Carotid artery plaque analysis device based on magnetic resonance imaging

Abstract

This invention relates to a carotid artery plaque analysis device based on magnetic resonance imaging (MRI). Breaking with conventional research focusing solely on carotid artery plaques, this device explores the relationship between plaques and brain structure. By analyzing early brain changes, particularly brain volume data, functional network topology data, and white matter fiber tract data, a device for analyzing carotid artery plaque properties is constructed. This device applies theoretical research to practical applications, qualitatively analyzing carotid artery plaque properties to determine whether plaques are warning signs with a high probability of rupture or safe plaques with a low probability of rupture. This analysis device can be used to predict and prevent cerebral ischemia. This device can be used for clinical early warning and preventive measures, significantly improving diagnostic accuracy and saving lives.

Description

Technical Field

[0001] The present invention relates to image analysis technology, in particular to a magnetic resonance image analysis technology. Background Art

[0002] Research on early prevention of stroke, especially ischemic stroke, is extremely necessary.

[0003] Previous studies have mostly focused on patients with moderate to severe carotid artery stenosis, with less attention paid to those with mild carotid stenosis. Clinically, the vulnerability of carotid atherosclerotic plaques, rather than the degree of stenosis, is the most important factor leading to sudden cerebral infarction. However, current anti-atherosclerotic plaque treatment strategies cannot completely prevent cerebral infarction. A growing number of studies have found that the default mode network (DMN, referring to the brain regions that are consistently active at rest, with key functional nodes including the medial prefrontal cortex (mPFC), precuneus, angular gyrus, anterior cingulate cortex (ACC), and posterior cingulate cortex (PCC)) is involved in regulating neuropsychiatric disorders mediated by inflammatory immune responses. In particular, the ACC and mPFC within the DMN may be higher-order cortical regions involved in systemic inflammatory activity, potentially modulating systemic inflammatory activity and contributing to the carotid atherosclerotic process. It is currently believed that the inflammatory immune response within the plaque is the key factor influencing the vulnerability of carotid artery plaques: the inflammatory response mediates the entire process of vulnerable plaque evolution and thrombotic events by promoting oxidative stress, lipid metabolism, new angiogenesis, and even feedback to central neurotransmitters. Studies have confirmed that the amygdala and hypothalamus are key brain structures regulating the inflammatory response in atherosclerosis. The higher cortical ACC and mPFC in the brain are interconnected through numerous nodes with subcortical structures (amygdala and hypothalamus) that control the autonomic nervous system. They regulate the activity of central neurotransmitters-mediated inflammatory markers within the plaque, coordinate hemodynamic changes near the vulnerable plaque, and promote rupture of vulnerable carotid artery plaques.

[0004] Therefore, how to explore brain changes in patients with carotid artery plaques without obvious stenosis, find early specific imaging markers of carotid artery plaques, and establish the relationship between the two are technical problems that need to be solved urgently in the field of early screening of high-risk stroke, especially ischemic stroke. Summary of the Invention

[0005] To solve the above technical problems, the present invention proposes a carotid artery plaque analysis device based on magnetic resonance imaging, comprising:

[0006] an acquisition module, for acquiring a magnetic resonance image;

[0007] a first extraction module, connected to the acquisition module, for extracting brain volume data based on the magnetic resonance image;

[0008] a second extraction module, connected to the acquisition module, for extracting functional network topology data based on the magnetic resonance image;

[0009] a third extraction module, connected to the acquisition module, for extracting white matter fiber bundle data based on the magnetic resonance image;

[0010] The control module is connected to the first extraction module, the second extraction module and the third extraction module, and is used to analyze the properties of the carotid artery plaque based on the brain volume data, the functional network topology data and the white matter fiber bundle data.

[0011] Furthermore, the first extraction module includes:

[0012] a first segmentation unit, configured to segment the magnetic resonance image to obtain a white matter segmentation image, a gray matter segmentation image, and a cerebrospinal fluid segmentation image;

[0013] The first calculation unit is used to calculate brain volume data according to the white matter segmentation image, the gray matter segmentation image and the cerebrospinal fluid segmentation image.

[0014] Furthermore, the second extraction module includes:

[0015] The second segmentation unit is used to segment the magnetic resonance image to obtain several brain regions;

[0016] The second computing unit regards each brain region as a node, which is used to determine the functional connection between two nodes and construct functional network topology data.

[0017] Furthermore, the second computing unit includes:

[0018] The correlation coefficient calculation unit is used to calculate the Pearson correlation coefficient between several brain regions;

[0019] The second judgment unit is used to determine whether the Pearson correlation coefficient between two brain regions exceeds a set threshold. If so, it is determined that there is a connection between the two brain regions; if not, it is determined that there is no connection between the two brain regions;

[0020] The network construction unit is used to construct functional network topology data based on the determination of whether there is a connection between two brain regions.

[0021] Furthermore, the third extraction module includes:

[0022] The brain region extraction unit is used to remove excess scalp before calculating the tensor and obtain a mask image to determine the tensor calculation range;

[0023] The third calculation unit calculates tensor parameters according to the tensor calculation range for analyzing white matter fiber bundle data.

[0024] Furthermore, the control module includes:

[0025] An integral calculation unit, used to calculate the integral of carotid artery plaque based on brain volume data, functional network topology data, and white matter fiber bundle data;

[0026] The qualitative analysis unit is connected to the integral calculation unit and is used to determine whether the integral of the carotid artery plaque exceeds a set threshold. If so, it is qualitatively classified as a warning plaque; if not, it is qualitatively classified as a safe plaque.

[0027] Furthermore, the integral of carotid artery plaque is calculated using formula (1):

[0028] P=K1*H1+ K2*H2+ K3*H3 (1)

[0029] Among them, P represents the integral of carotid artery plaque; K1, K2, and K3 represent the weights of brain volume data, functional network topology data, and white matter fiber bundle data, respectively; H1, H2, and H3 represent the individual integrals of brain volume data, functional network topology data, and white matter fiber bundle data, respectively.

[0030] Furthermore, the individual integrals of brain volume data were set as follows: if the whole brain volume was lower than the set threshold, 1 point was scored; if the gray matter volume of the right middle occipital gyrus exceeded the set threshold, 1 point was scored; if the gray matter volume of the bilateral lingual gyri exceeded the set threshold, 1 point was scored.

[0031] Furthermore, the individual integrals of the functional network topology are set as follows: if the centrality of the left middle frontal gyrus is higher than the set threshold, 1 point is scored; if the central betweenness of the right middle temporal gyrus is lower than the set threshold, 1 point is scored; if the efficacy of the left middle frontal gyrus and the right inferior parietal angular gyrus is higher than the set threshold, 1 point is scored; if the efficacy of the right middle temporal gyrus is lower than the set threshold, 1 point is scored.

[0032] Furthermore, the single integration of white matter fiber bundle data was set as follows: if the RD values ​​of the bilateral anterior thalamic radiation and inferior fronto-occipital fasciculus were greater than the set threshold, 1 point was scored; if the RD values ​​of the frontal part of the corpus callosum radiation and the uncinate fasciculus were greater than the set threshold, 2 points were scored; if the RD values ​​of the left corticospinal tract and cingulate were greater than the set threshold, 1 point was scored.

[0033] This invention provides a carotid artery plaque analysis device based on magnetic resonance imaging (MRI). This device breaks away from conventional research focusing solely on carotid artery plaques and explores the relationship between plaques and brain structure. By analyzing early brain changes, particularly brain volume data, functional network topology data, and white matter fiber tract data, a device for analyzing carotid artery plaque properties is constructed. This device applies theoretical research to practical applications, qualitatively analyzing carotid artery plaque properties to determine whether plaques are warning signs with a high probability of rupture or safe plaques with a low probability of rupture. This analysis device can be used to predict and prevent cerebral ischemia. This device can be used for clinical early warning and preventive measures, significantly improving diagnostic accuracy and saving lives. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a structural block diagram of the early warning device for carotid artery plaque rupture of the present invention. DETAILED DESCRIPTION

[0035] like Figure 1 As shown, the present invention provides a carotid artery plaque analysis device based on magnetic resonance imaging:

[0036] The acquisition module is configured to acquire magnetic resonance images. Specifically, the images may be acquired using, but not limited to, a head scan performed by a physician familiar with imaging using a magnetic resonance imaging device. Preferably, subsequent brain volume data extraction may include, but is not limited to, head scans using T1-weighted structural phase, T2-weighted structural phase, DWI, FLAIR, and three-dimensional brain volumetric magnetic resonance imaging (BRAVO) sequences. BRAVO sequence parameters include: pulse repetition time (TR): 7.792 ms, echo time (TE): 2.984 ms, flip angle: 7°, 188 slices, 1 mm slice thickness, 1 mm interslice spacing, 256x256 matrix, and voxel size: 1x1x1 mm³. To extract functional network topology data, rsfMRI can be used, but is not limited to it. Data acquisition uses an EPI sequence with the following parameters: TR: 2000 ms, TE: 30 ms, flip angle: 90°, 32 slices, 4 mm slice thickness, 64×64 matrix, voxel size: 3×3×3 mm³, and scan time of 360 s. To extract white matter fiber tract data, echo-planar imaging (DTI) can be used, but is not limited to it. Parameters include: TR (repetition time) = 12000 ms, TE (echo time) = 72.5 ms, matrix 256×256, field of view (FOV) = 230×230 mm², slice thickness 3 mm, 32 diffusion sensitivity gradient directions, and diffusion sensitivity coefficient (b) = 0 and 1000 s / mm². Conventional T1WI, T2WI, Flair, and diffusion-weighted imaging (DWI) scans were performed before DTI scanning to exclude previous and acute stroke and other intracranial organic lesions, and to evaluate the intracranial white matter hyperintensity.

[0037] The first extraction module is connected to the acquisition module and is configured to extract brain volume data based on the magnetic resonance images. Specifically, the brain volume data may include, but is not limited to, one or more parameters such as white matter volume, gray matter volume, cerebrospinal fluid volume, total brain volume (TCBV, which is the ratio of brain parenchymal volume to total cranial volume. Brain parenchymal volume is the sum of white matter volume and gray matter volume after excluding cerebrospinal fluid volume), and the white matter volume, gray matter volume, and cerebrospinal fluid volume of each brain region.

[0038] The second extraction module is connected to the acquisition module and is used to extract brain functional network topology data based on the magnetic resonance image. Specifically, the functional network topology data can be, but is not limited to, the connectivity of each brain region.

[0039] The third extraction module, connected to the acquisition module, is used to extract white matter fiber tract data from the MRI. Specifically, this white matter fiber tract data can be, but is not limited to, various DTI parameters. Tensor parameters such as FA, MD, AD, and RD are calculated for each subject for use in TBSS analysis.

[0040] The control module is connected to the first extraction module, the second extraction module, and the third extraction module and is configured to analyze the properties of the carotid artery plaque based on the brain volume data, the functional network topology data, and the white matter fiber tract data. Specifically, the control module may optionally, but is not limited to, determine whether the carotid artery plaque is a warning plaque with a high rupture probability or a safe plaque with a low rupture probability based on the three features of the brain volume data, the functional network topology data, and the white matter fiber tract data.

[0041] In this embodiment, the present invention proposes a carotid artery plaque analysis device based on magnetic resonance imaging. This device breaks away from conventional research on carotid artery plaques themselves and explores the relationship between plaques and brain structure. By analyzing early brain changes, particularly brain volume data, functional network topology data, and white matter fiber tract data, a carotid artery plaque property analysis device is constructed. This device applies theoretical research to practical applications, qualitatively analyzing the properties of carotid artery plaques. If a plaque is a warning plaque with a high rupture probability or a safe plaque with a low rupture probability, it can be used to predict and prevent cerebral ischemia. For example, if an examination result shows a carotid artery plaque without significant stenosis (an asymptomatic plaque), the carotid artery plaque analysis device of the present invention can qualitatively analyze the existing carotid artery plaque based on subtle changes in brain structure, determining its future trajectory and whether it is a warning plaque that may rupture and lead to ischemic stroke, or a safe plaque that will not rupture and lead to such a condition. This analysis device can be used for clinical early warning and preventive measures, significantly improving diagnostic accuracy and saving patients' lives. The carotid plaque analysis device of this invention investigates the vulnerability factors of carotid plaque (three early brain changes: brain volume data, functional network topology data, and white matter fiber tract data). This aims to develop a qualitative analysis device for carotid plaques based on magnetic resonance imaging (MRI) (functional brain imaging). This device facilitates early screening and differentiation of patients with similar intima-media thickness but varying stroke risks. The device explores the occurrence and development of carotid plaques from a neuromodulatory perspective, analyzes plaque properties, and deeply explores the neuromodulatory imaging features of carotid plaques of different types. Using intelligent technology, the device mines multimodal MRI data for prediction of carotid plaque types and analyzes the neuromodulatory mechanisms involved in carotid plaque evolution, providing guidance for early clinical screening of patients at increased risk of stroke.

[0042] Preferably, the first extraction module optionally includes, but is not limited to, a first segmentation unit for segmenting the magnetic resonance image to obtain a white matter segmentation image, a gray matter segmentation image, and a cerebrospinal fluid segmentation image. Specifically, a segmentation method for three-dimensional brain images can be used to segment and obtain the aforementioned individual images; a first calculation unit for calculating brain volume data based on the white matter segmentation image, the gray matter segmentation image, and the cerebrospinal fluid segmentation image. Preferably, the first extraction module also optionally includes, but is not limited to, a first preprocessing unit for preprocessing the magnetic resonance image to obtain a preprocessed magnetic resonance image. Specifically, the first extraction module optionally includes, but is not limited to, a first conversion unit for converting the magnetic resonance image into NIFIT format (generally, DICOM format, in which case format conversion is performed first) to obtain a converted first magnetic resonance image; and a first cropping unit for cropping a target region from the converted first magnetic resonance image and performing template matching to obtain a preprocessed first magnetic resonance image (specifically, optionally, but not limited to, removing the area below the head from the obtained magnetic resonance image to obtain a target region that only includes the head region, and then adjusting the image according to the Montreal template until template matching is completed). More specifically, the system may optionally include, but is not limited to, a first detection unit for detecting whether the quality of the white matter segmentation image, gray matter segmentation image, and cerebrospinal fluid segmentation image is qualified, and if unqualified, re-segmenting or discarding the image; a first smoothing unit for further smoothing the qualified white matter segmentation image, gray matter segmentation image, and cerebrospinal fluid segmentation image, preferably using a Gaussian smoothing kernel FWHM 8 for smoothing. More specifically, the first calculation unit may optionally include, but is not limited to, an element calculation subunit for calculating the white matter volume, gray matter volume, and cerebrospinal fluid volume based on the white matter segmentation image, gray matter segmentation image, and cerebrospinal fluid segmentation image; and a total brain volume calculation subunit for calculating the total brain volume as brain volume data based on the white matter volume, gray matter volume, and cerebrospinal fluid volume (specifically, the total cerebral brain volume (TCBV) is the ratio of the brain parenchymal volume to the total brain volume. The brain parenchymal volume is the sum of the white matter volume and the gray matter volume of the brain after excluding the cerebrospinal fluid volume). More specifically, it also includes a brain region volume calculation unit, which is used to calculate the white matter volume, gray matter volume and cerebrospinal fluid volume of each brain region as brain volume data based on the white matter segmentation image, gray matter segmentation image and cerebrospinal fluid segmentation image.

[0043] This embodiment provides the specific structure of the first extraction module, detailing how it extracts brain volume data from magnetic resonance images. It considers voxels as the basic building blocks of the brain, segmenting brain structural data into white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF). Statistical tests compare and analyze differences in the voxel composition of brain tissue components between groups, quantitatively detecting differences in brain tissue volume and density, and accurately displaying brain morphological abnormalities and the specific locations of the differentially located brain regions. It then extracts white matter volume, gray matter volume, cerebrospinal fluid volume, whole-brain volume, and the volumes of each local component.

[0044] More preferably, the second extraction module optionally includes, but is not limited to, a second segmentation unit for segmenting the magnetic resonance image to obtain a plurality of brain regions (the specific number may be, but is not limited to, 90); a second calculation unit for treating each brain region as a node, determining the functional connectivity between pairs of nodes, and constructing functional network topology data. Specifically, the second segmentation unit optionally includes, but is not limited to, using the AAL (Anatomical Automatic Labeling) brain template to segment the whole brain into a plurality of brain regions. The second calculation unit optionally includes, but is not limited to, constructing a functional network based on the average time series correlation between each brain region. Specifically, the average time series of each brain region in the rs-fMRI image data is the weighted sum of the time series of each brain region. More specifically, the second calculation unit optionally includes, but is not limited to, a correlation coefficient calculation unit for calculating the Pearson (r) correlation coefficient between each pair of brain regions; a second judgment unit for sequentially determining whether the Pearson correlation coefficient between each pair of brain regions exceeds a set threshold. If so, the pair of brain regions are deemed to be connected; if not, the pair of brain regions are deemed to be unconnected; and a network construction unit for constructing functional network topology data based on the judgment results. Specifically, the functional network topology data may be, but is not limited to, a correlation matrix of several * several (90 * 90). Preferably, the correlation matrix may be, but is not limited to, a binary matrix of several * several (90 * 90). If the Pearson correlation coefficient of any two brain regions exceeds a set threshold, i.e., there is a connection between the two regions, the value of the corresponding position in the matrix is ​​set to 1. If the Pearson correlation coefficient of any two brain regions does not exceed the set threshold, i.e., there is no connection between the two regions, the value of the corresponding position in the matrix is ​​set to 0. More specifically, the functional network topology data may be, but is not limited to, global network parameters, node network parameters, and network center nodes.More specifically, the second extraction module may optionally include but is not limited to: a second preprocessing unit for preprocessing the magnetic resonance image to obtain a preprocessed magnetic resonance image; specifically, the second preprocessing unit may optionally include but is not limited to: a second conversion subunit for converting the magnetic resonance image into NIFIT format (generally DICOM format, in which case format conversion is performed first); a second screening subunit for removing the head portrait data of the first five time points and continuing to process the remaining 175 time point data to reduce the impact of poor early data quality on the results; a second correction subunit including: time correction for eliminating the impact of different acquisition times of various parts of the brain on the results; position correction for adjusting each magnetic resonance image to the same position in the X, Y, and Z directions to correct the head position; standardization correction for aligning the head portrait data of each subject to the EPI template to reduce the impact of different head shapes on the analysis results; delinear drift: the rsfMRI image data signal has a time series, and the linear drift of this series is eliminated; regression covariates: reduce the impact of WM signals and CSF signals; filtering screening: select 0.01-0.08 Hz band; Rejection: If the subject's head data has a translation exceeding 3mm or a rotation angle exceeding 3°, the subject will be rejected.

[0045] This embodiment provides the specific structure of the second extraction module, detailing how it extracts functional network topology data from magnetic resonance images. The module first segments the brain into several regions. Using graph theory, the brain network is visualized as a three-dimensional graph consisting of nodes and edges, with brain regions or voxels representing nodes and the structural connections between nodes representing edges. This quantifies brain connectivity and allows analysis of changes in nodes or edges in the brain's structural functional network under various disease states, representing functional network topology data. Furthermore, a binary matrix of 0:1 is used to represent the functional network topology data, making it both computationally simple and easy to operate, as well as intuitive and understandable.

[0046] More preferably, the third extraction module optionally includes, but is not limited to, a brain region extraction unit for removing excess scalp before tensor calculation to generate a mask image to determine the range of tensor calculation; a third calculation unit for calculating tensor parameters (such as FA, MD, AD, and RD) for each subject for use in analyzing white matter fiber data. More specifically, the third extraction module optionally includes, but is not limited to, a third preprocessing unit for preprocessing the MRI images to generate preprocessed MRI images. Specifically, this includes data quality checks: checking basic data parameters, gradient directions, signal-to-noise ratio, and head motion; data format conversion: converting the original DICOM format to 4D NifTi format using Micron software; eddy current correction: eliminating head motion during scanning and deformation caused by head motion and eddy currents; and gradient direction correction: adjusting the original gradient direction based on the eddy current correction.

[0047] In this embodiment, the specific structure of the third extraction module is given, and how it extracts the specific form of white matter fiber bundle data based on the magnetic resonance image is explained in detail. The structure is simple and the data is accurate.

[0048] More preferably, the control module may optionally include, but is not limited to: an integral calculation unit for calculating the carotid artery plaque integral based on brain volume data, functional network topology data, and white matter fiber tract data; and a qualitative analysis unit, connected to the integral calculation unit, for determining whether the carotid artery plaque integral exceeds a set threshold. If so, the plaque is classified as a warning plaque; if not, the plaque is classified as a safe plaque. Specifically, the threshold can be determined by those skilled in the art based on actual circumstances. More specifically, the threshold can be determined based on basic information such as the examinee's age, gender, and medical history.

[0049] In this embodiment, a specific example of how the control module analyzes carotid artery plaques is given. Based on three brain characteristics: brain volume data, functional network topology data, and white matter fiber bundle data, a scoring mechanism is used to qualitatively analyze the properties of carotid artery plaques. The higher the score, the greater the probability of rupture and the stronger its warning nature. This method is simple and clear, and utilizes theoretical research on the correlation between plaque rupture and subtle changes in the brain, with extremely strong correlation and accuracy.

[0050] More preferably, the carotid artery plaque early warning device further includes an output module, connected to the control module, for outputting analysis results. Specifically, the output module may optionally output analysis results through, but is not limited to, sound, light, color, or text, such as a buzzer alarm, a flashing red LED light, a red display screen displaying warning plaque text, or other information to indicate the analysis results.

[0051] In this embodiment, the carotid artery plaque early warning device of the present invention is additionally provided with an output module, which can output the determination results in various forms for the examiner and the doctor to make a decision.

[0052] More preferably, the integral of the carotid artery plaque may be calculated using, but not limited to, formula (1):

[0053] P=K1*H1+ K2*H2+ K3*H3 (1)

[0054] Among them, P represents the integral of carotid artery plaque; K1, K2, and K3 represent the weights of brain volume data, functional network topology data, and white matter fiber bundle data, respectively; H1, H2, and H3 represent the individual integrals of brain volume data, functional network topology data, and white matter fiber bundle data, respectively.

[0055] In this embodiment, a preferred embodiment of the present invention for calculating the carotid plaque score is given. Based on theoretical research, three brain characteristics, namely brain volume data, functional network topology data, and white matter fiber bundle data, are used as influencing factors for judging carotid plaques, which can systematically and comprehensively qualitatively analyze the future trend of carotid plaques; and a weight coefficient is assigned to each influencing factor, and corresponding weights can be assigned according to the influence of each factor, which can further improve the precision and accuracy of carotid plaque analysis.

[0056] Preferably, for the individual integrals of brain volume data, the following settings may be selected, but are not limited to: if the whole brain volume is below a set threshold, a score of 1 is assigned, indicating a higher probability of a warning plaque (dangerous plaque); if the gray matter volume of the right middle occipital gyrus exceeds a set threshold, a score of 1 is assigned, indicating a higher probability of a vulnerable plaque; if the gray matter volume of the bilateral lingual gyri exceeds a set threshold, a score of 1 is assigned, indicating a higher probability of a vulnerable plaque. It is worth noting that the above-mentioned whole-brain or local volume parameters are merely examples of adaptability, and individual integrals may also be calculated based on, but are not limited to, other parameters, such as white matter volume, gray matter volume, and cerebrospinal fluid volume of any brain region or the whole brain, or even the correlation between white matter volume and gray matter volume.

[0057] Preferably, for the individual integrals of the functional network topology, the following settings may be selected but are not limited to: if the centrality of the left middle frontal gyrus is higher than a set threshold, a score of 1 is assigned, indicating a higher probability of a warning plaque (dangerous plaque); if the central betweenness of the right middle temporal gyrus is lower than a set threshold, a score of 1 is assigned, indicating a higher probability of a warning plaque (dangerous plaque); if the efficacy of the left middle frontal gyrus and the right inferior parietal angular gyrus is higher than a set threshold, a score of 1 is assigned, indicating a higher probability of a warning plaque (dangerous plaque); if the efficacy of the right middle temporal gyrus is lower than a set threshold, a score of 1 is assigned, indicating a higher probability of a warning plaque (dangerous plaque). It is worth noting that, similar to the individual integrals of the brain volume data described above, the above parameters are merely examples of adaptability.

[0058] Preferably, for the single integral of white matter fiber tract data, the following settings are optional but not limited to: If the RD values ​​of the bilateral anterior thalamic radiations, inferior fronto-occipital fasciculus, frontal portion of the corpus callosum radiation, and uncinate fasciculus, as well as the left corticospinal tract and cingulate gyrus, are greater than a set threshold, a score of 1 is assigned, indicating a higher probability of a warning plaque (dangerous plaque). Preferably, if the RD value of the main component, the frontal portion of the corpus callosum radiation, is greater than a set threshold, a score of 2 is assigned. This is because the corpus callosum is composed of cortico-cortical connecting fibers between the two cerebral hemispheres, most of which pass through deep white matter when connecting to the contralateral hemisphere, serving as a critical pathway for cognitive function between the two hemispheres. Therefore, microstructural disruption of the genu of the corpus callosum can lead to impairments in connectivity with the frontal lobe and other cognitive processes involved in attention and executive function. The inferior fronto-occipital fasciculus connects the frontal lobe with the occipital lobe and posterior temporal lobe, participating in information transmission between the lobes. It is worth noting that, similar to the single integral of brain volume data, the above parameters are merely examples of adaptation.

[0059] This example provides specific implementations for integrating individual brain volume data, functional network topology data, and white matter fiber tract data. Based on theoretical research findings, these data are converted into specific integral forms to construct a carotid plaque analysis device. This approach, applying theoretical insights to practical applications, facilitates early screening and differentiation of patients with similar intima-media thickness but varying stroke risks associated with carotid plaques. This approach explores the development and progression of carotid plaques from a neuromodulatory perspective, analyzes and predicts the rupture risk of different carotid plaque types, and provides guidance for early clinical screening of patients at increased risk of stroke, which is crucial for stroke prevention.

[0060] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0061] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the patent for this invention shall be determined by the appended claims.

Claims

1. A carotid artery plaque analysis device based on magnetic resonance imaging, characterized in that: include: an acquisition module, for acquiring a magnetic resonance image; a first extraction module, connected to the acquisition module, for extracting brain volume data based on the magnetic resonance image; a second extraction module, connected to the acquisition module, for extracting functional network topology data based on the magnetic resonance image; a third extraction module, connected to the acquisition module, for extracting white matter fiber bundle data based on the magnetic resonance image; a control module connected to the first extraction module, the second extraction module, and the third extraction module, for analyzing the properties of the carotid artery plaque based on the brain volume data, the functional network topology data, and the white matter fiber bundle data; An integral calculation unit, used to calculate the integral of carotid artery plaque based on brain volume data, functional network topology data, and white matter fiber bundle data; The qualitative analysis unit is connected to the integral calculation unit and is used to determine whether the integral of the carotid artery plaque exceeds a set threshold. If so, it is qualitatively classified as a warning plaque; If it does not exceed the limit, it is characterized as a safe patch; The integral of carotid artery plaque is calculated using formula (1): P=K1*H1+ K2*H2+ K3*H3 (1) Among them, P represents the integral of carotid artery plaque; K1, K2, and K3 represent the weights of brain volume data, functional network topology data, and white matter fiber bundle data, respectively; H1, H2, and H3 represent the individual integrals of brain volume data, functional network topology data, and white matter fiber bundle data, respectively. A weight coefficient is assigned to each influencing factor, and the corresponding weight is assigned according to the influence of each factor.

2. The carotid artery plaque analysis device according to claim 1, characterized in that: The first extraction module includes: a first segmentation unit, configured to segment the magnetic resonance image to obtain a white matter segmentation image, a gray matter segmentation image, and a cerebrospinal fluid segmentation image; The first calculation unit is used to calculate brain volume data according to the white matter segmentation image, the gray matter segmentation image and the cerebrospinal fluid segmentation image.

3. The carotid artery plaque analysis device according to claim 1, characterized in that: The second extraction module includes: The second segmentation unit is used to segment the magnetic resonance image to obtain several brain regions; The second computing unit regards each brain region as a node, which is used to determine the functional connection between two nodes and construct functional network topology data.

4. The carotid artery plaque analysis device according to claim 3, characterized in that: The second computing unit includes: The correlation coefficient calculation unit is used to calculate the Pearson correlation coefficient between several brain regions; The second judgment unit is used to determine whether the Pearson correlation coefficient between two brain regions exceeds a set threshold. If so, it is determined that there is a connection between the two brain regions; if not, it is determined that there is no connection between the two brain regions; The network construction unit is used to construct functional network topology data based on the determination of whether there is a connection between two brain regions.

5. The carotid artery plaque analysis device according to claim 1, characterized in that: The third extraction module includes: The brain region extraction unit is used to remove excess scalp before calculating the tensor and obtain a mask image to determine the tensor calculation range; The third calculation unit calculates tensor parameters according to the tensor calculation range for analyzing white matter fiber bundle data.

6. The carotid artery plaque analysis device according to claim 1, characterized in that: The individual integration of brain volume data was set as follows: if the whole brain volume was below the set threshold, 1 point was given; if the gray matter volume of the right middle occipital gyrus exceeded the set threshold, 1 point was given; If the gray matter volume of the bilateral lingual gyrus exceeds the set threshold, a score of 1 is scored.

7. The carotid artery plaque analysis device according to claim 6, characterized in that: The individual integrals of the functional network topology were set as follows: if the centrality of the left middle frontal gyrus was higher than the set threshold, 1 point was scored; if the central betweenness of the right middle temporal gyrus was lower than the set threshold, 1 point was scored; if the efficacy of the left middle frontal gyrus and the right inferior parietal angular gyrus was higher than the set threshold, 1 point was scored; if the efficacy of the right middle temporal gyrus was lower than the set threshold, 1 point was scored.

8. The carotid artery plaque analysis device according to claim 7, characterized in that: The single-item integration of white matter fiber bundle data was set as follows: if the RD values ​​of the bilateral anterior thalamic radiations and inferior fronto-occipital fasciculi were greater than the set threshold, 1 point was given; if the RD values ​​of the frontal part of the corpus callosum radiation and the uncinate fasciculus were greater than the set threshold, 2 points were given; if the RD values ​​of the left corticospinal tract and cingulate were greater than the set threshold, 1 point was given.

Images