Method, computer device, readable storage medium and program product for obtaining location of intracranial aneurysm in blood vessel segment

Abstract

The present application relates to a method, computer device, readable storage medium and program product for obtaining the segmented position of an intracranial aneurysm in a blood vessel, comprising: obtaining the three-dimensional position coordinates of the aneurysm; obtaining a segmented intracranial arterial tree; combining the three-dimensional position coordinates of the aneurysm and the segmented intracranial arterial tree to obtain a three-dimensional intracranial vascular model including the intracranial arterial tree and aneurysm position information; establishing a standard intracranial vascular model and performing segmented annotation; spatially registering the three-dimensional intracranial vascular model with the standard intracranial vascular model to obtain a registered three-dimensional intracranial vascular model; on the registered three-dimensional intracranial vascular model, selecting pixel points on the aneurysm whose distance from the intracranial arterial tree is within a preset threshold range, obtaining mapping points corresponding to the pixel points on the standard intracranial vascular model, and obtaining the segmented position of the aneurysm on the intracranial arterial tree based on the positions of the mapping points.

Description

Technical Field

[0001] The present application relates to the intersection of medical field and image processing field, and in particular to a method for obtaining the location of an intracranial aneurysm in a blood vessel segment, a computer device, a readable storage medium and a program product. Background Art

[0002] Intracranial aneurysms are abnormal bulges in the walls of intracranial arteries, with a prevalence of approximately 3.2%. The current gold standard for diagnosis is direct cerebral angiography (DSA). However, because DSA is invasive and expensive, CT angiography (CTA) or magnetic resonance imaging (MRA) is typically used for cerebral aneurysm screening in clinical practice.

[0003] After a CTA or MRA examination, a radiologist reviews the images to determine the presence of an aneurysm, as well as its location, morphological parameters, and other parameters. However, examining all intracranial vessels to determine abnormalities consumes the physician's time. Furthermore, screening for aneurysms using raw images relies heavily on the physician's knowledge and experience. Even experienced physicians can miss small aneurysms. Summary of the Invention

[0004] Based on this, it is necessary to provide a method for obtaining the location of intracranial aneurysms in blood vessel segments in response to the above technical problems.

[0005] The method of obtaining the segmented location of an intracranial aneurysm in a blood vessel in the present application comprises:

[0006] Aneurysm detection network is used to detect and segment aneurysms in intracranial medical images and obtain the three-dimensional position coordinates of the aneurysm.

[0007] The segmentation model is used to segment the arterial tree in the intracranial medical image to obtain the segmented intracranial arterial tree;

[0008] Combining the three-dimensional position coordinates of the aneurysm and the segmented intracranial arterial tree, a three-dimensional reconstruction is performed to obtain a three-dimensional intracranial vascular model including the intracranial arterial tree and aneurysm position information;

[0009] Establish a standard model of intracranial blood vessels, perform segmented annotation, and obtain a standard model of intracranial blood vessels with segmented annotation information;

[0010] Performing spatial registration on the intracranial vascular three-dimensional model and the intracranial vascular standard model to obtain a registered intracranial vascular three-dimensional model;

[0011] On the registered three-dimensional model of intracranial blood vessels, pixel points on the aneurysm that are within a preset threshold range from the intracranial arterial tree are selected to obtain the corresponding mapping points of the pixel points on the standard intracranial blood vessel model, and the segmented position of the aneurysm on the intracranial arterial tree is obtained according to the position of the mapping points.

[0012] Optionally, the pixel points and the mapping points correspond one to one, and for any pixel point, the point on the standard model of intracranial blood vessels that is closest to it is the corresponding mapping point.

[0013] Optionally, obtaining the segmented position of the aneurysm on the intracranial arterial tree according to the position of the mapping point specifically includes:

[0014] The segmented labeling information of the position corresponding to the maximum number of mapping points is obtained, and the segmented labeling information is translated into a specific segmented position to obtain the segmented position of the aneurysm on the intracranial arterial vascular tree.

[0015] Optionally, when obtaining the segmented annotation information corresponding to the position with the maximum number of mapping points, if the relative difference between the maximum number of mapping points appearing in two segments is less than a first threshold, it is determined that the specific segmented position of the aneurysm is at the intersection of the two segments.

[0016] Optionally, the intracranial vascular standard model is derived from an intracranial vascular standard model library, which includes at least two intracranial vascular standard models with different morphologies.

[0017] Performing spatial registration on the intracranial vascular three-dimensional model and the intracranial vascular standard model to obtain a registered intracranial vascular three-dimensional model specifically includes:

[0018] The intracranial vascular three-dimensional model and at least two different intracranial vascular standard models are spatially registered respectively, and a registration result with a relatively small spatial distance difference is selected to obtain a registered intracranial vascular three-dimensional model.

[0019] Optionally, the intracranial vascular standard models in the intracranial vascular standard model library are organized and classified according to gender and / or age.

[0020] Optionally, obtaining a three-dimensional intracranial vascular model including intracranial arterial tree and aneurysm location information further includes:

[0021] performing a closing operation on the three-dimensional model of the intracranial blood vessels with the aneurysm location information to obtain the three-dimensional model of the intracranial blood vessels that has completed the closing operation;

[0022] extracting the maximum connected domain of the three-dimensional intracranial blood vessel model that completes the closing operation, to obtain a three-dimensional intracranial blood vessel model without scattered points;

[0023] The Laplace smoothing algorithm is used to perform surface smoothing on the intracranial vascular three-dimensional model with scattered points removed, and a three-dimensional intracranial vascular model including the intracranial arterial tree and aneurysm location information is obtained.

[0024] Optionally, performing spatial registration on the three-dimensional intracranial blood vessel model and the standard intracranial blood vessel model further includes:

[0025] Perform spatial registration on the intracranial vascular three-dimensional model including the intracranial arterial tree and aneurysm location information and the intracranial vascular standard model to obtain a translation and rotation matrix;

[0026] According to the translation and rotation matrix, the three-dimensional model of the intracranial blood vessels is subjected to corresponding translation and rotation transformation to obtain a registered three-dimensional model of the intracranial blood vessels.

[0027] Optionally, spatial registration of the three-dimensional intracranial blood vessel model and the standard intracranial blood vessel model may include:

[0028] Obtaining all points of the intracranial artery tree, and obtaining the sum of distances from all points to the nearest point on the intracranial vascular standard model;

[0029] Obtaining a translation and rotation matrix that minimizes the sum of the distances by using singular value decomposition, and translating and rotating the three-dimensional model of the intracranial blood vessels according to the translation and rotation matrix;

[0030] The above steps are executed repeatedly until the sum of the distances reaches a second preset threshold, or the number of cycles reaches a third preset threshold.

[0031] The present application also provides a computer device, comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method for obtaining the location of an intracranial aneurysm in a blood vessel segment as described in the present application.

[0032] The present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method for obtaining the location of an intracranial aneurysm in a blood vessel segment as described in the present application are implemented.

[0033] The present application also provides a computer program product, comprising computer instructions, which, when executed by a processor, implement the steps of the method for obtaining the location of an intracranial aneurysm in a blood vessel segment as described in the present application.

[0034] The method for obtaining the segmented location of an intracranial aneurysm in a blood vessel in the present application has at least the following effects:

[0035] This application uses a registration method to spatially align the reconstructed 3D intracranial vascular model with a standard intracranial vascular model, achieving spatial matching. The specific segmented positions of the 3D reconstructed 3D intracranial vascular model are obtained using the segmentation information of the standard intracranial vascular model, allowing the location of the aneurysm vessel to be predicted without manual intervention. This application shortens the time required to diagnose aneurysms, improves diagnostic accuracy, reduces the misdiagnosis rate, and increases the convenience of collecting statistical data. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a flow chart of a method for obtaining the location of an intracranial aneurysm in a blood vessel segment in one embodiment of the present application;

[0037] Figure 2 A schematic diagram of a segmentation method of an intracranial vascular model in one embodiment of the present application;

[0038] Figure 3 Schematic diagram of a three-dimensional model of intracranial blood vessels before registration in one embodiment of the present application;

[0039] Figure 4 Schematic diagram of a three-dimensional model of intracranial blood vessels before matching and a standard model of intracranial blood vessels in one embodiment of the present application;

[0040] Figure 5 Schematic diagram of a standard model of intracranial blood vessels after registration in one embodiment of the present application;

[0041] Figure 6 This is a diagram of the internal structure of a computer device in one embodiment of the present application. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0043] like Figure 1 As shown, an embodiment of the present application provides a method for obtaining the location of an intracranial aneurysm in a blood vessel segment, comprising steps S100 to S600.

[0044] Step S100 , detecting and segmenting aneurysms in intracranial medical images using an aneurysm detection network to obtain the three-dimensional position coordinates of the aneurysm;

[0045] In this step, the intracranial artery tree is the subject's intracranial artery tree, and the aneurysm detection network can be, for example, the Transformer-based 3D image segmentation model UNETR. In step S100, if the aneurysm detection network is used alone to detect aneurysms, the result obtained is only the 3D location coordinates of the aneurysm, requiring further layer-by-layer search, and the detection result is not intuitive.

[0046] In step S100, if at least two aneurysms are detected, one of them is selected as the three-dimensional position coordinates of the aneurysm. It is understood that if there are more than one aneurysm, the position of each aneurysm is determined separately, so that the position determination result is more reliable.

[0047] Step S200 , segmenting the arterial tree in the intracranial medical image using a segmentation model to obtain a segmented intracranial arterial tree;

[0048] Specifically, a pre-trained semantic segmentation model is used to segment the intracranial artery tree with an aneurysm to obtain a segmented intracranial artery tree. The semantic segmentation model can be, for example, a 3D image segmentation model V-NET based on a volumetric fully convolutional neural network.

[0049] Step S300 , combining the three-dimensional position coordinates of the aneurysm and the segmented intracranial arterial tree to perform three-dimensional reconstruction to obtain a three-dimensional intracranial vascular model including the intracranial arterial tree and aneurysm position information;

[0050] In this step, the 3D reconstruction can be performed using an iso-rectangular surface extraction algorithm. Step S300 also includes sub-steps S310 to S330.

[0051] Step S310: Perform a closing operation on the intracranial vascular three-dimensional model with aneurysm location information to obtain a closed intracranial vascular three-dimensional model. The closing operation includes dilation and erosion processing to connect the blood vessels that may be segmented and broken.

[0052] Step S320: extract the largest connected domain of the closed intracranial vascular 3D model and remove scattered points. Specifically, retain the connected domain with more than 1000 voxels (i.e., 1*1*1 3D pixels) and remove scattered point noise.

[0053] Step S330 : Using a Laplace smoothing algorithm, the three-dimensional model of intracranial blood vessels with scattered points removed is smoothed.

[0054] Step S400: Establish a standard model of intracranial blood vessels, perform segmented annotation, and obtain a standard model of intracranial blood vessels with segmented annotation information;

[0055] Specifically, a complete and morphologically normal intracranial vascular model (i.e., an intracranial vascular standard model) is prepared for segmentation and each segment is labeled with a label. The label is the segmentation annotation information, for example, it can be a character or number sequence. The intracranial vascular standard model comes from the intracranial vascular standard model library, which includes at least two intracranial vascular standard models with different morphologies. The intracranial vascular standard models in the intracranial vascular standard model library are organized and classified according to gender and / or age. When in use, for example, the intracranial vascular standard model with a gender and / or age close to that of the patient can be preferentially retrieved for alignment. It can be understood that the intracranial arterial tree targeted in step S100 and step S200 is the same, and the intracranial vascular standard model established in step S400 is used as a reference.

[0056] Segmentation is used to assist in precise positioning in subsequent steps, see Figure 2 , the dotted lines in the figure serve as markers for the segmentation method. The segmentation method can be, for example: including the basilar artery, the anterior and posterior communicating arteries, the left and right anterior arteries (divided into 2 segments), the left and right middle arteries (divided into 2 segments), the left and right posterior arteries, the left and right vertebral arteries (divided into 4 segments), the left and right internal carotid arteries (divided into 7 segments), the left and right ophthalmic arteries, the left and right superior cerebellar arteries, and the left and right posterior cerebellar arteries. The segmentation method is only an example and can be further adjusted according to the accuracy requirements. It can be understood that aneurysms at the segmentation will not affect the results, which will be explained in detail in the subsequent steps.

[0057] Step S500, registering the intracranial vascular three-dimensional model with the intracranial vascular standard model to obtain a registered intracranial vascular three-dimensional model;

[0058] The registered intracranial vascular 3D model includes the intracranial arterial tree, aneurysm location information, and segmented annotation information. The registration method can be, for example, an iterative nearest neighbor method based on the least squares method. Of course, other registration methods can also be used.

[0059] Specifically, the 3D intracranial vascular model is spatially registered with at least two different standard intracranial vascular models. The registration result with the smallest spatial distance difference is selected to obtain a fully registered 3D intracranial vascular model. Separate registration can improve the applicability of the standard intracranial vascular model and ensure the reliability of the final result.

[0060] In step S500, the registration process specifically includes: registering the intracranial vascular three-dimensional model including the intracranial arterial tree and aneurysm location information with the intracranial vascular standard model to obtain a translation and rotation matrix;

[0061] According to the translation and rotation matrix, the corresponding translation and rotation transformation is performed on the intracranial vascular three-dimensional model to obtain the registered intracranial vascular three-dimensional model.

[0062] It can be understood that the corresponding translation and rotation transformations correspond to the translation and rotation matrices. Figure 3 , see the arrow in the figure, it can be seen that the aneurysm is on the vascular tree before registration. Figure 5 On the registered intracranial vascular 3D model, the positional relationship between the aneurysm and the standard intracranial vascular model can be seen. Figure 5 The standard model was made transparent, and it was seen that the aneurysm was completely hidden in the fourth segment of the carotid artery of the standard model.

[0063] Furthermore, in step S500, spatial registration of the intracranial vascular three-dimensional model and the intracranial vascular standard model is performed, specifically including:

[0064] Step S510, obtaining all points of the intracranial artery tree, and obtaining the sum of the distances from all points to the nearest point on the intracranial vascular standard model;

[0065] Step S520, using singular value decomposition to obtain a translation and rotation matrix that minimizes the sum of distances, and translating and rotating the three-dimensional model of the intracranial blood vessels according to the translation and rotation matrix;

[0066] Step S530 loops through steps S510 to S520 until the sum of the distances reaches a second preset threshold, or the number of loops reaches a third preset threshold. All points can be understood as all pixels forming the intracranial arterial tree. The second preset threshold can be, for example, the number of points in the intracranial vascular 3D model * 0.01 mm, and the third preset threshold can be, for example, 500-1000 times.

[0067] In step S600, on the registered three-dimensional intracranial vascular model, pixel points on the aneurysm that are within a preset threshold range from the intracranial arterial tree are selected, and the corresponding mapping points of these pixel points on the standard intracranial vascular model are obtained. The segmented position of the aneurysm on the intracranial arterial tree is obtained based on the positions of these mapping points.

[0068] It can be understood that there is a one-to-one correspondence between pixels and mapping points. For any pixel, the point on the standard intracranial vascular model closest to it is the corresponding mapping point. In this step, the segmented annotation information corresponding to the position with the maximum number of mapping points is obtained, and the segmented annotation information is translated into specific segment locations, thereby obtaining the segmented location of the aneurysm on the intracranial arterial tree.

[0069] In this step, setting an appropriate preset threshold can prevent the results from being interfered with and affected. The preset threshold can be, for example, 0.1mm to 2mm, such as 0.5mm. Pixels within the preset threshold range can be obtained on the three-dimensional model of the intracranial blood vessels after the registration is completed. The segmented annotation information corresponding to the position with the maximum number of pixels mapped in the standard model is obtained, that is, the number of pixels is first counted to determine the position with the most occurrences, and this position is identified as the specific location of the aneurysm.

[0070] Specifically, when obtaining segmented annotation information corresponding to the location with the maximum number of mapping points, if the relative difference between the maximum number of mapping points appearing in two segments is less than a first threshold, the specific segmented location of the aneurysm is determined to be at the intersection of the two segments. The relative difference is a percentage of one relative to the other, and the first threshold can be, for example, 70%.

[0071] It is understood that the segmented annotation information in step S400 corresponds to each segment, and the translation into specific locations in this step can visualize the specific location of the aneurysm. The accuracy of the specific location of the aneurysm depends on the accuracy of the segmentation method and can be set accordingly according to needs.

[0072] In steps S100 to S600, the specific segmented location of the aneurysm can be automatically determined, helping doctors quickly locate the aneurysm and possessing significant clinical value. This is particularly effective for small and medium-sized aneurysms, with diameters of less than 5 mm for small aneurysms, between 5 mm and 15 mm for medium aneurysms, and greater than 15 mm for large aneurysms.

[0073] In response to the existing practice of relying on physicians' knowledge and experience to screen for aneurysms using CTA or MRA imaging, the embodiments of this application utilize a pre-trained semantic segmentation model to segment the intracranial arterial tree with aneurysms. A registration method is used to register the 3D reconstructed intracranial vascular model with a standard intracranial vascular model, achieving spatial matching. The specific segmented locations of the 3D reconstructed intracranial vascular model are obtained using the segmentation information of the standard intracranial vascular model, allowing the location of the aneurysm to be predicted without manual intervention.

[0074] The various embodiments of this application shorten the doctor's diagnosis time, improve the diagnostic accuracy, and reduce the misdiagnosis rate. The methods provided by the various embodiments of this application can be used to complete preliminary screening of aneurysms in physical examination centers, making it easier to collect aneurysm locations from different populations, which is statistically significant. Under the premise of manual judgment, the prone locations of aneurysms in populations of different ages, regions, and genders can be counted in large quantities, which has far-reaching significance for studying the characteristics of aneurysm incidence.

[0075] It should be understood that although Figure 1The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. In addition, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0076] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 6 As shown. The computer device includes a processor, memory, a network interface, and a database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store segmented annotated data on aneurysm locations for populations of different ages, regions, and genders. The network interface of the computer device is used to communicate with an external terminal via a network connection. When executed by the processor, the computer program implements a method for obtaining the segmented location of an intracranial aneurysm in a blood vessel. The network interface of the computer device is used to communicate with an external terminal via a network connection. When executed by the processor, the computer program implements a method for obtaining the segmented location of an intracranial aneurysm in a blood vessel. The display screen of the computer device may be a liquid crystal display or an electronic ink display screen. The input device of the computer device may be a touch layer covering the display screen, or may be buttons, a trackball, or a touchpad provided on the computer device housing, or may be an external keyboard, touchpad, or mouse.

[0077] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0078] Step S100 , detecting and segmenting aneurysms in intracranial medical images using an aneurysm detection network to obtain the three-dimensional position coordinates of the aneurysm;

[0079] Step S200 , segmenting the arterial tree in the intracranial medical image using a segmentation model to obtain a segmented intracranial arterial tree;

[0080] Step S300 , combining the three-dimensional position coordinates of the aneurysm and the segmented intracranial arterial tree to perform three-dimensional reconstruction to obtain a three-dimensional intracranial vascular model including the intracranial arterial tree and aneurysm position information;

[0081] Step S400: Establish a standard model of intracranial blood vessels, perform segmented annotation, and obtain a standard model of intracranial blood vessels with segmented annotation information;

[0082] Step S500, registering the intracranial vascular three-dimensional model with the intracranial vascular standard model to obtain a registered intracranial vascular three-dimensional model;

[0083] In step S600, on the registered three-dimensional intracranial vascular model, pixel points on the aneurysm that are within a preset threshold range from the intracranial arterial tree are selected, and the corresponding mapping points of these pixel points on the standard intracranial vascular model are obtained. The segmented position of the aneurysm on the intracranial arterial tree is obtained based on the positions of these mapping points.

[0084] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0085] Step S100 , detecting and segmenting aneurysms in intracranial medical images using an aneurysm detection network to obtain the three-dimensional position coordinates of the aneurysm;

[0086] Step S200 , segmenting the arterial tree in the intracranial medical image using a segmentation model to obtain a segmented intracranial arterial tree;

[0087] Step S300 , combining the three-dimensional position coordinates of the aneurysm and the segmented intracranial arterial tree to perform three-dimensional reconstruction to obtain a three-dimensional intracranial vascular model including the intracranial arterial tree and aneurysm position information;

[0088] Step S400: Establish a standard model of intracranial blood vessels, perform segmented annotation, and obtain a standard model of intracranial blood vessels with segmented annotation information;

[0089] Step S500, registering the intracranial vascular three-dimensional model with the intracranial vascular standard model to obtain a registered intracranial vascular three-dimensional model;

[0090] In step S600, on the registered three-dimensional intracranial vascular model, pixel points on the aneurysm that are within a preset threshold range from the intracranial arterial tree are selected, and the corresponding mapping points of these pixel points on the standard intracranial vascular model are obtained. The segmented position of the aneurysm on the intracranial arterial tree is obtained based on the positions of these mapping points.

[0091] In one embodiment, a computer program product is provided, comprising computer instructions, which, when executed by a processor, implement the following steps:

[0092] Step S100 , detecting and segmenting aneurysms in intracranial medical images using an aneurysm detection network to obtain the three-dimensional position coordinates of the aneurysm;

[0093] Step S200 , segmenting the arterial tree in the intracranial medical image using a segmentation model to obtain a segmented intracranial arterial tree;

[0094] Step S300 , combining the three-dimensional position coordinates of the aneurysm and the segmented intracranial arterial tree to perform three-dimensional reconstruction to obtain a three-dimensional intracranial vascular model including the intracranial arterial tree and aneurysm position information;

[0095] Step S400: Establish a standard model of intracranial blood vessels, perform segmented annotation, and obtain a standard model of intracranial blood vessels with segmented annotation information;

[0096] Step S500, registering the intracranial vascular three-dimensional model with the intracranial vascular standard model to obtain a registered intracranial vascular three-dimensional model;

[0097] In step S600, on the registered three-dimensional intracranial vascular model, pixel points on the aneurysm that are within a preset threshold range from the intracranial arterial tree are selected, and the corresponding mapping points of these pixel points on the standard intracranial vascular model are obtained. The segmented position of the aneurysm on the intracranial arterial tree is obtained based on the positions of these mapping points.

[0098] In this embodiment, the computer program product includes a program code portion for executing the steps of the method for obtaining the location of an intracranial aneurysm in a blood vessel segment in each embodiment of the present application when the computer program product is executed by one or more computing devices. The computer program product can be stored on a computer-readable recording medium. The computer program product can also be provided for download via a data network (e.g., via a RAN, via the Internet and / or via an RBS). Alternatively or additionally, the method can be encoded in a field programmable gate array (FPGA) and / or an application-specific integrated circuit (ASIC), or the functionality can be provided for download with the aid of a hardware description language.

[0099] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0100] The technical features of the above embodiments may be combined in any manner. To simplify the description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there are no conflicts in the combination of these technical features, they should be considered to be within the scope of this specification. When technical features in different embodiments are reflected in the same figure, it can be regarded as that figure also discloses the combination examples of the various embodiments involved.

[0101] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A method for obtaining the location of an intracranial aneurysm in a blood vessel segment, characterized in that: include: Aneurysm detection network is used to detect and segment aneurysms in intracranial medical images and obtain the three-dimensional position coordinates of the aneurysm. The segmentation model is used to segment the arterial tree in the intracranial medical image to obtain the segmented intracranial arterial tree; Combining the three-dimensional position coordinates of the aneurysm and the segmented intracranial arterial tree, a three-dimensional reconstruction is performed to obtain a three-dimensional intracranial vascular model including the intracranial arterial tree and aneurysm position information; Establish a standard model of intracranial blood vessels, perform segmented annotation, and obtain a standard model of intracranial blood vessels with segmented annotation information; Performing spatial registration on the intracranial vascular three-dimensional model and the intracranial vascular standard model to obtain a registered intracranial vascular three-dimensional model; On the registered three-dimensional model of intracranial blood vessels, pixel points on the aneurysm that are within a preset threshold range from the intracranial arterial tree are selected to obtain the corresponding mapping points of the pixel points on the standard intracranial blood vessel model, and the segmented position of the aneurysm on the intracranial arterial tree is obtained according to the position of the mapping points.

2. The method for obtaining the location of an intracranial aneurysm in a blood vessel segment according to claim 1, characterized in that: The pixel points and the mapping points correspond one to one, For any pixel point, the point on the intracranial blood vessel standard model that is closest to it is the corresponding mapping point.

3. The method for obtaining the location of intracranial aneurysms in blood vessel segments according to claim 1, characterized in that: Obtaining the segmented position of the aneurysm on the intracranial arterial tree according to the position of the mapping point specifically includes: The segmented labeling information of the position corresponding to the maximum number of mapping points is obtained, and the segmented labeling information is translated into a specific segmented position to obtain the segmented position of the aneurysm on the intracranial arterial vascular tree.

4. The method for obtaining the location of intracranial aneurysms in blood vessel segments according to claim 3, characterized in that: When obtaining the segmented annotation information corresponding to the position with the maximum number of mapping points, if the relative difference between the maximum number of mapping points appearing in two segments is less than a first threshold, it is determined that the specific segmented position of the aneurysm is at the intersection of the two segments.

5. The method for obtaining the location of intracranial aneurysms in blood vessel segments according to claim 1, characterized in that: The intracranial vascular standard model is derived from an intracranial vascular standard model library, which includes at least two intracranial vascular standard models with different morphologies. Performing spatial registration on the intracranial vascular three-dimensional model and the intracranial vascular standard model to obtain a registered intracranial vascular three-dimensional model specifically includes: The intracranial vascular three-dimensional model and at least two different intracranial vascular standard models are spatially registered respectively, and a registration result with a relatively small spatial distance difference is selected to obtain a registered intracranial vascular three-dimensional model.

6. The method for obtaining the location of intracranial aneurysms in blood vessel segments according to claim 5, characterized in that: The intracranial vascular standard models in the intracranial vascular standard model library are organized and classified according to gender and / or age.

7. The method for obtaining the location of intracranial aneurysms in blood vessel segments according to claim 1, characterized in that: Obtaining a three-dimensional intracranial vascular model including intracranial arterial tree and aneurysm location information, further comprising: performing a closing operation on the three-dimensional model of the intracranial blood vessels with the aneurysm location information to obtain the three-dimensional model of the intracranial blood vessels that has completed the closing operation; extracting the maximum connected domain of the three-dimensional intracranial blood vessel model that completes the closing operation, to obtain a three-dimensional intracranial blood vessel model without scattered points; The Laplace smoothing algorithm is used to perform surface smoothing on the intracranial vascular three-dimensional model with scattered points removed, and a three-dimensional intracranial vascular model including the intracranial arterial tree and aneurysm location information is obtained.

8. The method for obtaining the location of intracranial aneurysms in blood vessel segments according to claim 1, characterized in that: Performing spatial registration of the intracranial blood vessel three-dimensional model and the intracranial blood vessel standard model, further comprising: Perform spatial registration on the intracranial vascular three-dimensional model including the intracranial arterial tree and aneurysm location information and the intracranial vascular standard model to obtain a translation and rotation matrix; According to the translation and rotation matrix, the three-dimensional model of the intracranial blood vessels is subjected to corresponding translation and rotation transformation to obtain a registered three-dimensional model of the intracranial blood vessels.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory, wherein: The processor executes the computer program to implement the steps of the method for obtaining the location of an intracranial aneurysm in a blood vessel segment according to any one of claims 1 to 8.

Images