Search

CN-122025192-A - Brain development characteristic evolution analysis system and method based on multi-mode brain image data

CN122025192ACN 122025192 ACN122025192 ACN 122025192ACN-122025192-A

Abstract

The invention discloses a brain development characteristic evolution analysis system and method based on multi-mode brain image data. The system comprises a brain development map construction module, a data acquisition module, a data processing module and an evolution matching evaluation module. The method comprises the steps of acquiring demographic information of healthy people in multiple age groups and multi-modal brain image data, constructing a three-dimensional group brain development map, connecting and activating the three-dimensional group brain development map, processing individual data by adopting the same flow, acquiring multi-modal brain feature vectors in a unified brain region space, mapping the multi-modal brain feature vectors to a corresponding group map space, calculating the deviation amount, combining the stability weight to obtain a fusion deviation index, identifying abnormal brain regions according to the fusion deviation index, calculating abnormal proportion and risk change rate, and realizing brain development feature evolution analysis. The invention fuses the population evolution priori and individual multi-modal data, can dynamically and accurately evaluate brain development, and solves the problems of static detection, population individual disjoint, multi-modal fusion deficiency and the like.

Inventors

  • LI AO
  • HUANG MINQIANG
  • WANG WEI

Assignees

  • 上海术理智能科技有限公司
  • 上海通用术理智慧医疗脑科学研究院

Dates

Publication Date
20260512
Application Date
20260414

Claims (10)

  1. 1. The brain development characteristic evolution analysis system based on the multi-mode brain image data is characterized by comprising the following components: The brain development map construction module is used for acquiring demographic information and multi-mode brain image data of healthy people in multiple age groups, preprocessing, brain region segmentation and feature extraction based on the multi-mode brain image data of the people in multiple age groups, extracting structural features, connection features and activation features, and constructing a group brain development map reflecting the evolution rule of the group brain structure, functions and connection modes along with the age by combining the demographic information of the people in multiple age groups; The data acquisition module is used for acquiring demographic information and multi-mode brain image data of the individual to be evaluated, and the multi-mode brain image data of the individual to be evaluated are consistent with the multi-mode brain image data acquired by the brain development map construction module in type; the data processing module is used for preprocessing the multi-modal brain data acquired by the data acquisition module, segmenting brain regions and extracting features by adopting the same processing method as the brain development map construction module to acquire individual multi-modal brain feature vectors of the individual under a unified brain region space coordinate system; The evolution matching evaluation module is used for mapping the individual multi-mode brain feature vector into a corresponding group brain development map space according to demographic information of an individual to be evaluated, obtaining a fusion deviation index by calculating deviation of the individual multi-mode brain feature vector relative to an evolution track of the same age group and combining stability weights of all mode features, quantitatively evaluating the evolution degree of an individual brain development state relative to a group reference according to the fusion deviation index, identifying an abnormal brain region based on the fusion deviation index, calculating the proportion of the abnormal brain region and the risk change rate, realizing brain development feature evolution analysis, and outputting a brain development feature evolution analysis result.
  2. 2. The brain development feature evolution analysis system according to claim 1, wherein the population brain development pattern constructed by the brain development pattern construction module includes a structural development pattern, a connective development pattern, and an activation development pattern; the structural development map is constructed based on extraction of brain region level structural characteristics and is used for establishing a distribution model of brain region volume, gray matter volume, cortex thickness and cortex surface area changing along with age; the connection development map is constructed based on spontaneous brain function signals in a resting state, and a distribution model of connection strength between brain networks and the whole brain connection level changing along with age is built by utilizing the time sequence correlation of the spontaneous brain function signals; the activation development map is constructed based on brain function response signals under task states, and a distribution model of brain region activation degree and multi-voxel mode similarity variation along with age under specific tasks is established by extracting activation characteristics.
  3. 3. The brain development feature evolution analysis system according to claim 1, wherein the multi-modal brain image data includes structural magnetic resonance imaging data, resting state functional magnetic resonance imaging data, functional magnetic resonance imaging data for specific cognitive tasks, and functional near infrared brain imaging data.
  4. 4. The brain development feature evolution analysis system according to claim 1, wherein the stability weight in the evolution matching evaluation module is determined according to the fluctuation range of each modal feature in the same-age healthy population, the smaller the fluctuation range is, the larger the weight is, the smaller the fluctuation range is, and the sum of the weights of the same brain region in different modes is 1.
  5. 5. The brain development characteristic evolution analysis system according to claim 1, wherein the evolution matching evaluation module identifies an abnormal brain region, adopts a preset abnormal discrimination threshold, judges a brain region as an abnormal brain region when the absolute value of a fusion deviation index of a brain region of an individual reaches or exceeds the threshold, and calculates the proportion of the abnormal brain region to all brain regions.
  6. 6. The brain development characteristic evolution analysis method based on the multi-mode brain image data is characterized by comprising the following steps of: The method comprises the steps of 1, obtaining demographic information and multi-mode brain image data of healthy people in multiple age groups, preprocessing, brain region segmentation and feature extraction based on the multi-mode brain image data of the people in multiple age groups, extracting structural features, connection features and activation features, and constructing a group brain development map reflecting the evolution rule of the group brain structure, functions and connection modes along with the age by combining the demographic information of the people in multiple age groups; Step 2, acquiring demographic information and multi-mode brain image data of an individual to be evaluated, wherein the multi-mode brain image data of the individual to be evaluated are consistent with the multi-mode brain image data types of healthy people in multiple age periods; Step 3, preprocessing, brain region segmentation and feature extraction are carried out on the multi-modal brain data of the individual to be evaluated by adopting a processing method which is the same as that for the multi-modal brain image data of the healthy crowd in the age range, so as to obtain individual multi-modal brain feature vectors of the individual under a unified brain region space coordinate system; And 4, mapping the individual multi-mode brain feature vector into a corresponding group brain development map space according to demographic information of an individual to be evaluated, obtaining a fusion deviation index by calculating deviation of the individual multi-mode brain feature vector relative to a same-age group evolution track and combining stability weights of all mode features, quantitatively evaluating the evolution degree of an individual brain development state relative to a group reference according to the fusion deviation index, identifying an abnormal brain region based on the fusion deviation index, calculating the abnormal brain region proportion and risk change rate, realizing brain development feature evolution analysis, and outputting brain development feature evolution analysis results.
  7. 7. The method according to claim 6, wherein the step 1 of extracting the structural features comprises performing brain region segmentation on the multi-mode brain image data by using a cortical segmentation map, counting the number of voxels belonging to each brain region and determining the voxel volume of the brain region by combining the physical volumes of monomeric elements, weighting the voxels in the brain region by combining a gray matter probability map to obtain the gray matter volume of the brain region, calculating the average Euclidean distance from the vertex coordinate set of the endothelial layer of the brain region to the adjacent white matter coordinate set to obtain the cortex thickness of the brain region, and counting the area of the corresponding cortex surface of the brain region to obtain the cortex surface area of the brain region.
  8. 8. The method according to claim 6, wherein the group brain development map in step 1 includes a connection development map, and the construction process of the connection development map includes dividing resting multi-mode brain image data into a plurality of brain networks by using a cortex segmentation map, calculating pearson correlation coefficients of BOLD signal time sequences between any two brain networks, generating a functional connection matrix, calculating the sum of connection strengths of a single brain network and all other brain networks and the average connection strength between all brain networks of the whole brain based on the functional connection matrix, and establishing a distribution model of the connection strength changing with age.
  9. 9. The method according to claim 6, wherein the population brain development map in step 1 comprises an activation development map, and the construction process of the activation development map comprises performing brain region segmentation on task state multi-mode brain image data by using a cortex segmentation map, performing statistics convergence on voxel signals in each brain region, calculating nerve activity activation degree related to a specific task, calculating multi-voxel pattern similarity in the brain region, and establishing a distribution model of brain region activation degree and multi-voxel pattern similarity with age.
  10. 10. The brain development feature evolution analysis method according to claim 6, wherein the specific method for realizing the brain development feature evolution analysis in step 4 includes the steps of: presetting an abnormal judgment threshold, comparing the absolute value of a fusion deviation index of each brain region of an individual to be evaluated with the threshold, screening out abnormal brain regions and forming an abnormal brain region set; counting the number of brain regions in the abnormal brain region set, and calculating the ratio of the total number of the brain regions to the total number of the whole brain regions to obtain the abnormal brain region ratio; calculating risk change rate based on abnormal brain region proportion estimated by an individual at least twice and corresponding estimation time interval, combining age average change trend of the abnormal brain region proportion in a group brain development map to obtain change deviation degree of the individual relative to the same-age healthy crowd, mapping the abnormal brain region proportion to normal model distribution of the same-age healthy crowd, determining the same-age risk percentile to realize cognition risk stratification, and predicting future abnormal brain region proportion of the individual, judging future cognition risk level and adapting to corresponding early warning measures by combining the risk change rate, the change deviation degree and the group evolution trend.

Description

Brain development characteristic evolution analysis system and method based on multi-mode brain image data Technical Field The invention belongs to the technical field of computer-aided biological information processing, and particularly relates to a brain development characteristic evolution analysis system and method based on multi-mode brain image data. Background With the continuous acceleration of the aging process of population, cognitive decline and related neurodegenerative diseases have become important problems affecting the quality of life and social public health of the middle-aged and elderly people. Cognitive aging is usually manifested by multiple functional degradations such as memory decline, attention loss, impaired executive function, and slow information processing speed, and the development process has the characteristics of long-term property, progressive property, obvious individual difference and the like. How to perform early prediction, accurate evaluation and dynamic monitoring on cognitive aging has become an important research direction in the fields of neuroscience, medical engineering, brain-computer interfaces and the like. The existing cognitive aging assessment method mainly depends on neuropsychological scale testing, clinical image examination and partial biomarker analysis. In recent years, with the development of multi-mode brain imaging technology, researchers gradually try to comprehensively analyze the cognitive aging process by fusing various brain signal data such as electroencephalogram, functional near infrared spectrum, functional magnetic resonance imaging, diffusion tensor imaging and the like. However, the following significant problems still remain in the existing cognitive aging assessment and brain image analysis technology system: 1. The limitation of static detection is that the existing method is mainly based on analysis of section data at single or small time points and is essentially static detection. Due to the lack of population evolution references in the full life cycle, continuous dynamic tracks of the cognitive aging process of an individual are difficult to characterize, so that the system cannot predict the trend of the future cognitive state of the individual. 2. Disjoint of population patterns from individual predictions-although partial brain development pattern studies already exist, these patterns are often responsible for population statistical analysis, with the results typically presented as an average model. In the face of individual data, the prior art lacks an effective mapping and alignment mechanism, and is difficult to convert the evolution priori knowledge of a large-scale population into a refined and interpretable positioning evaluation basis for a single individual. 3. The multi-mode fusion depth is insufficient, and the aging of the brain is a comprehensive result of structural atrophy, abnormal functional connection and change of task activation modes. The prior art often only depends on single-mode or simple multi-mode data splicing, and is difficult to identify heterogeneous evolution modes such as 'structure first change' or 'functional recombination'. Therefore, how to construct an evaluation scheme capable of systematically fusing a large-scale population evolution priori and individual longitudinal multi-modal data and realizing accurate and dynamic cognitive aging process and having evolution interpretability is a technical challenge to be solved in the current field. Disclosure of Invention Aiming at the technical problems of the static detection limitation, the dislocation of the group map and the individual prediction and the insufficient multi-modal fusion depth in the prior art, the invention provides a brain development characteristic evolution analysis system and method based on multi-modal brain image data. In order to achieve the technical purpose, the following technical scheme is adopted in the embodiment of the invention. In a first aspect, an embodiment of the present invention provides a system for analyzing evolution of brain development characteristics based on multi-modal brain image data, including: The brain development map construction module is used for acquiring demographic information and multi-mode brain image data of healthy people in multiple age groups, preprocessing, brain region segmentation and feature extraction based on the multi-mode brain image data of the people in multiple age groups, extracting structural features, connection features and activation features, and constructing a group brain development map reflecting the evolution rule of the group brain structure, functions and connection modes along with the age by combining the demographic information of the people in multiple age groups; The data acquisition module is used for acquiring demographic information and multi-mode brain image data of the individual to be evaluated, and the multi-mode brain image data of the individua