Brain function mapping system

Abstract

The present invention discloses a brain function map drawing system. The drawing system is implemented by a local computer device and / or a cloud computing platform. The drawing system includes: an initialization module for initializing an individual's brain function map using a brain function map template to obtain an initial individual brain function map, which divides the brain into multiple functional areas; a preprocessing module for dividing the initial individual brain function map into several large areas, each of which includes several functional areas; an iterative processing module for entering an iteration, each iteration process including: sequentially calculating the connectivity of each voxel in each large area with each functional area in the large area, adjusting each voxel to the functional area with the highest connectivity with the voxel, until all voxels are adjusted; and terminating the iteration when a termination condition is met to obtain a final individual brain function map.

Description

[0001] This application is a divisional application of a patent application with an application date of December 2, 2019, application number 201911214999.9, and invention name “Method and system for drawing brain function maps”. Technical Field

[0002] The present invention relates to the technical field of medical image processing, and in particular to a system for drawing brain function maps. Background Art

[0003] In the early 19th century, German neuroanatomist Korbinian Brodmann first mapped the human brain (Brodmann brain atlas), noting that different brain regions are responsible for different functions. Since then, brain mapping has been a key area of ​​research in brain science. Understanding the divisions of the human brain, the boundaries between regions, and the connections between these regions are of vital importance to both basic and clinical research in brain science.

[0004] For a long time, due to a lack of understanding of the human brain and technological limitations, brain mapping could only be done through a "group" approach. This involves studying the brains of a group of people and performing statistical analysis to create an "average" map. Group brain mapping can reveal many commonalities and patterns across human populations, making it highly valuable for scientific research. Currently, the most representative brain map in the United States is the group mapping approach, published in Nature in 2016 by Glasser et al. (Glasser et al., Nature, A multi-modal parcellation of human cerebral cortex, 2016).

[0005] However, in clinical practice, doctors focus on individual patients and require "individual" brain maps. In reality, human brains vary greatly; no two brains are exactly alike (Mueller et al., Neuron, Individual Variability in Functional Connectivity Architecture of the Human Brain, 2013). "Group" brain maps obliterate individual uniqueness, making them clinically valuable and even misleading. Therefore, mapping "individual" brains is a much-needed but unresolved problem.

[0006] Patent US 9,662,039 B2 (Liu et al.) describes a method for mapping "individual" brains. This method successfully mapped 18 functional networks, a significant advance. However, when applied to higher-precision brain maps (for example, 112 networks), the method suffers from stability issues and poor noise immunity, resulting in unreliable and noisy results. To our knowledge, there is currently no proven solution for mapping "individual" brains. Summary of the Invention

[0007] The main technical problem addressed by the present invention is to provide a method for mapping brain function, overcoming the low stability, weak noise immunity, and unreliable results of existing high-precision individual brain function maps. This method enables accurate and reliable mapping of an individual's whole-brain functional network, with high stability, high reliability, and low noise at varying levels of precision. Another technical problem addressed by the present invention is to provide a system for mapping brain function.

[0008] In order to solve the above technical problems, the technical solutions adopted by the present invention are as follows:

[0009] A first aspect of the present invention provides a method for drawing a brain function map, comprising: initializing an individual's brain function map using a brain function map template to obtain an initial individual brain function map, wherein the initial individual brain function map divides the individual's brain into multiple functional areas; dividing the initial individual brain function map into several large areas, each large area including several functional areas; entering an iteration process, wherein each iteration process comprises: sequentially calculating the connectivity of each voxel in each large area with each functional area in the large area, adjusting each voxel to the functional area with the highest connectivity with the voxel, until all voxels are adjusted; terminating the iteration when a termination condition is met, and obtaining a final individual brain function map. Furthermore, a corresponding report and / or image can be generated based on the final individual brain function map.

[0010] In a first possible implementation, each iteration process specifically includes: calculating a reference time series signal of each functional area based on the time series signals of all voxels in each functional area; determining an unadjusted voxel as the current voxel; calculating a correlation value between the time series signal of the current voxel and the reference time series signals of each functional area in the current region, and using the correlation value as the connectivity between the current voxel and each functional area in the current region, where the current region is the region to which the current voxel belongs; adjusting the current voxel to the functional area with the highest connectivity with the current voxel; determining whether all voxels have completed an adjustment, and if not, returning to the step of determining an unadjusted voxel as the current voxel; and terminating the iteration process if all voxels have completed an adjustment.

[0011] In combination with the first possible implementation method, in the second possible implementation method, before entering the iteration, it also includes: initializing the credibility of each voxel; after adjusting the current voxel to the functional area with the highest connectivity with the current voxel, it also includes: updating the credibility of the current voxel; calculating the reference time series signal of each functional area, including: for each functional area, calculating the average or median of the time series signals of all voxels in the functional area whose credibility is not lower than a preset threshold as the reference time series signal of the functional area.

[0012] In combination with the second possible implementation method, in a third possible implementation method, updating the credibility of the current voxel includes: selecting the maximum correlation value and the second largest correlation value from the calculated correlation values ​​of the time series signal of the current voxel and the reference time series signals of each functional area in the current region; calculating and updating the credibility of the current voxel, the credibility of the current voxel being equal to the ratio of the maximum correlation value to the second largest correlation value.

[0013] In combination with the first aspect or any one of the first to third possible implementations of the first aspect, in a fourth possible implementation, the iteration is terminated when the termination condition is met, including: terminating the iteration when the number of iterations reaches a preset number or reaches a convergence criterion.

[0014] In combination with the first aspect or any one of the first to fourth possible implementations of the first aspect, in a fifth possible implementation, a group brain function map is preselected or generated as a brain function map template.

[0015] In combination with the first aspect or any one of the first to fifth possible implementations of the first aspect, in a sixth possible implementation, a group brain function map is pre-selected or generated as a brain function map template, and the steps include: dividing the brain into several large regions; calculating a group-level functional connectivity map in each large region, wherein the functional connectivity map is a functional connectivity matrix of voxels in the large region and N regions of interest (ROIs) in the large region, where N is a natural number; preferably, N is a natural number not less than 100; based on the functional connectivity map as a feature, a clustering algorithm is used to divide each large region into multiple fine-grained functional partitions; comprehensively evaluating the indicators of the clustering algorithm and the indicators of maximizing functional homogeneity to determine multiple locally optimal numbers of partitions; using a merging algorithm to merge the fine-grained functional partitions in the large region to create a whole-brain group brain function map as the required brain function map template.

[0016] In combination with the first aspect or any one of the first to fifth possible implementations of the first aspect, in a seventh possible implementation, the initial individual brain function map is divided into several major regions, including: by dividing the left and right cortices of the brain into five major regions: frontal lobe, parietal lobe, occipital lobe, temporal lobe and pan-central sulcus area, the initial individual brain function map is divided into ten major regions.

[0017] A second aspect of the present invention provides a method for drawing a group brain function map, comprising:

[0018] Obtain brain MRI scan data from a group of people;

[0019] Divide the brain into several major regions;

[0020] Calculating a group functional connectivity map in each region, where the group functional connectivity map is a functional connectivity matrix of voxels in the region and N regions of interest in the region, where N is a natural number; preferably, N is a natural number not less than 100;

[0021] Based on the group functional connectivity map as a feature, each large area is divided into multiple fine-grained functional partitions using a clustering algorithm;

[0022] Comprehensively evaluate the clustering algorithm's indicators and the indicator of maximizing functional homogeneity to determine multiple locally optimal numbers of partitions;

[0023] A merging algorithm is used to merge the fine-grained functional partitions within the large region to create a group brain function map of the whole brain.

[0024] In a possible implementation of the second aspect, calculating the group functional connectivity map includes:

[0025] An individual functional connectivity map is calculated in each region. The group functional connectivity map is a functional connectivity matrix of voxels in the region and N regions of interest in the region, where N is a natural number.

[0026] Calculate the average of all individual functional connectivity maps in each region to obtain the group functional connectivity map

[0027] The third aspect of the present invention provides a brain function map drawing system, including: an initialization module, used to initialize the individual brain function map using a brain function map template to obtain an initial individual brain function map, which divides the brain into multiple functional areas; a preprocessing module, used to divide the initial individual brain function map into several large areas, each large area including several functional areas; an iterative processing module, used to enter the iteration, each iteration process including: sequentially calculating the connectivity of each voxel in each large area with each functional area in the large area, adjusting each voxel to the functional area with the highest connectivity with the voxel, until the adjustment of all voxels is completed; when the termination condition is met, the iteration is terminated to obtain the final individual brain function map.

[0028] The fourth aspect of the present invention provides a data processing system, comprising a processor and a memory; the memory is used to store computer-executable instructions, and when the data processing system is running, the processor executes the computer-executable instructions stored in the memory, so that the data processing system performs the method for drawing a brain function map as described in the first aspect or any possible implementation thereof.

[0029] A fifth aspect of the present invention provides a medical image processing system, comprising: a data acquisition system, a data output interaction system, and the data processing system as described in the third aspect;

[0030] The data acquisition system is used to acquire individual magnetic resonance scan data and upload the acquired data to the data processing system;

[0031] The data processing system is configured to execute the method for drawing a brain function map as described in any one of the first aspects to obtain a final individual brain function map, and generate a corresponding report and / or image based on the final individual brain function map;

[0032] The data output interactive system is used to obtain the report and / or image from the data processing system and display the report and / or image;

[0033] In which, the data processing system includes a local server and / or a cloud computing platform; when the data processing system includes both the local server and the cloud computing platform, the local server and the cloud computing platform jointly execute the method for drawing a brain function map described in the first aspect or any possible implementation thereof based on a load balancing strategy and / or a sharing strategy.

[0034] The sixth aspect of the present invention provides a computer-readable storage medium storing one or more programs, wherein the one or more programs include computer execution instructions. When the computer execution instructions are executed by a data processing system including a processor, the data processing system executes the method for drawing a brain function map as described in the first aspect or any possible implementation thereof.

[0035] As can be seen from the above, in some feasible embodiments of the present invention, by adopting the above technical solution, the following beneficial effects are achieved:

[0036] The present invention iteratively calculates an initial individual brain function map, continuously adjusting the functional area to which each voxel belongs through this iterative process. Ultimately, a high-precision individual brain function map can be created, achieving high stability, high reliability, and low noise at various levels of precision or resolution. In practice, the present invention has demonstrated stable and reliable results for partitioning the entire brain into 56, 112, and 213 functional areas, with various precisions, and has been successfully clinically validated. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments.

[0038] Figure 1 The figure is a flowchart of a method for drawing a brain function map provided by one embodiment of the present invention.

[0039] Figure 2 It is a flowchart of each iterative process in one embodiment of the present invention.

[0040] Figure 3 It is a schematic diagram of a process for generating a group brain function map template in one embodiment of the present invention.

[0041] Figure 4 It is a flowchart of a specific application scenario embodiment of the present invention.

[0042] Figure 5 It is a structural diagram of a brain function map drawing system provided by one embodiment of the present invention.

[0043] Figure 6 is a structural diagram of a data processing system provided by one embodiment of the present invention;

[0044] Figure 7-1 to Figure 7-3 is a network architecture diagram of a medical image processing system provided by an embodiment of the present invention;

[0045] Figure 8 This is the clinical validation result of the brain function areas mapped by the present invention and those measured by ECS;

[0046] Figure 9 This is the clinical verification result of the brain function areas mapped by the present invention and the brain function areas measured by long-term sampling task-state MRI;

[0047] Figure 10 is a brain function map drawn according to an embodiment of the present invention;

[0048] Figure 11 The brain function map drawn by an embodiment of the present invention and the brain function map drawn by patent US 9,662,039 B2 are shown. DETAILED DESCRIPTION

[0049] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described 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 efforts should fall within the scope of protection of the present invention.

[0050] The terms "including" and "having" and any variations thereof in the specification and claims of the present invention and the accompanying drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not limited to the listed steps or elements, but may optionally include steps or elements not listed, or may optionally include other steps or elements inherent to the process, method, product, or apparatus.

[0051] For ease of understanding, several terms involved in the present invention are first introduced:

[0052] Brain functional atlas (full name in English: brain functional atlas), referred to as "brain atlas" or "brain map" or "functional map" or "brain functional network map"; refers to the functional division of the cerebral cortex, marking the different areas of the brain responsible for different functions; the marked areas are also called "functional areas" or "functional networks."

[0053] Individualized brain functional atlas (English full name: individualized brain functional atlas), referred to as individual brain atlas, is a brain functional atlas that represents an individual.

[0054] The group level brain functional atlas (full name in English: group level brain functional atlas), referred to as the group brain atlas, is a statistical brain functional atlas that represents a group of many people.

[0055] Brain functional networks (full name: brain functional networks) represent the functional connections of the brain.

[0056] Functional magnetic resonance imaging (fMRI), also known as functional magnetic resonance imaging or MRI, shows the activated areas of the brain when stimulated by external stimuli.

[0057] Task functional magnetic resonance imaging (task fMRI) involves having subjects perform a designated task (for example, moving the tongue) during an MRI scan, in the hope of using MRI to identify the brain functional area or network responsible for the task.

[0058] Voxel, also known as voxel, is short for volume pixel. Conceptually, a voxel is similar to the smallest unit of two-dimensional space—the pixel—used in two-dimensional computer graphics. Voxel is the smallest unit of digital data used in three-dimensional space segmentation and is used in fields such as 3D imaging, scientific data, and medical imaging.

[0059] The BOLD (Blood Oxygen Level Dependency) signal is the blood oxygen level signal collected by functional magnetic resonance imaging.

[0060] Voxel time series signal: During functional magnetic resonance imaging, the BOLD signal of each voxel is generally collected for a period of time (for example, two minutes). The BOLD signal during this period is the time series signal of the voxel;

[0061] A region of interest (ROI) is an area that needs to be processed in machine vision and image processing, which is outlined from the image being processed in the form of a box, circle, ellipse, or irregular polygon.

[0062] The Pearson correlation coefficient is used to measure whether two data sets are on the same line, that is, to measure the linear relationship between interval variables. Its calculation formula is:

[0063]

[0064] The Pearson correlation coefficient (ρx,y) between two continuous variables (X,Y) is equal to their covariance cov(X,Y) divided by the product of their standard deviations (σX,σY). The coefficient always ranges from -1.0 to 1.0. Variables close to 0 are considered uncorrelated, while those close to 1 or -1 are considered strongly correlated.

[0065] The present invention is described in detail below through specific examples.

[0066] Please refer to Figure 1 , an embodiment of the present invention provides a method for drawing a brain function map, which may include the steps of:

[0067] S1. Data acquisition step: obtaining magnetic resonance imaging (MRI) scan data of individual brain regions of the subject (e.g., functional MRI scan data, specifically BOLD signals, which are related to brain functional areas; or structural MRI scan data, specifically MRI T1 signals, which are related to brain structure), and pre-selecting or generating a group brain function map as a brain function map template.

[0068] Functional magnetic resonance imaging (fMRI) is an emerging neuroimaging method that uses magnetic resonance imaging to measure changes in blood oxygen levels caused by neuronal activity and collect BOLD signals. It is currently mainly used to study the functional activities of the brain or other nervous systems of humans and animals. A resting-state fMRI scan is a scan in which the subject lies in a quiet state in the MRI scanner, with the whole body relaxed and not performing any tasks or systematic thinking. The present invention can particularly use resting-state fMRI scan data as raw data to draw brain function maps. Functional magnetic resonance scan data is voxel-based four-dimensional imaging data, including time series signals (BOLD signals) of all voxels.

[0069] It is necessary to obtain a group brain function map in advance, which serves as a brain function map template to initialize the individual brain function map. The brain function map template divides the brain into multiple functional areas, such as 56 or 112 or other numbers. Optionally, the brain function map template can divide the brain into several large areas according to structural information. Preferably, the brain is divided into two large areas of left and right cortex. Preferably, the brain is divided into two large areas according to high-level cortex and low-level cortex; more preferably, the brain is divided into four large areas according to high-level cortex and low-level cortex according to left and right cortex; more preferably, the left and right cortex of the brain are divided into five large areas according to structural information, namely the frontal lobe, parietal lobe, occipital lobe, temporal lobe and pan-central sulcus area, thus dividing the brain into ten large areas. All functional areas belong to these large areas respectively, and each large area includes several functional areas.

[0070] S2. Initialization step: Initialize the individual's brain function map using a brain function map template to obtain an initial individual brain function map. The initial individual brain function map divides the brain into multiple functional areas.

[0071] In the initialization step, for each large area of ​​the brain structure, the functional partitioning of the brain function map template can be projected onto the individual's reconstructed cerebral cortex to obtain an initialized individual brain function map. This "projection" is achieved by using mathematical methods such as interpolation to complete one-to-many, many-to-many, and many-to-one registrations based on the point-level mapping relationship between different images of the same individual. Different registration methods (rigid, radial, nonlinear, etc.) use different formulas and constraints. For example, the BBR algorithm provided by Douglas Greve et al. is a commonly used "projection" method (Douglas Greve and Bruce Fishl, Accurate and robust brain image alignment using boundary based registration, NeuroImage 2009).

[0072] Due to individual differences, this initial individual brain function map does not reflect the individual's true brain function partitioning, and requires subsequent iterative calculations to improve accuracy and individual precision.

[0073] S3. Preprocessing step: Divide the initial individual brain function map into several large regions, each of which includes several functional regions, where several means at least two.

[0074] For example, the left and right cortices of the brain can be divided into five major regions: the frontal lobe, parietal lobe, occipital lobe, temporal lobe, and pan-central sulcus area, and the initial individual brain function map can be divided into ten major regions. For another example, the brain can be divided into four regions in total based on the higher and lower cortices.

[0075] During this preprocessing step, the credibility of each voxel can also be initialized. This credibility indicates the degree of confidence that the voxel belongs to the functional area it is currently located in. Optionally, the initial value of the credibility of each voxel can be set to 1. In subsequent iterative calculations, the credibility will be continuously updated.

[0076] S4, iterative calculation step: Entering the iteration process, each iteration includes: sequentially calculating the connectivity of each voxel in each macroregion with each functional region in the macroregion, adjusting each voxel to the functional region with the highest connectivity, until all voxels are adjusted. When the termination condition is met, the iteration is terminated, and the final individual brain function map is obtained. In addition, based on the final individual brain function map, a corresponding report and / or image showing brain function zoning can be generated and output.

[0077] The iterative calculation step is a process of repeated iterative calculations. All voxels are adjusted in each iteration. After the iterative process is completed, it is judged whether the termination condition is met. If not, the result output by the previous iterative process is used as input to enter the next iteration and adjust all voxels again until the termination condition is met and the final individual brain function map is output.

[0078] In the final individual brain function map, each voxel has been repeatedly adjusted to the most likely functional area, resulting in a high degree of accuracy. By setting appropriate iteration termination conditions, a good balance between accuracy and computational time can be achieved, resulting in a highly accurate final individual brain function map.

[0079] Optionally, whether a convergence criterion is met or a preset number of iterations is reached can be used as an iteration termination condition. When the number of iterations reaches the preset number or the convergence criterion is met, the iteration is terminated.

[0080] Optionally, one convergence criterion is: the change between two iterations is very small: for example, the change between two iterations is less than 1%, preferably less than 0.1% or 0.5%, etc.; optionally, another convergence criterion is: all credibility reaches a certain threshold. According to experience, the threshold can be, for example, 2 to 10, preferably 3, 4 or 5; optionally, the preset number of times can be, for example, 5 to 1000 times. Preferably, the preset test is 100 times, 150 times, or 200 times.

[0081] Optionally, the connectivity between each voxel in each macroregion and each functional region in the macroregion can be calculated using the correlation value between the voxel's time series signal and the reference time series signal of the functional region. The correlation value represents the correlation, correlation coefficient, or similarity between the voxel's time series signal and the reference time series signal of the functional region. The higher the similarity between the two, the higher the correlation value, indicating that the two are closer and the connectivity is higher (or larger or stronger).

[0082] Furthermore, various correlation calculation methods may be used to calculate the correlation value, which is not limited herein. For example, the Pearson correlation value between the time series signal of the voxel and the reference time series signal of the functional area may be calculated to represent the connectivity between the voxel and the functional area.

[0083] Please refer to Figure 2 In some embodiments, each iteration process may specifically include:

[0084] S41. Calculate a reference time series signal for each functional region based on the time series signals of all voxels in each functional region. For example, for each functional region, the average or median of the time series signals of all voxels in the functional region whose confidence level is not less than a preset threshold may be calculated as the reference time series signal for the functional region. Optionally, during the first iteration, the average or median of the time series signals of all voxels in the functional region may be calculated as the reference time series signal for the functional region.

[0085] S42: Determine an unadjusted voxel as the current voxel.

[0086] S43. Calculate the correlation value between the time series signal of the current voxel and the reference time series signals of each functional area in the current region, and use the correlation value as the connectivity between the current voxel and each functional area in the current region, where the current region is the region to which the current voxel belongs.

[0087] S44. Adjust the current voxel to the functional area with the highest connectivity to the current voxel.

[0088] By adjusting the voxel to the functional area with the highest connectivity (i.e., the largest correlation value), the accuracy of functional area division is improved, and the individual brain functional map is updated. Optionally, if there are multiple equal maximum correlation values, the current voxel can be adjusted or reallocated to any functional area with the largest correlation value.

[0089] S45. Update the credibility of the current voxel.

[0090] Optionally, the maximum correlation value and the second largest correlation value can be selected from the calculated correlation values ​​of the time series signal of the current voxel and the reference time series signals of each functional area in the current region; when calculating and updating the credibility of the current voxel, the credibility of the current voxel is set to be equal to the ratio of the maximum correlation value to the second largest correlation value.

[0091] S46 , determining whether all voxels have been adjusted once; if not, proceeding to step 42 , selecting the next voxel, and continuing to execute the above steps 43 and 44 ; if yes, proceeding to step 47 .

[0092] S47: If all voxels have completed one adjustment, then end the iteration process.

[0093] The end of this iteration process indicates that a round of adjustment and update of the individual brain function map has been completed; next, if the iteration termination condition has not been met, the next iteration process will be entered based on the updated individual brain function map obtained in this iteration process.

[0094] Please refer to Figure 3In some embodiments, after collecting functional magnetic resonance imaging (fMRI) scan data of a group, a group brain function map can be generated as a brain function map template by the following method, comprising the steps of:

[0095] A1. Divide the brain into several major regions based on structural information. Optionally, the left and right cortices of the brain can be divided into five major regions each, namely the frontal lobe, parietal lobe, occipital lobe, temporal lobe, and pan-central sulcus region, thus dividing the brain into ten major regions.

[0096] A2. Calculate a population functional connectivity profile within each region. This is the functional connectivity matrix of the voxels within the region and the regions of interest (ROIs) within the region. This can be accomplished by the following steps:

[0097] ① For each voxel in each region of each individual, calculate the functional connectivity of the voxel to each ROI in the region to obtain a multi-dimensional vector, such as a 1000-dimensional vector or a 4098-dimensional vector. Preferably, the ROI can be obtained by uniform sampling;

[0098] ② Average the multi-vectors of each voxel in each region of all individuals in the group to obtain the group functional connectivity vector of each voxel, and form a matrix of the functional connectivity vectors of all voxels to obtain the group functional connectivity map (or group functional connectivity matrix).

[0099] A3. Using these functional connectivity maps as features, clustering algorithms are used to classify each voxel within each region, ultimately dividing each region into dozens of "clusters" (fine-grained functional partitions). Clustering algorithms include but are not limited to K-means and its related algorithms, spectral clustering, and Gaussian mixed models.

[0100] A4. Comprehensively evaluate the clustering algorithm's metrics and the maximization of functional homogeneity to determine multiple locally optimal partitions. For example, the clustering algorithm's metrics utilize the "distance" between voxels in 1000-dimensional space to group voxels with close "distances" into the same fine-grained functional partition, thereby achieving "local optimality."

[0101] A5. Use a merging algorithm to merge the fine-grained functional partitions of multiple large regions to create a whole-brain population-level functional partitioning template. Merging methods include, but are not limited to, direct merging, followed by edge trimming and edge voxel redistribution. The resulting population brain functional map template can have varying resolutions, ranging from dozens to hundreds of functional regions.

[0102] It can be understood that the above-mentioned solutions of the embodiments of the present invention can be specifically implemented in local or cloud-based data processing systems such as local servers and cloud computing platforms.

[0103] To facilitate a better understanding of the technical solution provided by the embodiments of the present invention, an implementation method in a specific scenario is used as an example to introduce it below.

[0104] Please refer to Figure 4 , is a flowchart of a method for drawing a brain function map in an embodiment of a specific application scenario, the steps comprising:

[0105] B1. Obtain individual functional magnetic resonance imaging (fMRI) scan data, including time series signals of all voxels.

[0106] B2. Select a brain function map template.

[0107] B3. Initialize the individual's brain function map using the brain function map template to obtain an initial individual brain function map.

[0108] B4. Initialize the credibility of each voxel to the maximum credibility of 1.

[0109] B5. Divide the individual brain map into multiple local areas, that is, multiple large areas, each of which covers multiple functional areas.

[0110] B6. Enter iteration.

[0111] B7. Determine whether the iteration termination condition is met. If not, proceed to step B8; if so, proceed to step B16.

[0112] B8. Calculate the reference signal (full name: reference time series signal) of all functional areas: For each functional area, select voxels with a credibility not less than a certain threshold to calculate the reference time series signal of the functional area.

[0113] B9. Perform the following calculations on each voxel in turn.

[0114] B10. Determine whether all voxels have been calculated once. If so, return to step B7; if not, determine the next voxel that has not been calculated and adjusted as the current voxel and go to step B11.

[0115] B11. For each functional area located in the same region as the current voxel, calculate the connectivity between the current voxel and the functional area.

[0116] B12. Sort the connectivity of all functional areas in the same region as the current voxel.

[0117] B13. Adjust individual brain function maps: assign the current voxel to the functional area with the strongest connectivity.

[0118] B14. Calculate the confidence level that the current voxel belongs to the adjusted functional area.

[0119] B15. Return to step B10.

[0120] B16. If the iteration termination conditions are met, end the iteration and generate the corresponding report and / or image.

[0121] As can be seen from the above, in some feasible embodiments of the present invention, a method for drawing a brain function map is provided. This method iteratively calculates the initial individual brain function map, and continuously adjusts the functional area to which each voxel belongs through the iterative process. Ultimately, a high-precision individual brain function map can be drawn, which can achieve high stability, high reliability and low noise at a variety of different accuracies or resolutions.

[0122] It should be noted that the present invention first divides the region and then adjusts or reallocates voxels in each region based on the region. Compared with the whole-brain individualization method of the prior art that directly allocates voxels across the entire brain, this regional individualization method, on the one hand, naturally introduces general neuroscience laws such as "voxels of the same fine-grained functional partition should belong to the same region" as additional partitioning information by adjusting voxels within each region, effectively improving the ability to resist noise and interference. On the other hand, since voxel adjustment is performed within a smaller range such as a region, the processing accuracy and calculation speed can be effectively improved.

[0123] In the practice of the present invention, the whole brain can be divided into 56 functional areas, 112 functional areas, 213 functional areas and other functional areas with different accuracies, and the results are stable and reliable and have been successfully clinically verified.

[0124] Verification method 1, please refer to Figure 8 , Figure 8 The left column is the cerebral cortex of one person, and the right column is the cerebral cortex of another person: the first row uses the intraoperative direct electrical cortical stimulation (ECS) technology to locate the functional area (which is also the commonly used clinical method for locating functional areas), and the second row uses the method of the present invention to locate the functional area. It can be seen that the position of the functional area located by the method of the present invention is consistent with the position located by the ECS technology, which verifies the accuracy of the positioning of the present invention.

[0125] Verification method 2: Please refer to Figure 9 ,Verify the movement network, different color parts are part of the movement network, for example, one part controls hand movement, and another part controls body movement (face, head, tongue).

[0126] Figure 9 The right column shows a motor network diagram obtained through extensive experimental sampling using traditional methods (task-based functional MRI). The specific method involves having healthy individuals perform tasks, such as tongue movements or finger pinching, while scanning under a functional MRI scanner. This activity is performed for 1.5 hours. After 10 consecutive days of scanning, the brain functional areas associated with movement are located. Due to the large number of experimental samples, this method can accurately identify the corresponding functional areas.

[0127] Figure 9 In the figure, the left column shows the method of the present invention. It can be seen that the localization area on the left is basically the same as the localization area on the right. The yellow boundary in the right figure is the result of the comparison between the present invention and the figure on the right, showing that the motor function network obtained by the two methods has a very high degree of overlap. In addition, the present invention does not require any tasks to be performed, and the sampling time is short (twenty minutes on a 3T MRI machine). Compared with task-state functional MRI, the present invention has obvious advantages. The short time to confirm the individual brain functional area is more conducive to clinical application.

[0128] Please refer to Figure 10, a brain functional map obtained using a specific mapping method of the present invention. The specific mapping steps are as follows: A young, normal subject undergoes structural and resting-state functional MRI scans and performs signal preprocessing. A brain map template of 213 functional regions at the population level is selected for initialization. The population-level partitions are then projected onto the individual cerebral cortex using nonlinear registration. The individual cerebral cortex is structurally divided into five major regions on each side: the frontal lobe, parietal lobe, occipital lobe, temporal lobe, and pancentral sulcus region. Individualized computational iterations are then performed on each region within the individual cortex. Taking the 17 functional regions of the left frontal lobe as an example, in the first iteration of individualization, the time series signals of all voxels within each initialized functional region are averaged. The resulting average signal serves as the reference signal for that region, totaling 17 reference signals. The Pearson coefficient is then calculated for the correlation between the time series signal of each voxel in the left frontal lobe and the 17 reference signals, and the 17 correlation values ​​are sorted in descending order. The voxel is then reassigned to the functional region with the highest correlation value, and its credibility is calculated. The confidence level is the ratio of the maximum correlation value to the second-largest correlation value, and the confidence level ranges from [1 to +∞]. Assigning a voxel to a functional region with a confidence level of ≥3 is considered highly confident. For example, if the correlation between voxel 110 and partition 10 is 0.78, the maximum correlation value, and its correlation with partition 8 is 0.18, the second-largest correlation value, then the confidence level is 0.78 / 0.18 = 4.3, indicating that we are highly confident in assigning this voxel to partition 10. For another example, if the strongest functional connectivity of voxel 125 is 0.35 with partition 8 and its second-strongest functional connectivity is 0.29 with partition 4, then the confidence level is 0.35 / 0.29 = 1.2, indicating that we are less confident in assigning this voxel to partition 8. Repeat the above steps for all voxels to complete the first individualized update. Starting from the second iteration, 17 reference signals are recalculated each time. This reference signal is calculated by averaging the time series signals of voxels with a confidence score greater than 3 within each partition. If no voxel in a partition has a confidence score greater than 3, the top 5% of voxels with the highest confidence scores are used as replacements. Each voxel in the large region is then reassigned to a new partition and its confidence score estimated based on its correlation with the reference signal. After all voxels within the large region are updated, the next iteration begins. This continues until the similarity between the individual partitions of the two iterations reaches 99% or higher, or after 110 iterations. Finally, after all large regions have been individualized, 10 large regions are merged using a direct merging method. This completes the individualized functional partitioning, resulting in an individual brain functional map of 213 functional regions. Boundaries are added to the boundaries of each functional region in the individual brain functional map to more clearly distinguish them.

[0129] Please refer to Figure 11, is a comparison diagram of the effects of the brain function map drawn by the method of the present invention and the brain function map drawn by the prior art.

[0130] like Figure 11 As shown in (a) on the left, the method of the present invention is used to draw functional areas, and the brain is divided into functional areas by individualizing large areas. Figure 11 As shown in (b) on the right, the functional areas are drawn using the technical solution of patent US 9,662,039B2, which divides the brain into functional areas in a simultaneous individualized manner across the entire brain.

[0131] As can be seen in the image (b) on the right, there is a lot of obvious noise, as shown by 101 and 102. Noise is a small area in the image that violates neuroscience principles. Neuroscience principles include: at the same resolution, a functional area of ​​the nervous system should generally not be within other functional areas. If a red area appears green inside, it indicates a problem with the processing result. The boundaries of each functional area are usually smooth to a certain extent. If the boundaries are jagged or there are multiple discontinuous areas in a functional area, it is usually caused by noise interference in the processing method.

[0132] In order to better implement the above solutions of the embodiments of the present invention, relevant devices for cooperating in implementing the above solutions are also provided below.

[0133] Please refer to Figure 5 , an embodiment of the present invention provides a system for drawing a brain function map, which may include:

[0134] Initialization module 51, used to initialize the individual brain function map using the brain function map template to obtain an initial individual brain function map, which divides the brain into multiple functional areas;

[0135] A pre-processing module 52 is used to divide the initial individual brain function map into several large regions, each of which includes several functional areas;

[0136] The iterative processing module 53 is used to enter the iteration. Each iteration process includes: calculating the connectivity of each voxel in each large area with each functional area in the large area in turn, adjusting each voxel to the functional area with the highest connectivity with the voxel, until all voxels are adjusted; when the termination condition is met, the iteration is terminated to obtain the final individual brain function map.

[0137] It can be understood that the functions of each functional module of the brain function map drawing system in the embodiment of the present invention can be specifically implemented according to the method in the above method embodiment. The specific implementation process can refer to the relevant description in the above method embodiment, which will not be repeated here.

[0138] It is understandable that the brain function map drawing system of the embodiment of the present invention can be implemented in different forms.

[0139] On the one hand, it can be implemented by a local computer device, such as a server. The server obtains functional magnetic resonance imaging (fMRI) scan data from the MRI device, performs computational processing, obtains the final individual brain function map, and further generates and outputs corresponding reports and / or images showing brain functional partitioning.

[0140] Alternatively, cloud computing platforms can be used to implement this. First, a server can acquire functional MRI scan data from the MRI device and upload the data to a cloud platform. The cloud platform then performs computational processing to obtain a final individual brain function map and can further generate and output corresponding reports and / or images showing brain functional partitioning. The cloud platform can also transmit the final individual brain function map, report, and / or image back to the server.

[0141] Alternatively, this could be achieved using a data processing system comprised of a server and a cloud computing platform. First, the server could acquire functional MRI scan data from the MRI device and collaborate with the cloud platform to perform computational processing to obtain a final individual brain function map. This could then be used to generate and output reports and / or images showing brain functional partitioning.

[0142] As can be seen from the foregoing, some feasible embodiments of the present invention provide a system for mapping brain function. This system is used to iteratively calculate an initial individual brain function map, continuously adjusting the functional area to which each voxel belongs through an iterative process. Ultimately, a high-precision individual brain function map can be drawn, achieving high stability, high reliability, and low noise at various levels of precision or resolution. In practice, the present invention can divide the entire brain into 56 functional areas, 112 functional areas, 213 functional areas, and other different levels of precision, with stable and reliable results, which have been successfully clinically validated.

[0143] Please refer to Figure 6 , an embodiment of the present invention further provides a data processing system 60, which may include:

[0144] Processor 61, memory 62, communication interface 63, bus 64,

[0145] The processor 61, memory 62, and communication interface 63 communicate with each other through the bus 64; the communication interface 63 is used to receive and send data; the memory 62 is used to store computer execution instructions. When the data processing system is running, the processor 61 executes the computer execution instructions stored in the memory 62, so that the data processing system executes the method for drawing brain function maps as described in the method embodiment above.

[0146] The data processing system 60 may be composed of a local computer device such as a local server or a cloud computing platform, or may be composed of both.

[0147] Please refer to Figure 7-1 to Figure 7-3 , an embodiment of the present invention further provides a medical image processing system, comprising: a data acquisition system 71, a data output interaction system 72, and, the data processing system as described above; the data processing system comprises a local server 73 and / or a cloud computing platform 74;

[0148] The data acquisition system 71 includes a magnetic resonance device, etc., for collecting functional magnetic resonance imaging data of an individual and uploading the collected data to the data processing system;

[0149] The data processing system is used to execute the above Figure 1 The method for drawing a brain function map described in the embodiment obtains a final individual brain function map, and based on the final individual brain function map, generates corresponding data processing results such as a report and / or image;

[0150] The data output interaction system 72 includes a display device, input and output devices, etc., and is used to obtain data processing results in the form of reports and / or images from the data processing system and display the data processing results;

[0151] Wherein, the data processing system may only include a local server, such as Figure 7-1 As shown; it can also include only cloud computing platforms, such as Figure 7-2 Alternatively, it can also include both local servers and cloud computing platforms, such as Figure 7-3 shown.

[0152] When the data processing system includes both the local server and the cloud computing platform, the local server and the cloud computing platform jointly execute the method for drawing brain function maps provided by the present invention based on a load balancing strategy and / or a sharing strategy.

[0153] The interaction between the local server and the cloud computing platform may include, for example:

[0154] 1. The work that the local server cannot complete is assigned to the cloud computing platform. A load balancing strategy between the cloud computing platform and the local server may be, for example: when the load of the local server exceeds a certain set value, such as 70%, new work is transferred to the cloud computing platform for processing.

[0155] 2. When users have sharing requirements or other location needs, they can hand over the work to the cloud computing platform.

[0156] An embodiment of the present invention also provides a computer storage medium, a computer-readable storage medium storing one or more programs, wherein the one or more programs include computer execution instructions, and when the computer execution instructions are executed by a data processing system including a processor, the data processing system executes the method for drawing a brain function map as described in the method embodiment above.

[0157] It should be noted that for the aforementioned method embodiments, for simplicity of description, they are all expressed as a series of action combinations. However, those skilled in the art should be aware that the present invention is not limited to the order of the actions described, because according to the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present invention.

[0158] Those skilled in the art will clearly understand that for the convenience and brevity of description, the specific working process of the system and device described above can refer to the corresponding process in the aforementioned method embodiment. For parts not described in detail in an embodiment, please refer to the relevant description of other embodiments.

[0159] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0160] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0161] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0162] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program code.

[0163] The above is a detailed introduction to the method and system for drawing brain function maps provided by the embodiments of the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core ideas. At the same time, for those skilled in the art, according to the ideas of the present invention, there may be changes in the specific implementation methods and application scopes. In summary, the contents of this specification should not be understood as limiting the present invention.

Claims

1. A brain function map drawing system, characterized in that: The rendering system is implemented by a local computer device and / or a cloud computing platform. The drawing system includes: an initialization module, configured to initialize the individual's brain function map using a brain function map template to obtain an initial individual brain function map, wherein the initial individual brain function map divides the brain into multiple functional areas; A preprocessing module is used to divide the initial individual brain function map into several large regions, each of which includes several functional areas; The iterative processing module is used to enter the iteration. Each iteration process includes: calculating the connectivity of each voxel in each large area with each functional area in the large area in turn, adjusting each voxel to the functional area with the highest connectivity with the voxel, until the adjustment of all voxels is completed; when the termination condition is met, the iteration is terminated to obtain the final individual brain function map.

2. The drawing system according to claim 1, wherein: When the drawing system is implemented by a local computer device, The local computer device is a server, which communicates with the magnetic resonance device to obtain magnetic resonance scanning data, perform calculations and processing, and obtain the final individual brain function map.

3. The drawing system according to claim 2, wherein: The server generates and outputs corresponding reports and / or images showing brain functional divisions.

4. The drawing system according to claim 2, wherein: When the mapping system is implemented through a cloud computing platform, the server communicates with the magnetic resonance device and the cloud computing platform respectively. The server acquires magnetic resonance scanning data from the magnetic resonance device and uploads the data to a cloud computing platform, which then performs computational processing to obtain the final individual brain function map.

5. The drawing system according to claim 4, wherein: The cloud computing platform generates and outputs corresponding reports and / or images showing brain functional divisions.

6. The drawing system according to claim 5, wherein: The cloud computing platform transmits the final individual brain function map, report and / or image back to the server.

7. The drawing system according to claim 1, wherein: When the mapping system is implemented by a server and a cloud computing platform forming a data processing system, the server communicates with the magnetic resonance device and the cloud computing platform respectively. First, the server acquires magnetic resonance scanning data from the magnetic resonance device, and performs computational processing in collaboration with the cloud computing platform to obtain the final individual brain function map.

8. The drawing system according to claim 7, wherein: The server or cloud computing platform generates and outputs corresponding reports and / or images showing brain functional divisions.

9. The drawing system according to claim 7, wherein: The server and the cloud computing platform jointly implement brain function mapping based on a load balancing strategy and / or a sharing strategy to obtain a final individual brain function map.

10. The drawing system according to any one of claims 2 to 9, characterized in that: Before obtaining the initial individual brain function map, data acquisition is also performed to obtain magnetic resonance imaging data of the subject's brain. The magnetic resonance scan data includes functional magnetic resonance scan data and structural magnetic resonance scan data, The functional magnetic resonance scan data includes a time series signal, and the structural magnetic resonance scan data includes a magnetic resonance T1 signal.

11. The drawing system according to any one of claims 1 to 9, characterized in that: The mapping system is used to iteratively calculate the initial individual brain function map, continuously adjust the functional area to which each voxel belongs through the iterative process, and finally map the individual brain function map.

12. The drawing system according to any one of claims 1 to 9, characterized in that: Each iteration process specifically includes: Calculate the reference time series signal of each functional area according to the time series signals of all voxels in each functional area; Determine an unadjusted voxel as the current voxel; Calculating correlation values ​​between a time series signal of a current voxel and reference time series signals of each functional area in a current region, using the correlation values ​​as connectivity between the current voxel and each functional area in the current region, where the current region is the region to which the current voxel belongs; Adjust the current voxel to the functional area with the highest connectivity to the current voxel; Determine whether all voxels have completed an adjustment. If not, return to the step of determining an unadjusted voxel as the current voxel. If all voxels have completed an adjustment, the iteration process ends.

13. The drawing system according to claim 12, wherein: Before entering the iteration, it also includes: initializing the credibility of each voxel; After adjusting the current voxel to the functional area with the highest connectivity to the current voxel, the method further includes: updating the credibility of the current voxel; Calculating the reference time series signal of each functional area includes: for each functional area, calculating the average value or median of the time series signals of all voxels in the functional area whose credibility is not less than a preset threshold as the reference time series signal of the functional area.

14. The drawing system according to claim 13, wherein: The Pearson correlation value between the time series signal of the voxel and the reference time series signal of the functional area is calculated and used to represent the connectivity between the voxel and the functional area.

15. The drawing system according to claim 13, wherein: The updating of the credibility of the current voxel includes: Selecting the maximum correlation value and the second largest correlation value from the calculated correlation values ​​of the time series signal of the current voxel and the reference time series signals of each functional area in the current region; The credibility of the current voxel is calculated and updated. The credibility of the current voxel is equal to the ratio of the maximum correlation value to the second largest correlation value.

16. The drawing system according to claim 12, wherein: The brain function map template is a pre-selected or generated group brain function map, and the steps include: Divide the brain into several major regions; A group functional connectivity map is calculated in each region. The functional connectivity map is a functional connectivity matrix of voxels in the region and N regions of interest in the region, where N is a natural number. Based on the functional connectivity map as a feature, each large region is divided into multiple fine-grained functional partitions using a clustering algorithm; Comprehensively evaluate the clustering algorithm's indicators and the indicator of maximizing functional homogeneity to determine multiple locally optimal numbers of partitions; The merging algorithm is used to merge the fine-grained functional partitions within the large region to create a whole-brain group brain function map as the required brain function map template.

17. The drawing system according to claim 16, wherein: The step of dividing the brain into several large regions is to divide the brain into several large regions according to the structural information in the brain magnetic resonance scanning data of a group of people.

Images