Brain metabolism network epicenter detection method and device

Abstract

The invention provides a brain metabolism network epicenter detection method and device. The method comprises the following steps: establishing a healthy brain network based on MRI brain imaging data of a healthy subject, wherein the healthy brain network comprises a structural network and a functional network which have the same brain network nodes; calculating a brain metabolism damage degree value L of each brain network node based on FDG-PET brain image data of a disease patient and a healthy subject; calculating probabilities P1 and P2 that each brain network node may be a brain metabolism network epicenter based on a healthy brain structure network and a functional network; comprehensively ranking the brain network nodes according to the sizes of L, P1 and P2, wherein several brain network nodes ranked at the front are brain metabolism network epicenter. The method can accurately detect the epicenter of the brain metabolism network, and has good robustness.

Description

Technical Field

[0001] The present invention belongs to the field of brain network technology, and specifically relates to a method and device for detecting the epicenter of a brain metabolic network. Background Art

[0002] The human brain is composed of tens of billions of neurons, and the network formed by the interconnections between them is the basis for realizing complex brain functional activities. Studies have shown that the onset and deterioration of major brain diseases are related to the destruction of brain networks. Connectome structure plays an important role in the spread of diseases, especially in the trans-neuronal propagation of misfolded proteins. Studying the epicenter of brain networks (the core starting point of functional network damage) is of great significance in understanding the course of neurodegenerative diseases. Neurodegenerative diseases are not limited to specific brain regions, but extend to a wider range of functional neural networks. Understanding the structural and functional characteristics of these networks can help identify disease transmission pathways and potential intervention points.

[0003] The starting core brain region of major brain diseases, i.e., the epicenter of the brain network, plays a very important role in determining the pathogenesis and proposing new treatment plans. At present, brain network research generally uses inter-group statistical comparative analysis to find out which network connections have significant changes in the patient group compared with normal people. However, this method makes it difficult to effectively detect the epicenter of the brain network.

[0004] In view of this, the present invention proposes a method and device for detecting the epicenter of a brain metabolic network. Summary of the invention

[0005] In order to solve the above problems existing in the prior art, the present invention provides a method and device for detecting the epicenter of a brain metabolic network, which can obtain the brain function damage area most relevant to brain diseases and the accurate position of the epicenter of the brain metabolic network.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions.

[0007] In a first aspect, the present invention provides a method for detecting the epicenter of a brain metabolic network, comprising the following steps:

[0008] Establish a healthy brain network based on MRI brain imaging data of healthy subjects, including structural networks and functional networks with the same brain network nodes;

[0009] Based on the FDG-PET brain imaging data of disease patients and healthy subjects, the degree of brain metabolic damage L of each brain network node was calculated;

[0010] Based on the healthy brain structural network and functional network, the probability P1 and P2 of each brain network node being the epicenter of the brain metabolic network are calculated respectively;

[0011] The brain network nodes are comprehensively ranked according to the size of L, P1, and P2, and the top few brain network nodes are the epicenters of the brain metabolic network.

[0012] Furthermore, the method for establishing a healthy brain structure network includes:

[0013] S1. Obtain MRI brain imaging data of healthy subjects, including diffusion tensor imaging DTI and functional brain imaging fMRI;

[0014] S2, converting MRI brain imaging data into NIFTI format data;

[0015] S3, using the diffusion tensor model to perform tensor fitting on the DTI data, and using the least squares method to obtain the optimal solutions of the diffusion tensor matrix characteristic roots λ1, λ2, λ3 and their corresponding eigenvectors e1, e2, e3, where λ1≥λ2≥λ3;

[0016] S4. Calculate the diffusion anisotropy fraction FA of each pixel point. The formula is:

[0017]

[0018] In the formula, MT=(λ1+λ2+λ3) / 3;

[0019] S5. Use the FACT algorithm to perform fiber tracking on the DTI image to obtain the whole-brain white matter fiber map of each healthy subject: take each pixel in the DTI image as the starting seed point, take the direction of the main eigenvector e1 as the tracking direction, and use the directions of the second and third eigenvectors e2 and e3 to correct the tracking direction; when the FA of the pixel is less than 0.2, stop tracking the fiber bundle; and then use the next pixel as the starting seed point to perform fiber tracking until all pixels are traversed;

[0020] S6. According to the brain area division of the human brain function map, the whole brain is divided into multiple brain network nodes, and the fiber bundle connection density between two brain network nodes is used as the value SC of the edge connecting the two brain network nodes to obtain the healthy brain structure network; the calculation formula of SC is:

[0021]

[0022] Where N is the number of healthy subjects, N fib is the number of healthy subjects whose fiber bundles between two brain network nodes are greater than 0, fib i is the number of fiber bundles between two nodes of the i-th healthy subject, i=1,2,...,N.

[0023] Furthermore, the methods for establishing a healthy brain function network include:

[0024] Convert fMRI data into 4-dimensional NIFTI format data including time information;

[0025] Based on the fMRI signals of brain network nodes at multiple consecutive time points, the Pearson correlation coefficient between two brain network nodes is calculated, and the correlation coefficient is used as the value FC of the edge connecting the two brain network nodes to obtain a healthy brain function network.

[0026] Furthermore, the method for establishing a healthy brain functional network also includes: preprocessing the fMRI images of healthy subjects, including time correction, head motion correction, spatial normalization, Gaussian smoothing, white matter signal regression, cerebrospinal fluid signal regression, Friston24 parameter head motion regression, and low-frequency filtering.

[0027] Furthermore, the method for establishing a healthy brain function network also includes sparse processing of the healthy brain function network: if SC ij =0, then set FC ij =0, i and j are the serial numbers of the two brain network nodes.

[0028] Furthermore, the calculation method of the degree of brain metabolic damage L includes:

[0029] The FDG-PET brain imaging data of the patient group and the healthy group were obtained and converted into NIFTI format data; the FDG-PET brain images were standardized to the MNI space using the DARTEL algorithm, and the Gaussian kernel was used to perform Gaussian smoothing on the spatially standardized data;

[0030] Divide the pixel value of each pixel point by a reference value to obtain a standardized uptake score SUVR, and save a SUVR image with SUVR as the pixel value, wherein the reference value is equal to the average pixel value of the brain area not affected by brain diseases;

[0031] Calculate the brain metabolism damage value l of the i-th pixel i , the formula is:

[0032]

[0033]

[0034] Where, t i is the t value of the two-sample t test for the i-th pixel, are the mean and variance of the SUVR value of the i-th pixel of the n1 member of the patient group, are the mean and variance of the SUVR value of the i-th pixel of the n2 members in the healthy group;

[0035] The average value of the degree of brain metabolic damage of the pixels in the area corresponding to each brain network node is calculated to obtain the degree of brain metabolic damage L of each brain network node.

[0036] Furthermore, the calculation method of P1 includes:

[0037] Calculate the P1 intermediate probability value A of the i-th brain network node i , the formula is:

[0038]

[0039] Where, L ij is the brain metabolic damage degree value of the jth neighbor node of the ith brain network node, SC ij is the value of the edge connecting the i-th brain network node and its j-th neighbor node in the healthy brain structure network, j = 1, 2, ..., N i , N i is the number of neighbor nodes;

[0040] Calculate P1 using the formula:

[0041]

[0042] Where P 1i is the value of P1 of the ith brain network node, A ij is the P1 intermediate probability of the jth neighbor node of the i-th brain network node.

[0043] Furthermore, the calculation method of P2 includes:

[0044] Calculate the P2 intermediate probability value B of the i-th brain network node i , the formula is:

[0045]

[0046] In the formula, FC ij The value of the edge connecting the i-th brain network node and its j-th neighbor node in the healthy brain function network;

[0047] Calculate P2 using the formula:

[0048]

[0049] Where P 2i is the value of P2 of the ith brain network node, B ij is the P2 median probability of the jth neighbor node of the i-th brain network node.

[0050] Furthermore, the method also includes evaluating the influence range of the epicenter of the brain metabolic network, including:

[0051] The SUVR images of the patient group and the healthy group were merged in four dimensions including the subject information to obtain the PET image subject sequence of each group;

[0052] Taking the brain network nodes as nodes of the PET network, based on the standardized uptake scores SUVR of all subjects at the nodes, the Pearson correlation coefficient between the two nodes is calculated, and the correlation coefficient is used as the value Pet of the edge connecting the two nodes of the PET network;

[0053] Using healthy brain structural networks to sparsely process PET networks: If SC ij =0, then set Pet ij =0;

[0054] Calculate the Pet for each member of the patient group ij Pet with each member of the health group ij The difference between the two groups was statistically significant using the permutation test.

[0055] The edges corresponding to the differences with significant differences and connected to the brain metabolic network epicenter node are determined, and the area formed by the nodes included in the edges is the influence range of the brain metabolic network epicenter.

[0056] In a second aspect, the present invention provides a brain metabolic network epicenter detection device, comprising:

[0057] A network building module, used to build a healthy brain network based on MRI brain imaging data of healthy subjects, including a structural network and a functional network with the same brain network nodes;

[0058] An L value calculation module is used to calculate the brain metabolism damage degree value L of each brain network node based on the FDG-PET brain imaging data of disease patients and healthy subjects;

[0059] A probability calculation module, used to calculate the probability P1 and P2 that each brain network node may be the epicenter of the brain metabolic network based on the healthy brain structural network and functional network respectively;

[0060] The epicenter determination module is used to comprehensively rank the brain network nodes according to the size of L, P1, and P2. The top few brain network nodes are the epicenters of the brain metabolic network.

[0061] Compared with the prior art, the present invention has the following beneficial effects.

[0062] The present invention establishes a healthy brain network based on MRI brain imaging data of healthy subjects, calculates the brain metabolism impairment value L of each brain network node based on FDG-PET brain imaging data of disease patients and healthy subjects, calculates the probability P1 and P2 that each brain network node may be the epicenter of the brain metabolism network based on the healthy brain structure network and functional network, and comprehensively ranks the brain network nodes according to the size of L, P1, and P2. The top several brain network nodes are the epicenters of the brain metabolism network, and the automatic detection of the epicenter of the brain metabolism network is realized. The present invention can accurately detect the epicenter of the brain metabolism network and has good robustness. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 The present invention is a flowchart of a metabolic network epicenter detection method according to an embodiment of the present invention.

[0064] Figure 2 Distribution diagrams of the probabilities P1 and P2 that a brain network node may be the epicenter of the brain metabolic network for the healthy brain structural network (A) and functional network (B).

[0065] Figure 3 The present invention is a block diagram of a metabolic network epicenter detection device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0066] In order to make the purpose, technical solution and advantages of the present invention clearer and more understandable, the present invention is further described below in conjunction with the accompanying drawings and specific implementation methods. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0067] Figure 1 This is a flow chart of a metabolic network epicenter detection method according to an embodiment of the present invention, comprising the following steps:

[0068] Step 101, establishing a healthy brain network based on MRI brain imaging data of healthy subjects, including a structural network and a functional network with the same brain network nodes;

[0069] Step 102, calculating the brain metabolic impairment level value L of each brain network node based on the FDG-PET brain imaging data of the patient group and the healthy group;

[0070] Step 103, respectively calculating the probability P1 and P2 that each brain network node may be the epicenter of the brain metabolic network based on the healthy brain structural network and functional network;

[0071] Step 104, comprehensively rank the brain network nodes according to the size of L, P1, and P2, and the top few brain network nodes are the epicenters of the brain metabolic network.

[0072] In this embodiment, step 101 is mainly used to establish a healthy brain network. This embodiment establishes a healthy brain network based on MRI brain imaging data of healthy subjects. The healthy brain network includes a structural network and a functional network, both of which are composed of nodes and edges connecting nodes, and the value of the edge represents the association relationship between the two nodes connected by the edge. The two networks use the same brain network nodes, and both use AAL90 brain atlas as the nodes of the brain structure network. The following embodiments will respectively give the specific establishment methods of these two networks.

[0073] In this embodiment, step 102 is mainly used to calculate the brain metabolism impairment value L of each brain network node. The larger the brain metabolism impairment value L of the brain network node, the greater the possibility that the node is the epicenter of the metabolic network, so the brain network node belonging to the epicenter of the metabolic network can be determined according to the size of L. This embodiment is based on the FDG-PET brain imaging data of the patient group and the healthy group of subjects to calculate the brain metabolism impairment value of each brain network node. The members of the patient group are generally Alzheimer's disease (AD) patients. Compared with traditional structural imaging techniques, PET (positron emission tomography) imaging has unique advantages and can provide information on brain function and metabolism. By using specific radioactive tracers, PET can detect early metabolic abnormalities in neurodegenerative diseases, even before any obvious structural changes occur. For example, in Alzheimer's disease, PET can identify the presence of reduced FDG (glucose metabolism) uptake, amyloid plaques and pathological tau proteins in the preclinical stage of the disease, which is conducive to early diagnosis and disease monitoring.

[0074] In this embodiment, step 103 is mainly used to calculate the probability that each brain network node may be the epicenter of the brain metabolic network. Based on the healthy brain structural network and functional network, the probability P1 and P2 of each brain network node being the epicenter of the brain metabolic network are calculated respectively. Obviously, the larger the value of P1 and P2, the greater the possibility that the corresponding brain network node may be the epicenter of the brain metabolic network. Therefore, the node belonging to the epicenter of the brain metabolic network can be determined according to the size of P1 and P2.

[0075] In this embodiment, step 104 is mainly used to determine the location of the epicenter of the brain metabolic network. According to the above, the size of L, P1, and P2 of the brain network node is closely related to whether the node is the epicenter of the brain metabolic network. In principle, the approximate location of the epicenter of the brain metabolic network can be estimated according to any one of the parameters. In order to improve the positioning accuracy, this embodiment comprehensively ranks the brain network nodes according to the size of L, P1, and P2, and selects the first few brain network nodes (such as the first 3 to 5) as the epicenter of the brain metabolic network. There are many methods for comprehensive ranking. For example, L, P1, and P2 can be weighted and summed, and ranked according to the size of the sum; they can also be ranked and screened according to the importance of the three parameters; they can also be ranked separately, and the top few are taken, and then the intersection is calculated. This embodiment does not limit the specific method of comprehensive ranking.

[0076] As an optional embodiment, the method for establishing a healthy brain structure network includes:

[0077] S1. Obtain MRI brain imaging data of healthy subjects, including diffusion tensor imaging DTI and functional brain imaging fMRI;

[0078] S2, converting MRI brain imaging data into NIFTI format data;

[0079] S3, using the diffusion tensor model to perform tensor fitting on the DTI data, and using the least squares method to obtain the optimal solutions of the diffusion tensor matrix characteristic roots λ1, λ2, λ3 and their corresponding eigenvectors e1, e2, e3, where λ1≥λ2≥λ3;

[0080] S4. Calculate the diffusion anisotropy fraction FA of each pixel point. The formula is:

[0081]

[0082] In the formula, MT=(λ1+λ2+λ3) / 3;

[0083] S5. Use the FACT algorithm to perform fiber tracking on the DTI image to obtain the whole-brain white matter fiber map of each healthy subject: take each pixel in the DTI image as the starting seed point, take the direction of the main eigenvector e1 as the tracking direction, and use the directions of the second and third eigenvectors e2 and e3 to correct the tracking direction; when the FA of the pixel is less than 0.2, stop tracking the fiber bundle; and then use the next pixel as the starting seed point to perform fiber tracking until all pixels are traversed;

[0084] S6. According to the brain area division of the human brain function map, the whole brain is divided into multiple brain network nodes, and the fiber bundle connection density between two brain network nodes is used as the value SC of the edge connecting the two brain network nodes to obtain the healthy brain structure network; the calculation formula of SC is:

[0085]

[0086] Where N is the number of healthy subjects, N fib is the number of healthy subjects whose fiber bundles between two brain network nodes are greater than 0, fib i is the number of fiber bundles between two nodes of the i-th healthy subject, i=1,2,...,N.

[0087] This embodiment provides a technical solution for establishing a healthy brain structure network. Establishing a healthy brain structure network mainly includes the division of brain network nodes and the calculation of edge values, among which the edge value calculation is the top priority. This embodiment calculates the edge value according to the number or density of fibers between two nodes connected by each edge by establishing a whole-brain white matter fiber map of healthy subjects.

[0088] Steps S1 and S2 are used for data acquisition and data format conversion. First, the MRI brain imaging data of healthy subjects are obtained, including diffusion tensor imaging DTI and brain functional imaging fMRI; then the MRI image (DICOM data) is converted into processable NIFTI format data. Since a healthy brain structure network is to be established, only MRI images of healthy subjects are required. Diffusion tensor imaging DTI is currently the only imaging technology that can perform non-invasive detection of white matter fiber structure in the living human brain. It can display the white matter fiber bundles of the living brain in three dimensions, and can display the changes in the course of the fiber bundles in a three-dimensional and intuitive manner, thereby achieving fine imaging of human nerve fibers.

[0089] Steps S3 to S5 are used to generate a whole-brain white matter fiber map for each healthy subject. Step S3 calculates the three eigenvalues ​​of the diffusion tensor matrix and their corresponding eigenvalues. Step S4 calculates the diffusion anisotropy fraction FA of each pixel based on the three eigenvalues, and the calculation formula is as shown in formula (1). Step S5 performs fiber tracking on the DTI image based on the three eigenvalues ​​and the diffusion anisotropy fraction to obtain a whole-brain white matter fiber map for each healthy subject, specifically, to determine all the pixels connected to each fiber bundle. In actual calculations, fiber bundles with higher resolution than the DTI image are obtained through image interpolation.

[0090] Step S6 is used for brain network node division and edge value calculation. This embodiment divides the whole brain into multiple brain network nodes according to the brain region division of the human brain function map (such as the AAL90 brain map), and uses the fiber bundle connection density between two nodes as the edge value SC connecting the two nodes. The calculation formula of SC is as shown in formula (2). Formula (2) is a two-piece function with SC as the dependent variable and the number of fiber bundle connection roots fib between two nodes as the independent variable. When the number of fiber roots between two nodes is greater than 0, the number of healthy subjects N fibLess than half of the number of healthy subjects N (i.e. N fib <0.5N), the value of SC is 0, which is equivalent to no edge connection between the two nodes; when N fib When ≥0.5N, the value of SC is equal to the mean number of fiber bundles fib between two nodes of all healthy subjects.

[0091] As an optional embodiment, the method for establishing a healthy brain function network includes:

[0092] Convert fMRI data into 4-dimensional NIFTI format data including time information;

[0093] Based on the fMRI signals of brain network nodes at multiple consecutive time points, the Pearson correlation coefficient between two brain network nodes is calculated, and the correlation coefficient is used as the value FC of the edge connecting the two brain network nodes to obtain a healthy brain function network.

[0094] This embodiment provides a technical solution for establishing a healthy brain function network. The healthy brain function network uses the same brain network nodes as the structural network, but the difference is that the edge value FC is calculated based on the brain functional imaging fMRI data. First, the fMRI data is converted into 4-dimensional (3-dimensional + time dimension) NIFTI format data that can be processed and includes time information; then the Pearson correlation coefficient of the two brain network nodes is calculated, that is, the Pearson correlation coefficient of the fMRI signals of the two nodes at multiple consecutive time points is calculated, and the correlation coefficient is the edge value FC connecting the two nodes.

[0095] As an optional embodiment, the method for establishing a healthy brain functional network also includes: preprocessing the fMRI images of healthy subjects, including time correction, head motion correction, spatial normalization, Gaussian smoothing, white matter signal regression, cerebrospinal fluid signal regression, Friston24 parameter head motion regression, and low-frequency filtering.

[0096] This embodiment provides a technical solution for preprocessing fMRI images. In order to obtain high-quality fMRI image data, thereby improving the accuracy of healthy brain functional networks, fMRI images must be preprocessed before edge value calculation. The preprocessing content includes time correction, head motion correction, Gaussian smoothing, low-frequency filtering, etc.

[0097] As an optional embodiment, the method for establishing a healthy brain function network further includes performing sparse processing on the healthy brain function network: if SC ij =0, then set FC ij =0, i and j are the serial numbers of the two brain network nodes.

[0098] This embodiment provides a technical solution for sparse processing of healthy brain functional networks. The so-called sparse processing is to reduce the number of edges of the healthy brain functional network, so as to reduce the amount of calculation and improve the running speed without affecting the accuracy of epicenter detection. This embodiment uses the healthy brain structural network to perform sparse processing on the functional network. The technical principle is: if two nodes are not connected by edges in the structural network (that is, the edge value is 0), make these two nodes also have no edge connection in the functional network, that is, if there is an edge connection originally, delete the edge; if there is no edge connection originally, keep the edgeless connection.

[0099] As an optional embodiment, the calculation method of the degree of brain metabolic damage L includes:

[0100] The FDG-PET brain imaging data of the patient group and the healthy group were obtained and converted into NIFTI format data; the FDG-PET brain images were standardized to the MNI space using the DARTEL algorithm, and the Gaussian kernel was used to perform Gaussian smoothing on the spatially standardized data;

[0101] Divide the pixel value of each pixel point by a reference value to obtain a standardized uptake score SUVR, and save a SUVR image with SUVR as the pixel value, wherein the reference value is equal to the average pixel value of the brain area not affected by brain diseases;

[0102] Calculate the brain metabolism damage value l of the i-th pixel i , the formula is:

[0103]

[0104] Where, t i is the t value of the two-sample t test for the i-th pixel, are the mean and variance of the SUVR value of the i-th pixel of the n1 member of the patient group, are the mean and variance of the SUVR value of the i-th pixel of the n2 members in the healthy group;

[0105] The average value of the degree of brain metabolic damage of the pixels in the area corresponding to each brain network node is calculated to obtain the degree of brain metabolic damage L of each brain network node.

[0106] This embodiment provides a technical solution for calculating the degree of brain metabolic damage L. As mentioned above, PET imaging can provide information on brain function and metabolism. By using specific radioactive tracers, PET can detect early metabolic abnormalities of neurodegenerative diseases. Therefore, this embodiment calculates the degree of brain metabolic damage based on glucose-positron emission tomography imaging FDG-PET.

[0107] First, data acquisition and processing. In this embodiment, the degree of brain metabolic impairment is calculated by comparing patients and healthy people. Therefore, it is necessary to obtain FDG-PET brain image data of patients and healthy subjects, and convert them into NIFTI format data, and then use the DARTEL algorithm to standardize the FDG-PET brain images to MNI space, and use Gaussian kernel to perform Gaussian smoothing on the data after space standardization.

[0108] Then the standardized uptake fraction SUVR of the pixel point is calculated. A brain area not affected by brain disease is selected as the reference area, and the average pixel value (also called PET metabolic value) of the reference area is calculated to obtain the benchmark value; then the pixel value of each pixel point is divided by the benchmark value to obtain the standardized uptake fraction SUVR of each pixel point.

[0109] Finally, the brain metabolism damage value L is calculated. First, the brain metabolism damage value l of each pixel is calculated, and then the brain metabolism damage value L of each node is obtained by calculating the average value of the brain metabolism damage value l of the pixel points in each node area. In this embodiment, the l value of each pixel point is obtained by calculating the t value of the double sample t test for each pixel point. The double sample includes the SUVR value x of each pixel point patient group n1 members i1 and the SUVR value x of n2 members in the healthy group i2 , the t value of each pixel can be obtained according to formula (4). Then, the positive or negative value of t is used to determine whether the patient group has decreased brain metabolism. If t<0, there is decreased brain metabolism, and the absolute value of t (i.e. -t) is the degree of brain metabolic damage l; otherwise, there is no decreased brain metabolism, l = 0, see the piecewise function shown in formula (3).

[0110] As an optional embodiment, the calculation method of P1 includes:

[0111] Calculate the P1 intermediate probability value A of the i-th brain network node i , the formula is:

[0112]

[0113] Where, L ij is the brain metabolic damage degree value of the jth neighbor node of the ith brain network node, SC ij is the value of the edge connecting the i-th brain network node and its j-th neighbor node in the healthy brain structure network, j = 1, 2, ..., N i , N i is the number of neighbor nodes;

[0114] Calculate P1 using the formula:

[0115]

[0116] Where P1i is the value of P1 of the ith brain network node, A ij is the P1 intermediate probability of the jth neighbor node of the i-th brain network node.

[0117] This embodiment provides a technical solution for calculating P1. P1 is the probability that each node may be the epicenter, calculated based on the healthy brain structure network. This embodiment uses the brain metabolism damage degree L of each node's neighboring nodes to weighted sum the edge value SC of the structural network connecting the node and the neighboring nodes to obtain the intermediate probability A for calculating P1, as shown in formula (5); then calculate the mean value of A of each node's neighboring nodes to obtain P1, as shown in formula (6).

[0118] As an optional embodiment, the calculation method of P2 includes:

[0119] Calculate the P2 intermediate probability value B of the i-th brain network node i , the formula is:

[0120]

[0121] In the formula, FC ij The value of the edge connecting the i-th brain network node and its j-th neighbor node in the healthy brain function network;

[0122] Calculate P2 using the formula:

[0123]

[0124] Where P 2i is the value of P2 of the ith brain network node, B ij is the P2 median probability of the jth neighbor node of the i-th brain network node.

[0125] This embodiment provides a technical solution for calculating P2. P2 is the probability that each node may be the epicenter, calculated based on the healthy brain function network. The calculation method of P2 is almost the same as that of P1. It is only necessary to replace SC in the calculation formula of P1 with FC. See formulas (7) and (8) for details.

[0126] Figure 2 The distribution maps of P1 and P2 obtained by actual calculation are given, the upper part A is the distribution map of P1, and the lower part B is the distribution map of P2. This example finally determines that the epicenter of the AD patient is the right angular gyrus, the left posterior cingulate gyrus, the left angular gyrus, the right anterior temporal gyrus, and the right middle temporal gyrus.

[0127] As an optional embodiment, the method further includes evaluating the influence range of the epicenter of the brain metabolic network, including:

[0128] The SUVR images of the patient group and the healthy group were merged in four dimensions including the subject information to obtain the PET image subject sequence of each group;

[0129] Taking the brain network nodes as nodes of the PET network, based on the standardized uptake scores SUVR of all subjects at the nodes, the Pearson correlation coefficient between the two nodes is calculated, and the correlation coefficient is used as the value Pet of the edge connecting the two nodes of the PET network;

[0130] Using healthy brain structural networks to sparsely process PET networks: If SC ij =0, then set Pet ij =0;

[0131] Calculate the Pet for each member of the patient group ij Pet with each member of the health group ij The difference between the two groups was statistically significant using the permutation test.

[0132] The edges corresponding to the differences with significant differences and connected to the brain metabolic network epicenter node are determined, and the area formed by the nodes included in the edges is the influence range of the brain metabolic network epicenter.

[0133] This embodiment provides a technical solution for evaluating the influence range of the epicenter of the brain metabolic network. This embodiment establishes a PET network based on the SUVR images of the patient group and the healthy group, and then calculates the PET network edge value Pet of each member of the patient group ij Pet with each member of the health group ij The difference is calculated and the significance of the difference is statistically tested by permutation test. Finally, the influence range of the epicenter is determined according to the node areas corresponding to the differences with significant differences.

[0134] First, a PET image test sequence is established. In this embodiment, the SUVR images of the patient group and the healthy group are respectively merged in four dimensions including the subject information (dimensions), so as to obtain a PET image test sequence consisting of each group of SUVR images of each subject.

[0135] Then the PET network is established. The nodes of the PET network are still the brain network nodes obtained previously, and the edge values ​​are still obtained by calculating the Pearson correlation coefficient between two nodes. The data used to calculate the correlation coefficient is a sequence composed of the standardized uptake scores SUVR of all subjects at each node. Next, the PET network is sparsely processed using the healthy brain structure network to simplify the PET network.

[0136] Finally, determine the influence range of the epicenter. First, perform a permutation test on the aforementioned differences to obtain the edges and nodes corresponding to the differences with significant differences. The area formed by the nodes connected to the epicenter node and the epicenter node is the influence range of the epicenter.

[0137] Figure 3 This is a schematic diagram of the composition of a brain metabolic network epicenter detection device according to an embodiment of the present invention, the device comprising:

[0138] A network establishment module 11 is used to establish a healthy brain network based on MRI brain imaging data of healthy subjects, including a structural network and a functional network with the same brain network nodes;

[0139] An L value calculation module 12 is used to calculate the brain metabolism damage degree value L of each brain network node based on the FDG-PET brain imaging data of disease patients and healthy subjects;

[0140] A probability calculation module 13, used to calculate the probability P1, P2 that each brain network node may be the epicenter of the brain metabolic network based on the healthy brain structural network and functional network respectively;

[0141] The epicenter determination module 14 is used to comprehensively rank the brain network nodes according to the sizes of L, P1, and P2, and the first few brain network nodes are the epicenters of the brain metabolic network.

[0142] The device of this embodiment can be used to perform Figure 1 The technical solution of the method embodiment shown has similar implementation principles and technical effects, which will not be repeated here.

[0143] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for detecting the epicenter of a brain metabolic network, characterized in that: The following steps are involved: Establish a healthy brain network based on MRI brain imaging data of healthy subjects, including structural networks and functional networks with the same brain network nodes; Based on the FDG-PET brain imaging data of disease patients and healthy subjects, the degree of brain metabolic damage L of each brain network node was calculated; Based on the healthy brain structural network and functional network, the probability P1 and P2 of each brain network node being the epicenter of the brain metabolic network are calculated respectively; The brain network nodes are comprehensively ranked according to the size of L, P1, and P2, and the top few brain network nodes are the epicenters of the brain metabolic network.

2. The brain metabolic network epicenter detection method according to claim 1, characterized in that: Methods for building a healthy brain structure network include: S1. Obtain MRI brain imaging data of healthy subjects, including diffusion tensor imaging DTI and functional brain imaging fMRI; S2, converting MRI brain imaging data into NIFTI format data; S3, using the diffusion tensor model to perform tensor fitting on the DTI data, and using the least squares method to obtain the optimal solutions of the diffusion tensor matrix characteristic roots λ1, λ2, λ3 and their corresponding eigenvectors e1, e2, e3, where λ1≥λ2≥λ3; S4. Calculate the diffusion anisotropy fraction FA of each pixel point. The formula is: In the formula, MT=(λ1+λ2+λ3) / 3; S5. Use the FACT algorithm to perform fiber tracking on the DTI image to obtain the whole-brain white matter fiber map of each healthy subject: take each pixel in the DTI image as the starting seed point, take the direction of the main eigenvector e1 as the tracking direction, and use the directions of the second and third eigenvectors e2 and e3 to correct the tracking direction; when the FA of the pixel is less than 0.2, stop tracking the fiber bundle; and then use the next pixel as the starting seed point to perform fiber tracking until all pixels are traversed; S6. According to the brain area division of the human brain function map, the whole brain is divided into multiple brain network nodes, and the fiber bundle connection density between two brain network nodes is used as the value SC of the edge connecting the two brain network nodes to obtain the healthy brain structure network; the calculation formula of SC is: Where N is the number of healthy subjects, N fib is the number of healthy subjects whose fiber bundles between two brain network nodes are greater than 0, fib i is the number of fiber bundles between two nodes of the i-th healthy subject, i=1,2,...,N.

3. The method for detecting the epicenter of the brain metabolic network according to claim 2, characterized in that: Methods for establishing a healthy brain function network include: Convert fMRI data into 4-dimensional NIFTI format data including time information; Based on the fMRI signals of brain network nodes at multiple consecutive time points, the Pearson correlation coefficient between two brain network nodes is calculated, and the correlation coefficient is used as the value FC of the edge connecting the two brain network nodes to obtain a healthy brain function network.

4. The method for detecting the epicenter of the brain metabolic network according to claim 3, characterized in that: The method for establishing a healthy brain functional network also includes: preprocessing the fMRI images of healthy subjects, including time correction, head motion correction, spatial normalization, Gaussian smoothing, white matter signal regression, cerebrospinal fluid signal regression, Friston24 parameter head motion regression, and low-frequency filtering.

5. The method for detecting the epicenter of brain metabolic network according to claim 3, characterized in that: The method for establishing a healthy brain function network also includes sparse processing of the brain function healthy brain function network: if SC ij =0, then set FC ij =0, i and j are the serial numbers of the two brain network nodes.

6. The method for detecting the epicenter of brain metabolic network according to claim 3, characterized in that: The calculation method of the degree of brain metabolic damage L includes: The FDG-PET brain imaging data of the patient group and the healthy group were obtained and converted into NIFTI format data; the FDG-PET brain images were standardized to the MNI space using the DARTEL algorithm, and the Gaussian kernel was used to perform Gaussian smoothing on the spatially standardized data; The pixel value of each pixel point is divided by a reference value to obtain a standardized uptake score SUVR, and the result is saved as a SUVR image with SUVR as the pixel value, wherein the reference value is equal to the average pixel value of the brain area not affected by brain diseases; Calculate the brain metabolism damage value l of the i-th pixel i , the formula is: Where, t i is the t value of the two-sample t test for the i-th pixel, are the mean and variance of the SUVR value of the i-th pixel of the n1 member of the patient group, are the mean and variance of the SUVR value of the i-th pixel of the n2 members in the healthy group; The average value of the degree of brain metabolic damage of the pixels in the area corresponding to each brain network node is calculated to obtain the degree of brain metabolic damage L of each brain network node.

7. The method for detecting the epicenter of brain metabolic network according to claim 6, characterized in that: The calculation method of P1 includes: Calculate the P1 intermediate probability value A of the i-th brain network node i , the formula is: Where, L ij is the brain metabolic damage degree value of the jth neighbor node of the ith brain network node, SC ij is the value of the edge connecting the i-th brain network node and its j-th neighbor node in the healthy brain structure network, j = 1, 2, ..., N i , N i is the number of neighbor nodes; Calculate P1 using the formula: Where P 1i is the value of P1 of the ith brain network node, A ij is the P1 intermediate probability of the jth neighbor node of the i-th brain network node.

8. The method for detecting the epicenter of the brain metabolic network according to claim 7, characterized in that: The calculation method of P2 includes: Calculate the P2 intermediate probability value B of the i-th brain network node i , the formula is: In the formula, FC ij The value of the edge connecting the i-th brain network node and its j-th neighbor node in the healthy brain function network; Calculate P2 using the formula: Where P 2i is the value of P2 of the ith brain network node, B ij is the P2 median probability of the jth neighbor node of the i-th brain network node.

9. The method for detecting the epicenter of brain metabolic network according to claim 6, characterized in that: The method also includes assessing the scope of influence of the epicenter of the brain metabolic network, including: The SUVR images of the patient group and the healthy group were merged in four dimensions including the subject information to obtain the PET image subject sequence of each group; Taking the brain network nodes as the nodes of the PET network, based on the standardized uptake scores SUVR of all subjects at the nodes, the Pearson correlation coefficient between the two nodes is calculated, and the correlation coefficient is used as the value Pet of the edge connecting the two nodes of the PET network; Using healthy brain structural networks to sparsely process PET networks: If SC ij =0, then set Pet ij =0; Calculate the Pet for each member of the patient group ij Pet with each member of the health group ij The difference between the two groups was statistically significant using the permutation test. The edges corresponding to the differences with significant differences and connected to the brain metabolic network epicenter node are determined, and the area formed by the nodes included in the edges is the influence range of the brain metabolic network epicenter.

10. A brain metabolic network epicenter detection device, characterized in that: include: A network building module, used to build a healthy brain network based on MRI brain imaging data of healthy subjects, including a structural network and a functional network with the same brain network nodes; An L value calculation module is used to calculate the brain metabolism damage degree value L of each brain network node based on the FDG-PET brain imaging data of disease patients and healthy subjects; A probability calculation module, used to calculate the probability P1 and P2 that each brain network node may be the epicenter of the brain metabolic network based on the healthy brain structural network and functional network respectively; The epicenter determination module is used to comprehensively rank the brain network nodes according to the size of L, P1, and P2. The top few brain network nodes are the epicenters of the brain metabolic network.

Images