder and classifier, combining multimodal training data to cover different imaging conditions and a unified representation of vascular structures in binary images, enabling the model to have stronger cross-modal generalization performance and avoiding overfitting caused by single-modal training. Finally, end-to-end training optimizes feature transfer efficiency, achieving accurate identification of vascular adhesion locations, solving the problem of misjudgment of adhesions caused by volume effects in traditional methods, and improving the reliability of vascular segmentation results and the stability of subsequent clinical applications.

[0040] Figure 2 This is an exemplary flowchart illustrating the overall training of a detection model for detecting vascular adhesions according to an embodiment of this application. Figure 2 As shown, in step S201, multimodal image data is acquired, including digital subtraction angiography data, CT angiography data, and magnetic resonance angiography data. Next, in step S202, vessel segmentation is performed based on the multimodal image data to obtain vessel maps and information. Specifically, vessel segmentation can be performed using image processing methods or deep learning methods, and vessel information (e.g., the position and radius of each pixel) can be obtained using, for example, the SKNW library. In step S203, a maximum bounding box is performed on a single vessel to obtain the cropped single vessel. In step S204, training data is constructed based on the multimodal image data, the binary map of the cropped single vessel, and the annotation results. Further, the training data is input into the detection model, and in step S205, an encoder is used to perform encoding operations to obtain the encoding results. In step S206, a classifier is used to classify the encoding results, and in step S207, a loss function (e.g., cross-entropy loss) is calculated to train the detection model for vascular adhesion detection, so that in step S208, the trained detection model is ready.

[0041] Figure 3 This is an exemplary schematic diagram illustrating a portion of a blood vessel segment according to an embodiment of this application. For example... Figure 3 The image shown is a partial vascular segment image X. When training a detection model for detecting vascular adhesions, binary images of corresponding single vascular segments (e.g., vascular segment 1-3-6; vascular segment 5-2-1) can be extracted based on this partial vascular segment image X. Specific binarization operations can be found in the description above. Next, training data is constructed based on multimodal image data, the binary images of single vessels, and the annotation results.

[0042] Taking vessel segment 1-3-6 as an example, if its corresponding binary image is denoted as X_binary, since vessel segment 1-3-6 is an adherent vessel, its annotation value is 1, thus obtaining the training data [[X,X_binary],1]. Similarly, for vessel segment 5-2-1, which does not pass through adherent vessels, the value is 0, and the training data [X,X_binary'],0] can be obtained. As mentioned above, a maximum bounding box can be performed on a single vessel to obtain the cropped single vessel to construct the training data [[X,X_binary_box],1]. This data is then input into the detection model, encoded using an encoder, and then classified using a classifier to obtain the detection result of vessel adhesion detection, thereby training the vessel adhesion detection model.

[0043] Figure 4This is an exemplary flowchart illustrating a method 400 for detecting vascular adhesions according to an embodiment of this application. Figure 4 As shown, the method 400 includes: Step S401: acquiring multimodal image data, wherein the multimodal image data includes digital subtraction angiography data, CT angiography data, and magnetic resonance angiography data; Step S402: performing vessel segmentation based on the multimodal image data to obtain vessel maps and vessel information; Step S403: extracting binary images of single vessels based on the vessel maps and vessel information; Step S404: inputting the multimodal image data and the binary images of single vessels into a trained detection model to perform vessel adhesion detection, thereby obtaining detection results. Further details regarding the extraction of binary images of single vessels and the training of the detection model can be found above. Figure 1 , Figure 2 The description of the process is not repeated here. According to the detection results, if the classification operation pred is 1, it indicates that there is adhesion at the intermediate endpoint; other values ​​indicate that there is no adhesion.

[0044] In some embodiments, in detecting vascular adhesions, the maximum bounding box of a single vessel can be cropped. The binary image of the cropped single vessel and the original multimodal data are then input into a trained detection model to perform vascular adhesion detection and obtain the detection results. Similarly, based on the detection results, if the classification operation `pred` is 1, it indicates that adhesion exists at the intermediate endpoint; other values ​​indicate that adhesion does not exist.

[0045] Figure 5 This is an exemplary structural block diagram illustrating a device 500 according to an embodiment of this application. It is understood that the device 500 may include the apparatus of the embodiments of this application, and the device implementing the solution of this application may be a single device (e.g., a computing device) or a multifunctional device including various peripheral devices.

[0046] like Figure 5 As shown, the device of this application may further include a central processing unit (“CPU”) 511, which may be a general-purpose CPU, a special-purpose CPU, or other information processing and program execution unit. Further, the device 500 may also include a mass storage device 512 and a read-only memory (“ROM”) 513, wherein the mass storage device 512 may be configured to store various types of data, including various multimodal data, vascular maps and vascular information, binary maps of single vessels, algorithm data, intermediate results, and various programs required to run the device 500. The ROM 513 may be configured to store power-on self-tests for the device 500, initialization of various functional modules in the system, drivers for the system's basic input / output, and data and instructions required to boot the operating system.

[0047] Optionally, device 500 may also include other hardware platforms or components, such as the tensor processing unit (“TPU”) 514, graphics processing unit (“GPU”) 515, field-programmable gate array (“FPGA”) 516, and machine learning unit (“MLU”) 517 shown. It is understood that although various hardware platforms or components are shown in device 500, they are merely exemplary and not limiting, and those skilled in the art can add or remove appropriate hardware as needed. For example, device 500 may include only a CPU, associated storage devices, and interface devices to implement the method of training a detection model for detecting vascular adhesions or the method for detecting vascular adhesions described in this application.

[0048] In some embodiments, to facilitate data transmission and interaction with external networks, the device 500 of this application further includes a communication interface 518, through which it can connect to a local area network / wireless local area network (“LAN / WLAN”) 505, and further through the LAN / WLAN to connect to a local server 506 or to the Internet (“Internet”) 507. Alternatively or additionally, the device 500 of this application can also directly connect to the Internet or cellular network via the communication interface 518 based on wireless communication technology, such as wireless communication technology based on 3G (“3G”), 4G (“4G”), or 5G (“5G”). In some application scenarios, the device 500 of this application can also access the server 508 and database 509 of an external network as needed to obtain various known algorithms, data, and modules, and can remotely store various data, such as various data or instructions for presenting, for example, multimodal data, vascular maps and vascular information, binary maps of single vessels, etc.

[0049] The peripheral devices of device 500 may include a display device 502, an input device 503, and a data transmission interface 504. In one embodiment, the display device 502 may include, for example, one or more speakers and / or one or more visual displays, configured to train the detection model for detecting vascular adhesions of this application or to provide voice prompts and / or image / video display for detecting vascular adhesions. The input device 503 may include, for example, a keyboard, mouse, microphone, posture capture camera, or other input buttons or controls, configured to receive audio data input and / or user commands. The data transmission interface 504 may include, for example, a serial interface, parallel interface, or Universal Serial Bus interface (“USB”), Small Computer System Interface (“SCSI”), Serial ATA, FireWire (“FireWire”), PCI Express, and High Definition Multimedia Interface (“HDMI”), configured for data transmission and interaction with other devices or systems. According to the scheme of this application, the data transmission interface 504 can receive multimodal data acquired from CT or MRI equipment and transmit multimodal data or various other types of data or results to device 500.

[0050] The CPU 511, mass storage 512, ROM 513, TPU 514, GPU 515, FPGA 516, MLU 517, and communication interface 518 of the device 500 of this application can be interconnected via bus 519, and can interact with peripheral devices through this bus. In one embodiment, the CPU 511 can control other hardware components in the device 500 and its peripheral devices through this bus 519.

[0051] The above combination Figure 5 A device that can be used to execute this application is described. It should be understood that the device structure or architecture described herein is merely exemplary, and the implementation method and implementation entity of this application are not limited thereto, but can be changed without departing from the spirit of this application.

[0052] Based on the foregoing description in conjunction with the accompanying drawings, those skilled in the art will understand that the embodiments of this application can also be implemented by software programs. Therefore, this application also provides a computer-readable storage medium storing computer-readable instructions for training a detection model for detecting vascular adhesions or for detecting vascular adhesions. When executed by one or more processors, these computer-readable instructions can be used to implement the embodiments of this application in conjunction with the accompanying drawings. Figure 1 The method described for training a detection model for detecting vascular adhesions; or the implementation of this application in conjunction with the appendix. Figure 4 The method described is for detecting vascular adhesions.

[0053] It should be noted that although the operations of the method of this application are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. On the contrary, the steps depicted in the flowchart can be performed in a different order. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0054] It should be understood that when the terms "first," "second," "third," and "fourth," etc., are used in the claims, specification, and drawings of this application, they are used only to distinguish different objects and not to describe a specific order. The terms "comprising" and "including" as used in the specification and claims of this application indicate the presence of the described features, integrals, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or collections thereof.

[0055] It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application. As used in this specification and claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this specification and claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations.

[0056] Although the embodiments of this application are described above, the content is merely an example adopted for the purpose of facilitating understanding of this application and is not intended to limit the scope and application scenarios of this application. Any person skilled in the art described in this application may make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed in this application, but the scope of patent protection of this application shall still be determined by the scope defined in the appended claims.

[0057] Furthermore, the collection and acquisition of various data in this application comply with relevant laws and regulations and are authorized by the data providers. Any organization or individual that needs to obtain external data shall obtain authorization in accordance with the law and ensure data security, and shall not illegally collect, use, process, or transmit unauthorized or unprotected data, nor shall it illegally buy, sell, provide, or disclose unauthorized or unprotected data.

Claims

1. A method for training a detection model for detecting vascular adhesions, comprising: Acquire multimodal image data, including digital subtraction angiography data, CT angiography data, and magnetic resonance angiography data, wherein the multimodal image data respectively provide details of the vascular lumen, the spatial relationship between the blood vessel and the surrounding tissue, and display hemodynamic characteristics; Based on the multimodal image data, blood vessel segmentation is performed to obtain blood vessel maps and blood vessel information; The target pixel on a single blood vessel and the radius at the target pixel are determined based on the blood vessel map and the blood vessel information. The pixels within the radius of the target pixel are set to the target value, and the remaining pixels are set to 0, so as to extract the binary image of the single blood vessel and mark whether it passes through the adhesion blood vessel. The single blood vessel is cut using the maximum bounding box to obtain the cut single blood vessel. Training data is constructed based on the multimodal image data, the binary image of the cropped single blood vessel, and the annotation results; The training data is input into the detection model to detect vascular adhesions and obtain detection results, thereby training the detection model for vascular adhesion detection. This also includes: Based on the vascular diagram and the vascular information, determine whether a pixel passes through an adherent blood vessel; The annotation result is obtained by marking pixels that pass through or do not pass through adherent blood vessels based on the judgment result, wherein the blood vessel information includes at least the position information and radius information of each pixel on the blood vessel map.

2. The method according to claim 1, wherein the detection model includes an encoder and a classifier, and the training data is input into the detection model to perform vascular adhesion detection to obtain detection results, thereby training the vascular adhesion detection model: Based on the training data, the encoder is used to perform encoding operations to obtain the encoding result; The classifier is used to classify the encoded results to obtain the detection results of vascular adhesion detection, so as to train the detection model of vascular adhesion detection.

3. The method according to claim 2, wherein the detection model includes a machine learning model or a deep learning model.

4. A method for detecting vascular adhesions, comprising: Acquire multimodal image data, wherein the multimodal image data includes digital subtraction angiography data, CT angiography data, and magnetic resonance angiography data; Based on the multimodal image data, blood vessel segmentation is performed to obtain blood vessel maps and blood vessel information; Based on the vascular map and the vascular information, a binary image of a single blood vessel is extracted; and The multimodal image data and the binary image of the single blood vessel are input into the detection model trained according to any one of claims 1-3 to perform vascular adhesion detection and obtain the detection results.

5. An apparatus comprising: processor; as well as A memory storing computer instructions for training a detection model for detecting vascular adhesions, which, when executed by a processor, cause the method according to any one of claims 1-3 to be implemented; or storing computer instructions for detecting vascular adhesions, which, when executed by a processor, cause the method according to claim 4 to be implemented.

6. A computer-readable storage medium storing computer program instructions for training a detection model for detecting vascular adhesions, wherein the computer program instructions, when executed by one or more processors, cause the method according to any one of claims 1-3 to be implemented; or storing computer program instructions for detecting vascular adhesions, wherein the computer program instructions, when executed by one or more processors, cause the method according to claim 4 to be implemented.

Images