Method and system for blood vessel analysis based on multi-contrast magnetic resonance images

Abstract

The application provides a blood vessel analysis method and system based on multi-contrast magnetic resonance images, and relates to the technical field of medical image processing.In the embodiment of the application, multi-contrast brain magnetic resonance scanning sequences (including T1W and TOF) can provide various contrast information, solve the problem that manual preprocessing of the skull part is required when only a single TOF image is used to extract blood vessel parameters in the prior art, and can additionally provide a transformation result in a standard brain space.In addition, compared with a manual and semi-automatic measurement method currently adopted, the embodiment of the application can realize a fully automatic parameter extraction process, is more time-saving and labor-saving, has no human factor intervention, and the result is more stable.In addition, the embodiment of the application can provide various blood vessel parameters of multiple blood vessel groups, and can quantize a blood vessel variation degree according to the transformation result in the standard brain space.

Description

Technical Field

[0001] The present invention relates to the technical field of medical image processing, and in particular to a blood vessel analysis method and system based on multi-contrast magnetic resonance imaging. Background Art

[0002] Current methods for assessing and analyzing brain vascular images are manual, requiring experienced personnel to evaluate images through visual inspection and manual measurement. However, this method is subject to significant subjectivity and consumes significant time and resources. However, some researchers have proposed automated and semi-automated programs for analyzing physiological parameters in brain vascular images. Using computer algorithms and libraries, automated analysis and assessment of large-scale brain vascular images can be achieved.

[0003] iCafe is a semi-automatic artery modeling tool developed for TOF images. It integrates modules such as vascular centerline tracking algorithms, arterial anatomical structure labeling, automatic vascular segmentation and analysis, and vascular structure visualization. Related studies have proposed a process for quantifying the degree of visualization of small intracranial blood vessels. Through image preprocessing, threshold segmentation, skeleton extraction, and other operations, physiological parameters such as vascular radius can be obtained. Some researchers are also committed to using the morphological parameters of intracranial arteries to construct statistical maps to improve understanding of the distribution of human brain arteries and subtle morphological changes. Related studies have also proposed a method for processing intracranial vascular statistical maps. This method uses a linear registration method to align TOF images to a standard brain template and calculates statistical maps including signal intensity, vascular probability, and average vessel diameter.

[0004] As can be seen, current algorithms primarily focus on extracting parameters from single-contrast images and fail to utilize information from multi-contrast images. Furthermore, most current algorithms rely on manual or semi-automatic parameter extraction (such as in skull preprocessing and categorization of different vascular groups in TOF images), which is time-consuming and labor-intensive. The selection of some parameters is influenced by subjective factors, making it difficult to standardize large amounts of data. Furthermore, current algorithms primarily focus on extracting a single parameter, namely, vascular radius, and are unable to provide a comprehensive set of parameters for multiple vascular groups. Summary of the Invention

[0005] The present invention provides a blood vessel analysis method and system based on multi-contrast magnetic resonance imaging to at least partially solve the above problems.

[0006] A first aspect of the present invention provides a blood vessel analysis method based on multi-contrast magnetic resonance imaging, the method comprising:

[0007] Obtaining DICOM format multi-contrast magnetic resonance images, performing format conversion and standardization processing on the DICOM format multi-contrast magnetic resonance images to obtain a to-be-processed image file, the to-be-processed image file comprising: T1W image data and TOF image data;

[0008] Performing registration preprocessing operation on the to-be-processed image file using a general brain image processing tool, obtaining a brain mask in TOF original space, performing TOF skull removal operation, and standard brain space registration operation;

[0009] Performing structured processing on the TOF image data in the original space to obtain a blood vessel segmentation image; converting the blood vessel segmentation image into a three-dimensional skeleton, converting the three-dimensional skeleton into a graph structure, generating a topological structure file comprising nodes and edges, the topological structure file representing the connectivity and original structure of the skeleton, determining the length, tortuosity and surface area of the blood vessels based on the topological structure file; obtaining the segmentation results of the bilateral internal carotid arteries, bilateral middle cerebral arteries, bilateral anterior cerebral arteries, bilateral posterior cerebral arteries and basilar artery based on the blood vessel segmentation image and a pre-trained classification network;

[0010] Determining the vessel radius, length, tortuosity, surface area and volume of different branch arteries based on the segmentation results of the bilateral internal carotid arteries, bilateral middle cerebral arteries, bilateral anterior cerebral arteries, bilateral posterior cerebral arteries and basilar artery; performing radius result assignment on the blood vessel segmentation image with respect to a statistical atlas in the standard brain space, searching for the nearest center line of a point in the distance mask, obtaining the radius value of the center line point, and obtaining a vessel radius file; transforming the vessel radius file and the blood vessel segmentation image to a standard brain template based on the registration preprocessing operation, and generating a statistical atlas of vessel radius distribution.

[0011] Optionally, the registration preprocessing operation comprises:

[0012] Obtaining a non-linear transformation file of the T1W image in the original space to the standard brain template, obtaining a linear transformation file of the TOF image in the original space to the T1W space in the original space, and transforming a TOF post-processing parameter file by using the non-linear transformation file in combination with the linear transformation file.

[0013] Optionally, the structured processing on the TOF image data in the original space to obtain the blood vessel segmentation image comprises:

[0014] Analyzing the gray histogram of the TOF image data to determine the maximum peak value, the frequency value after the maximum peak value decreases rapidly with the increase of brightness, and finally approaches the x-axis;

[0015] Selecting two points of the maximum peak value and the maximum brightness frequency to draw a straight line, and taking the point farthest from the straight line in the gray curve as the segmentation threshold.

[0016] The volumes of all connected domains are calculated based on the segmentation threshold, and connected domains larger than 95% of the total volume are retained to obtain a blood vessel segmentation image.

[0017] Optionally, determining the length, curvature, and surface area of ​​the blood vessel based on the topological structure file includes:

[0018] Traverse all edges, obtain the blood vessel segmentation blocks around each skeleton node on the edge, calculate the radius of the largest inscribed sphere around the skeleton node, and use it as the corresponding blood vessel radius on the skeleton node;

[0019] Obtain the vascular skeleton and its topological connection structure to be calculated, and determine the vascular length, curvature, and surface area point by point based on the position information and radius information of the skeleton nodes on each edge.

[0020] Optionally, the blood vessel length, curvature, and surface area are determined point by point based on the position information and radius information of the skeleton nodes on each edge, including:

[0021] ;

[0022] ;

[0023] ;

[0024] in,( x i ，y i ，z i )and( x i+1 ，y i+1 ，z i+1 ) represent the position information of two adjacent points on the edge that need to be calculated; length represents the length of the blood vessel, surface represents the vascular surface area, radius represents the vessel radius, tortuosity Indicates curvature, norm Represents the straight-line distance between the start and end points of a vessel.

[0025] Optionally, performing format conversion and standardization processing on the multi-contrast magnetic resonance image in DICOM format to obtain an image file to be processed includes:

[0026] Use the dcm2niix conversion tool to convert the original Dicom format file to Nifti format;

[0027] The obtained Nifti files were normalized using the median value of the files.

[0028] The second aspect of the present application provides a blood vessel analysis system based on multi-contrast magnetic resonance images, comprising:

[0029] A file management module is configured to acquire DICOM format multi-contrast magnetic resonance images, perform format conversion and standardization processing on the DICOM format multi-contrast magnetic resonance images, and obtain a to-be-processed image file, wherein the to-be-processed image file comprises T1W image data and TOF image data.

[0030] A registration preprocessing module is configured to perform registration preprocessing operation on the to-be-processed image file using a general brain image processing tool, acquire a brain mask in TOF original space, perform TOF skull removal operation, and standard brain space registration operation.

[0031] A structured processing module is configured to perform structured processing on the TOF image data in the original space, and acquire a blood vessel segmentation image; convert the blood vessel segmentation image into a three-dimensional skeleton, convert the three-dimensional skeleton into a graph structure, generate a topological structure file containing nodes and edges, wherein the topological structure file represents the connectivity and original structure of the skeleton, and determine the blood vessel length, tortuosity, and surface area based on the topological structure file; and obtain the segmentation results of bilateral internal carotid arteries, bilateral middle cerebral arteries, bilateral anterior cerebral arteries, bilateral posterior cerebral arteries, and basilar arteries based on the blood vessel segmentation image and a pre-trained classification network.

[0032] A physiological parameter generation module is configured to determine the blood vessel radius, length, tortuosity, surface area, and volume of different branch arteries based on the segmentation results of the bilateral internal carotid arteries, bilateral middle cerebral arteries, bilateral anterior cerebral arteries, bilateral posterior cerebral arteries, and basilar arteries; perform radius result distribution on the blood vessel segmentation image with respect to a statistical atlas in the standard brain space, search for the nearest center line of a point in the distance mask, acquire the radius value of the center line point, and obtain a blood vessel radius file; and transform the blood vessel radius file and the blood vessel segmentation image to a standard brain template based on the registration preprocessing operation, and generate a blood vessel radius distribution statistical atlas.

[0033] The third aspect of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the blood vessel analysis method based on multi-contrast magnetic resonance images according to the first aspect of the present application when executed.

[0034] The fourth aspect of the present application provides a computer readable storage medium, which stores a computer program, wherein the program is executed by a processor to implement the blood vessel analysis method based on multi-contrast magnetic resonance images according to the first aspect of the present application.

[0035] A fifth aspect of the present invention provides a computer program product comprising a computer program / instruction, which is used by a processor to implement the steps of the vascular analysis method based on multi-contrast magnetic resonance imaging as described in the first aspect of the present invention.

[0036] In an embodiment of the present invention, a multi-contrast brain magnetic resonance imaging sequence (including T1W brain structural images and TOF cerebral vascular images) is utilized. T1W and TOF can provide multiple contrast information, solving the problem of manual preprocessing of the skull when vascular parameter extraction using only a single TOF image in the past, and can also provide additional transformation results in a standard brain space. In addition, compared to the currently used manual and semi-automatic measurement methods, the embodiment of the present invention can achieve a fully automatic parameter extraction process, saving more manpower and time, without human intervention, and with more stable results. In addition, the embodiment of the present invention can provide multiple vascular parameters for multiple vascular groups and can quantify the degree of vascular variation based on the transformation results in a standard brain space.

[0037] The multi-contrast MRI-based vascular analysis method proposed in this embodiment of the present invention enables fully automated parameter extraction from brain vascular images. Compared to traditional algorithms, it offers more stable performance and is capable of batch processing large amounts of data. Furthermore, this embodiment of the present invention provides multiple parameters for the entire brain's vascular system and branch arteries based on multi-contrast images, as well as their quantitative parameter distribution under a standard brain, enabling comprehensive assessment of vascular images. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the description of the present invention. 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 any creative work.

[0039] Figure 1 is a flowchart of the steps of the blood vessel analysis method based on multi-contrast magnetic resonance imaging provided by the present invention;

[0040] Figure 2 is a detailed flowchart of the blood vessel analysis method based on multi-contrast magnetic resonance imaging provided by the present invention;

[0041] Figure 3 This is a sample image of a multi-contrast brain magnetic resonance scan sequence in the vascular analysis method based on multi-contrast magnetic resonance imaging provided by the present invention;

[0042] Figure 4 This is a structural block diagram of a blood vessel analysis system based on multi-contrast magnetic resonance imaging provided by the present invention. DETAILED DESCRIPTION

[0043] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0044] The diagnosis and treatment of cerebral vascular diseases rely on the accurate interpretation of brain vascular images. However, current methods for evaluating brain vascular images mainly focus on some basic image quality parameters, such as signal-to-noise ratio and other commonly used imaging assessment parameters. These parameters often cannot fully reflect the complex structure and functional status of blood vessels. In order to improve the accuracy of diagnosis and treatment efficacy, it is particularly important to develop a quantitative assessment scheme for brain vascular imaging. Currently, the commonly used brain vascular imaging scan sequences in clinical practice include T1-weighted brain structural images (T1 Weighted, T1W) and time-of-flight magnetic resonance angiography (Time-of-Flight, TOF) for cerebral vascular images.

[0045] T1W images are used to visualize brain anatomical structures and pathologies, such as gray and white matter, tumors, and hemorrhages. They play a vital role in the diagnosis of brain diseases and are therefore a fundamental sequence for clinical examinations.

[0046] TOF is a commonly used non-invasive cerebrovascular imaging technique that uses the inflow effect to obtain blood flow information. High-signal areas in the image represent the vascular lumen. This technique is commonly used to detect cerebral aneurysms, arteriovenous malformations, and other cerebrovascular abnormalities, and is also an important tool for the initial screening and monitoring of cerebrovascular disease. Extracting physiological parameters from large-scale brain MRI databases is essential, providing a reliable basis for subsequent scientific research and clinical applications.

[0047] To accurately assess brain vascular imaging, current researchers are focusing on vascular morphology, data maps, and parameters related to several major arterial branches. Morphologically, researchers are focusing on vascular radius, length, volume, and surface area. Data maps can be constructed by statistically analyzing cerebrovascular data from large populations to construct average vascular imaging maps, average vascular occurrence probability maps, and average vascular radius maps, thereby quantifying and analyzing subtle vascular changes in individual subjects.

[0048] Therefore, an embodiment of the present invention proposes a blood vessel analysis method based on multi-contrast magnetic resonance images. The method automatically, accurately and reliably utilizes information provided by multi-contrast magnetic resonance images to perform a comprehensive evaluation of blood vessel-related parameters.

[0049] In the embodiment of the present application, the multi-contrast magnetic resonance scanning sequence (including T1W and TOF) T1W and TOF can provide multiple contrast information, solve the problem of manual preprocessing of the skull part when only using single TOF image to extract the blood vessel parameters in the past, and can additionally provide the transformation result under the standard brain space. In addition, compared with the manual and semi-automatic measurement method currently used, the embodiment of the present application can realize a fully automatic parameter extraction process, which is more time-saving and labor-saving, and the result is more stable without human factor intervention. In addition, the embodiment of the present application can provide multiple blood vessel parameters of multiple blood vessel groups, and can quantify the degree of blood vessel variation according to the transformation result under the standard brain space.

[0050] Specifically, the embodiment of the present application develops a blood vessel analysis method based on multi-contrast magnetic resonance images, Figure 1 The step flow chart of the method is shown, Figure 2 The detailed step flow chart of the method is shown, as Figure 1 As shown, the method comprises the following steps:

[0051] S101, acquiring a multi-contrast magnetic resonance image in a first format (DICOM format), performing format conversion and standardization processing on the multi-contrast magnetic resonance image in the first format to obtain a to-be-processed image file, the to-be-processed image file comprising brain structure image (T1W image) data and brain blood vessel image (TOF image) data.

[0052] In the embodiment of the present application, the contrast magnetic resonance image comprises T1W and TOF, and specifically as Figure 3 As shown, it shows a multi-contrast magnetic resonance scanning sequence sample image, wherein the (a) part is a brain structure image T1W, and the (b) part is a brain blood vessel image TOF.

[0053] In the embodiment of the present application, the DICOM (Digital Imaging and Communications in Medicine, medical digital imaging and communication) format is mainly used for storing and transmitting medical images. The characteristics are rich metadata, including patient information, examination parameters, device information, etc., which is convenient for clinical diagnosis and image processing; hierarchical structure: using information object definition (IOD) and modular structure, supporting multiple image types and compression methods; multi-file slice: usually one file is generated for each slice, which is convenient for displaying different levels of structure, and is a series of two-dimensional layers.

[0054] In practical application, the original first format file can be converted into a more general second format (Nifti format).

[0055] The Nifti format is also used to store medical imaging data, making it suitable for scientific research and analysis. Its structure is simple, and the raw data is stored as a three-dimensional image, making it easy to process and analyze.

[0056] In practical applications, it is only necessary to collect and export images in DICOM format from the machine to complete the subsequent parameter extraction process based on the technical solution provided by the embodiment of the present invention.

[0057] In an optional implementation, step S101 includes the following sub-steps:

[0058] S1011, use the dcm2niix conversion tool to convert the original DICOM format file to Nifti format.

[0059] S1012: Normalize the obtained Nifti file using the median value of the file.

[0060] like Figure 2 As shown, the multi-contrast vascular image automated parameter extraction algorithm provided by the embodiment of the present invention mainly includes four stages: file management, registration preprocessing, structured processing flow, and physiological parameter generation. In the file management part, the dcm2niix conversion tool can be used to convert the file format, convert the original Dicom format file into the more general Nifti format, and simultaneously generate a Json file containing rich information. After completing the file format conversion and naming standardization, data standardization processing is performed on the acquired Nifti file. The specific operation is: use the median of the file for normalization processing to effectively prevent the data from being offset during the processing process. This step is of great significance for ensuring the accuracy and reliability of the data analysis results.

[0061] S102, using a general brain image processing tool to perform a registration preprocessing operation on the image file to be processed, obtaining a brain mask in the TOF original space, performing a TOF skull removal operation, and a standard brain space registration operation.

[0062] In an embodiment of the present invention, the general brain image processing tool may be an FSL tool.

[0063] In embodiments of the present invention, through a registration preprocessing operation, based on two contrast images (TOF and T1W), the TOF image can be transformed into T1W space, then reconfigured into a standard brain space. Within this standard space, a unified TOF skull removal operation can then be performed. In TOF images, the presence of skull bones may interfere with vascular segmentation, necessitating a prior skull removal operation. Other related brain vascular image assessment algorithms require deep learning algorithms or manual parameter adjustment to achieve skull removal. In contrast, in embodiments of the present invention, by uniformly reconfiguring the TOF image into the standard brain space and then obtaining a transformation matrix, the skull bones can be automatically removed. Furthermore, in embodiments of the present invention, after skull removal, the TOF image can be directly segmented using thresholding to calculate optimal segmentation points, thereby automatically extracting the segmented vascular image. This eliminates the need for manual intervention in the subsequent automatic annotation and topology generation processes.

[0064] In an optional embodiment, the registration preprocessing operation includes: obtaining a nonlinear transformation file of the T1W image in the original space to a standard brain template, obtaining a linear transformation file of the TOF image in the original space to linearly align it to the T1W space of the original space, and using the nonlinear transformation file in combination with the linear transformation file to transform the TOF post-processing parameter file.

[0065] After completing the file management, this patent uses the FSL tool to perform registration preprocessing operations on all Nifti files. In short, in the embodiment of the present invention, the registration preprocessing operation includes:

[0066] Use FSL tools to perform region of interest cropping, skull removal, linear registration, and nonlinear registration on the original T1W image to obtain the transformation field from the original scan space to a 0.5 mm standard brain template.

[0067] Perform TOF image registration. First, linearly register the TOF image in native space to the T1W image in native space. Utilize the previously acquired nonlinear transformation file from T1W native space to standard brain space, combined with the linear transformation file from TOF native space to T1W space, transform the standard brain mask to obtain a brain mask in TOF native space, and perform TOF skull removal.

[0068] Using the same transformation method (TOF→T1W→standard brain), such as Figure 2 As shown in the registration preprocessing section, the vascular segmentation image and vascular radius distribution image generated in subsequent steps are converted to the standard brain space.

[0069] The standard brain template is a standard brain template widely used in neuroimaging research. It is based on the brain MRI scan data of multiple normal individuals, and is spatially normalized through nonlinear deformation technology to create an average, standardized brain structure model. The advantage of the standard brain template is that it provides a unified coordinate system, which allows effective comparison and analysis between different studies. In an embodiment of the present invention, a standard brain template with a resolution of 0.5mm is used. This is because the template has a high spatial accuracy and can better capture the details of the brain structure. By aligning individual images to the standard brain template, the embodiment of the present invention can not only achieve spatial standardization of imaging data, but also promote the integration and analysis of brain imaging data across studies and populations, which is of great significance for revealing the universal laws of brain structure and function.

[0070] Specifically, the TOF image (TOF_orig) is linearly registered to the T1W structure (T1W_orig), and then mapped to the standard brain space through nonlinear registration. The process is divided into three parts:

[0071] Linear registration from TOF_orig to T1W_orig; nonlinear registration from T1W_orig to standard brain space; final result verification and mask processing.

[0072] Specifically, the linear registration (flirt) process of TOF_orig (TOF image) → T1W_orig (T1W image) includes:

[0073] Use FSL's flirt tool to transform the TOF image to the T1W structure (TOF_orig_to_T1W_orig). Generate a transformation matrix to record the spatial mapping relationship. Combine multiple transformation matrices using convert_xfm for multi-step registration or cross-modality alignment. Mask generation and application (mask): Perform brain extraction (T1W_roi_brain) on the T1W image and generate a brain mask (T1W_brain_mask) to exclude non-brain tissue such as the skull. Apply the mask to the TOF image (TOF_orig_brain) to ensure that subsequent analysis focuses only on the brain region.

[0074] The nonlinear registration process from T1W_orig (T1W image) to standard brain space includes:

[0075] Use maskFOV to roughly crop the region of interest, use bet to remove the skull area, use standard_space_roi to further reduce the region of interest to focus on a specific area in the standard space, use linear registration (flirt) to obtain a preliminary registration to the specific area transformation file in the standard space, and use nonlinear registration (fnirt) to register the T1W image to the standard brain template.

[0076] convertwrap combines the linear transformation field from TOF_orig to T1W_orig and the nonlinear transformation field from T1W_orig to standard brain space to map TOF images to standard brain space.

[0077] S103, performing structured processing on the TOF image data in the original space to obtain a vascular segmentation image; converting the vascular segmentation image into a three-dimensional skeleton, converting the three-dimensional skeleton into a graph structure, generating a topology structure file containing nodes and edges, wherein the topology structure file represents the connectivity and original structure of the skeleton, and determining the vascular length, curvature, and surface area based on the topology structure file; obtaining segmentation results of the bilateral internal carotid arteries, bilateral middle cerebral arteries, bilateral anterior cerebral arteries, bilateral posterior cerebral arteries, and basilar arteries based on the vascular segmentation image and a pre-trained classification network.

[0078] In an embodiment of the present invention, TOF brain image data in the original space is subjected to structured processing to obtain a file for subsequent parameter extraction.

[0079] In an optional embodiment, performing structured processing on the TOF image data in the original space to obtain a blood vessel segmentation image includes:

[0080] S1, analyze the grayscale histogram of the TOF image data to determine the maximum peak. After this maximum peak, the frequency value decreases rapidly as the brightness increases, and finally approaches the x-axis.

[0081] S2: Select the two points with the maximum peak value and the maximum brightness frequency to draw a straight line, and use the point farthest from the straight line in the grayscale curve as the segmentation threshold.

[0082] S3, calculating the volumes of all connected domains based on the segmentation threshold, retaining connected domains that are larger than 95% of the total volume, and obtaining a blood vessel segmentation image.

[0083] In the embodiment of the present application, the gray histogram of the image is analyzed, and some peak values can be observed, one of which is the most significant. The frequency value decreases rapidly with the increase of brightness after the maximum peak value, and finally approaches the x-axis. The two points of the maximum peak value and the maximum brightness frequency are selected to draw a straight line, and the point farthest from the straight line in the gray curve is the segmentation threshold. Then the volume of all connected domains is calculated, and the connected domains greater than 95% of the total volume are retained, which is the blood vessel segmentation result, and the blood vessel segmentation image is obtained.

[0084] Subsequently, in the embodiment of the present application, the Skeleton3D function in MATLAB can be used to convert the segmentation image into a skeleton, which is used as the blood vessel center line, and then the Skel2Graph3D function is used to convert the three-dimensional skeleton into a graph structure, generating a topological structure file containing nodes and edges, representing the connectivity of the skeleton and reflecting its original structure, which is used for subsequent analysis. The algorithm will traverse all edges to obtain the blood vessel segmentation cut block around each node on the edge, and calculate the radius of the maximum inscribed sphere in the region as the corresponding blood vessel radius of the skeleton node. In the embodiment of the present application, the blood vessel volume is defined as the total volume of the blood vessel segmentation result in the region of interest (which can be determined according to the subsequent segmentation results of the bilateral internal carotid artery, bilateral middle cerebral artery, bilateral anterior cerebral artery, bilateral posterior cerebral artery, and basilar artery). For the calculation of the blood vessel length, tortuosity, and surface area, first, the blood vessel skeleton and its topological connection structure to be calculated are obtained, and the blood vessel length, tortuosity, and surface area are calculated point by point according to the position information and radius information of the nodes on each edge.

[0085] In an alternative embodiment, determining the blood vessel length, tortuosity, and surface area based on the topological structure file comprises:

[0086] S11, traversing all edges, obtaining the blood vessel segmentation cut block around each skeleton node on the edge, and calculating the radius of the maximum inscribed sphere around the skeleton node as the corresponding blood vessel radius of the skeleton node.

[0087] S12, obtaining the blood vessel skeleton and its topological connection structure to be calculated, and determining the blood vessel length, tortuosity, and surface area point by point according to the position information and radius information of the skeleton nodes on each edge.

[0088] In an alternative embodiment, the blood vessel length, tortuosity, and surface area are calculated using the following formulas.

[0089] ;

[0090] ;

[0091] ;

[0092] wherein, xi ，y i ，z i )and( x i+1 ，y i+1 ，z i+1 ) represent the position information of two adjacent points on the edge that need to be calculated; length represents the length of the blood vessel, surface represents the vascular surface area, radius represents the vessel radius, tortuosity Indicates curvature, norm Represents the straight-line distance between the start and end points of a vessel.

[0093] In practical applications, after determining the segmentation results of the bilateral internal carotid arteries, bilateral middle cerebral arteries, bilateral anterior cerebral arteries, bilateral posterior cerebral arteries, and basilar arteries, the vascular radius, length, curvature, surface area, and volume of different branch arteries can be determined based on the topological structure and the above formula.

[0094] In an embodiment of the present invention, automatic artery labeling is performed using a pre-trained nnU-net network. This pre-trained nnU-net network can output segmentation results for the bilateral internal carotid arteries, bilateral middle cerebral arteries, bilateral anterior cerebral arteries, bilateral posterior cerebral arteries, and basilar arteries based on an input vascular segmentation file.

[0095] The segmentation neural network (nnU-net) used in this embodiment is a type of deep learning network and a well-established, publicly available segmentation algorithm with excellent performance on a variety of datasets. This model can segment both 2D and 3D images.

[0096] like Figure 2 As shown, in an embodiment of the present invention, in the structured processing flow, the processes of blood vessel segmentation, centerline extraction, automatic blood vessel labeling and topological structure generation are completed.

[0097] S104, based on the segmentation results of the bilateral internal carotid arteries, bilateral middle cerebral arteries, bilateral anterior cerebral arteries, bilateral posterior cerebral arteries, and basilar arteries, the vascular radius, length, curvature, surface area, and volume of different branch arteries are determined; with respect to a statistical atlas under a standard brain space, the radius results are assigned to the vascular segmentation image, the center line closest to a point in the mask is searched, the radius value of the center line point is obtained, and a vascular radius file is obtained; based on the registration preprocessing operation, the vascular radius file and the vascular segmentation image are transformed into a standard brain template to generate a statistical atlas of vascular radius distribution.

[0098] like Figure 2As shown, in the physiological parameter generation stage, the vessel radius, length, tortuosity, surface area and volume of different branch arteries are calculated according to the automatic labeling result of the artery.

[0099] For the statistical atlas under the standard brain space, first, the radius result is assigned to the vessel segmentation mask, the nearest center line point in the distance mask is searched, and the radius value of the center line point is obtained. Finally, the vessel radius file and the vessel segmentation file are transformed into the standard brain space, and can be quantitatively compared with the average atlas in the database.

[0100] In the embodiment of the application, in the process of calculating the statistical atlas, first, each part of the vessel needs to be assigned a corresponding radius value. For the calculation of the radius value, the embodiment of the application proposes that the nearest center line point to the vessel segmentation mask is obtained, and the radius value of the center line point is obtained. By using this strategy, the entire vessel region can be fully assigned with a radius, and each point on the vessel is assigned with a radius. Then, the vessel segmentation file with the assigned vessel radius is transformed into the standard standard brain space, and the average value is obtained, which can be quantitatively compared with the average atlas in the database. Specifically, the radius possessed by each position of the vessel image can be quantitatively compared with the radius possessed by each position of the vessel image in the average atlas.

[0101] The blood vessel analysis method based on multi-contrast magnetic resonance images provided by the embodiment of the application can realize full-automatic extraction of parameters of brain vessel images, has more stable performance compared with traditional algorithms, and can perform batch processing on large-scale data. Moreover, the embodiment of the application can provide various parameters including whole brain vessels and branch arteries based on multi-contrast images, and provide quantitative parameter distribution under the standard brain, so as to realize comprehensive evaluation of the vessel images.

[0102] Based on the same inventive concept, the application further provides a blood vessel analysis system based on multi-contrast magnetic resonance images, like Figure 4 As shown, the blood vessel analysis system 400 based on multi-contrast magnetic resonance images comprises:

[0103] A file management module 401 is configured to acquire multi-contrast magnetic resonance images in DICOM format, perform format conversion and standardization processing on the multi-contrast magnetic resonance images in DICOM format, and obtain a to-be-processed image file, wherein the to-be-processed image file comprises T1W image data and TOF image data.

[0104] A registration preprocessing module 402 is configured to perform a registration preprocessing operation on the to-be-processed image file using a general brain image processing tool, acquire a brain mask in a TOF original space, perform a TOF skull removal operation, and perform a standard brain space registration operation.

[0105] The structured processing module 403 is configured to perform structured processing on the TOF image data in the original space to obtain a vascular segmentation image; convert the vascular segmentation image into a three-dimensional skeleton, convert the three-dimensional skeleton into a graph structure, generate a topology file containing nodes and edges, wherein the topology file represents the connectivity and original structure of the skeleton; determine the vascular length, curvature, and surface area based on the topology file; and obtain segmentation results for the bilateral internal carotid arteries, bilateral middle cerebral arteries, bilateral anterior cerebral arteries, bilateral posterior cerebral arteries, and basilar arteries based on the vascular segmentation image and a pre-trained classification network;

[0106] Physiological parameter generation module 404 is configured to determine the vascular radius, length, curvature, surface area, and volume of different branch arteries based on the segmentation results of the bilateral internal carotid arteries, bilateral middle cerebral arteries, bilateral anterior cerebral arteries, bilateral posterior cerebral arteries, and basilar arteries; assign radius results to the vascular segmentation image based on a statistical atlas in a standard brain space, search for the centerline closest to a point in the mask, obtain the radius value of the centerline point, and obtain a vascular radius file; and transform the vascular radius file and the vascular segmentation image into a standard brain template based on the registration preprocessing operation to generate a statistical atlas of vascular radius distribution.

[0107] Optionally, the registration pre-processing module 402 is configured to:

[0108] Obtain a nonlinear transformation file of the original space T1W image to the standard brain template, obtain a linear transformation file of the original space TOF image linearly aligned to the original space T1W space, and use the nonlinear transformation file combined with the linear transformation file to transform the TOF post-processing parameter file.

[0109] Optionally, the structured processing module 403 is configured to:

[0110] Analyze the grayscale histogram of the TOF image data to determine the maximum peak. After this peak, the frequency value decreases rapidly as the brightness increases, and eventually approaches the x-axis.

[0111] Select the two points of the maximum peak and the maximum brightness frequency to draw a straight line, and use the point farthest from the straight line in the grayscale curve as the segmentation threshold;

[0112] The volumes of all connected domains are calculated based on the segmentation threshold, and connected domains larger than 95% of the total volume are retained to obtain a blood vessel segmentation image.

[0113] Optionally, the structured processing module 403 is further configured to:

[0114] Traverse all edges, obtain the blood vessel segmentation blocks around each skeleton node on the edge, calculate the radius of the largest inscribed sphere around the skeleton node, and use it as the corresponding blood vessel radius on the skeleton node;

[0115] Obtain the vascular skeleton and its topological connection structure to be calculated, and determine the vascular length, curvature, and surface area point by point based on the position information and radius information of the skeleton nodes on each edge.

[0116] Optionally, the blood vessel length, curvature, and surface area are determined point by point based on the position information and radius information of the skeleton nodes on each edge, including:

[0117] ;

[0118] ;

[0119] ;

[0120] in,( x i ，y i ，z i )and( x i+1 ，y i+1 ，z i+1 ) represent the position information of two adjacent points on the edge that need to be calculated; length represents the length of the blood vessel, surface represents the vascular surface area, radius represents the vessel radius, tortuosity Indicates curvature, norm Represents the straight-line distance between the start and end points of a vessel.

[0121] Optionally, the file management module 501 is used to:

[0122] Use the dcm2niix conversion tool to convert the original Dicom format file to Nifti format;

[0123] The obtained Nifti files were normalized using the median value of the files.

[0124] Based on the same inventive concept, the present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when executed by the processor, the steps of the vascular analysis method based on multi-contrast magnetic resonance imaging as described in any of the above embodiments are implemented.

[0125] Based on the same inventive concept, the present invention further provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the vascular analysis method based on multi-contrast magnetic resonance imaging described in any of the above embodiments.

[0126] Based on the same inventive concept, the present application provides a computer program product comprising computer programs / instructions which, when executed by a processor, implement the steps of the blood vessel analysis method based on multi-contrast magnetic resonance images according to any one of the above embodiments.

[0127] Each of the embodiments in the present specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be understood by referring to each other.

[0128] Those skilled in the art will understand that embodiments of the present application can be provided as methods, apparatuses, or computer program products. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0129] The present application is described with reference to flowcharts and / or block diagrams of the method, terminal device (apparatus), and computer program product according to the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable terminal device to produce a machine, so that the instructions executed by the processor of the computer or other programmable terminal device produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that performs the functions specified in one or more blocks.

[0130] These computer program instructions can also be stored in a computer-readable memory that can direct the computer or other programmable terminal device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including instruction apparatuses that implement the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that performs the functions specified in one or more blocks.

[0131] These computer program instructions can also be loaded into a computer or other programmable terminal device, so that a series of operation steps are performed on the computer or other programmable terminal device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable terminal device provide a process for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1A step that specifies a function in one or more boxes.

[0132] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0133] Finally, it should be noted that, in the present invention, relational terms such as first and second, etc., are merely used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises", or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or terminal device that includes a series of elements includes not only those elements, but also other elements that are not explicitly listed, or also includes elements that are inherent to such process, method, article, or terminal device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes the element.

[0134] The above describes in detail the vascular analysis method based on multi-contrast magnetic resonance imaging provided by the present invention. Specific examples are used in the present invention to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only intended to help understand the method and core concept of the present invention. At the same time, for those skilled in the art, based on the concept of the present invention, there may be changes in the specific implementation methods and application scope. In summary, the contents of this specification should not be understood as limiting the present invention.

Claims

1. A blood vessel analysis method based on multi-contrast magnetic resonance imaging, characterized in that: The method comprises: Acquire a multi-contrast magnetic resonance image in DICOM format, perform format conversion and standardization processing on the multi-contrast magnetic resonance image in DICOM format, and obtain an image file to be processed, wherein the image file to be processed includes: T1W image data and TOF image data; A general brain image processing tool is used to perform a registration preprocessing operation on the image file to be processed, obtain a brain mask in the TOF original space, perform a TOF skull removal operation, and perform a standard brain space registration operation; the TOF image is linearly registered to the T1W image, and then mapped to the standard brain standard space through nonlinear registration. The nonlinear registration process of the T1W image to the standard brain space includes: using maskFOV to roughly crop the region of interest, using bet to remove the skull area, using standard_space_roi to further reduce the region of interest to focus on a specific area in the standard space, using linear registration (flirt) to obtain a transformation file for the specific area in the standard space, and using nonlinear registration (fnirt) to register the T1W image to the standard brain template; The TOF image data in the original space is structured to obtain a vascular segmentation image; the vascular segmentation image is converted into a three-dimensional skeleton, and the three-dimensional skeleton is converted into a graph structure to generate a topology structure file containing nodes and edges. The topology structure file represents the connectivity and original structure of the skeleton, and the vascular length, curvature, and surface area are determined based on the topology structure file; based on the vascular segmentation image and a pre-trained classification network, segmentation results of the bilateral internal carotid arteries, bilateral middle cerebral arteries, bilateral anterior cerebral arteries, bilateral posterior cerebral arteries, and basilar arteries are obtained; Based on the segmentation results of the bilateral internal carotid arteries, bilateral middle cerebral arteries, bilateral anterior cerebral arteries, bilateral posterior cerebral arteries, and basilar arteries, the vascular radius, length, curvature, surface area, and volume of different branch arteries are determined; based on the statistical atlas under the standard brain space, the radius results are assigned to the vascular segmentation image, and the center line closest to a point in the mask is searched to obtain the radius value of the center line point to obtain a vascular radius file; based on the alignment preprocessing operation, the vascular radius file and the vascular segmentation image are transformed into a standard brain template to generate a statistical atlas of vascular radius distribution.

2. The blood vessel analysis method based on multi-contrast magnetic resonance imaging according to claim 1, characterized in that: The registration preprocessing operation includes: Obtain a nonlinear transformation file of the original space T1W image to the standard brain template, obtain a linear transformation file of the original space TOF image linearly aligned to the original space T1W space, and use the nonlinear transformation file combined with the linear transformation file to transform the TOF post-processing parameter file.

3. The blood vessel analysis method based on multi-contrast magnetic resonance imaging according to claim 1, characterized in that: Perform structured processing on the TOF image data in the original space to obtain the vascular segmentation image, including: Analyze the grayscale histogram of the TOF image data to determine the maximum peak. After this peak, the frequency value decreases rapidly as the brightness increases, and eventually approaches the x-axis. Select the two points of the maximum peak and the maximum brightness frequency to draw a straight line, and use the point farthest from the straight line in the grayscale curve as the segmentation threshold; The volumes of all connected domains are calculated based on the segmentation threshold, and connected domains larger than 95% of the total volume are retained to obtain a blood vessel segmentation image.

4. The blood vessel analysis method based on multi-contrast magnetic resonance imaging according to claim 1, characterized in that: Determining the length, curvature, and surface area of ​​the blood vessel based on the topological structure file includes: Traverse all edges, obtain the blood vessel segmentation blocks around each skeleton node on the edge, calculate the radius of the largest inscribed sphere around the skeleton node, and use it as the corresponding blood vessel radius on the skeleton node; Obtain the vascular skeleton and its topological connection structure to be calculated, and determine the vascular length, curvature, and surface area point by point based on the position information and radius information of the skeleton nodes on each edge.

5. The blood vessel analysis method based on multi-contrast magnetic resonance imaging according to claim 4, characterized in that: The length, curvature, and surface area of ​​the blood vessels are determined point by point based on the position and radius information of the skeleton nodes on each edge, including: ； ； ； in,( x i ，y i ，z i )and( x i+1 ，y i+1 ，z i+1 ) represent the position information of two adjacent points on the edge that need to be calculated; length represents the length of the blood vessel, surface represents the vascular surface area, radius represents the vessel radius, tortuosity Indicates curvature, norm Represents the straight-line distance between the start and end points of a vessel.

6. The blood vessel analysis method based on multi-contrast magnetic resonance imaging according to claim 1, characterized in that: Performing format conversion and standardization processing on the multi-contrast magnetic resonance image in the DICOM format to obtain an image file to be processed, including: Use the dcm2niix conversion tool to convert the original Dicom format file to Nifti format; The obtained Nifti files were normalized using the median value of the files.

7. A vascular analysis system based on multi-contrast magnetic resonance imaging, characterized in that: The multi-contrast magnetic resonance imaging-based vascular analysis system includes: A file management module is used to obtain multi-contrast magnetic resonance images in DICOM format, perform format conversion and standardization processing on the multi-contrast magnetic resonance images in DICOM format, and obtain image files to be processed, wherein the image files to be processed include: T1W image data and TOF image data; A registration preprocessing module is used to perform registration preprocessing operations on the image file to be processed using a general brain image processing tool, obtain a brain mask in the TOF original space, perform a TOF skull removal operation, and perform standard brain space registration operations; the TOF image is linearly registered to the T1W image, and then mapped to the standard brain standard space through nonlinear registration. The nonlinear registration process of the T1W image to the standard brain space includes: using maskFOV to roughly crop the region of interest, using bet to remove the skull area, using standard_space_roi to further reduce the region of interest to focus on a specific area in the standard space, using linear registration (flirt) to obtain a transformation file for preliminary registration to the specific area in the standard space, and using nonlinear registration (fnirt) to register the T1W image to the standard brain template; a structured processing module for performing structured processing on the TOF image data in the original space to obtain a vascular segmentation image; converting the vascular segmentation image into a three-dimensional skeleton, converting the three-dimensional skeleton into a graph structure, generating a topology file containing nodes and edges, wherein the topology file characterizes the connectivity and original structure of the skeleton, and determining the vascular length, curvature, and surface area based on the topology file; and obtaining segmentation results of the bilateral internal carotid arteries, bilateral middle cerebral arteries, bilateral anterior cerebral arteries, bilateral posterior cerebral arteries, and basilar arteries based on the vascular segmentation image and a pre-trained classification network; The physiological parameter generation module is used to determine the vascular radius, length, curvature, surface area, and volume of different branch arteries based on the segmentation results of the bilateral internal carotid arteries, bilateral middle cerebral arteries, bilateral anterior cerebral arteries, bilateral posterior cerebral arteries, and basilar arteries; assign radius results to the vascular segmentation image based on the statistical atlas in the standard brain space, search for the center line closest to a point in the mask, obtain the radius value of the center line point, and obtain a vascular radius file; based on the registration preprocessing operation, transform the vascular radius file and the vascular segmentation image into a standard brain space template to generate a statistical atlas of vascular radius distribution.

8. 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 blood vessel analysis method based on multi-contrast magnetic resonance imaging according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the blood vessel analysis method based on multi-contrast magnetic resonance imaging according to any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the steps of the blood vessel analysis method based on multi-contrast magnetic resonance imaging according to any one of claims 1 to 6 are implemented.

Images