Reliable Brain Functional Network Analysis Methods And Applications

Brain image measures calculated based on brain image data can deeply explore brain functions,reveal the relationships between various brain regions,and their roles in cognitive and emotional processes.These measures can serve as objective biomarkers and provide important indicators and evidence for the diagnosis and treatment of mental disorders.The main task of this paper is to obtain stable and reliable brain function network templates based on brain image data;Explore the differences in functional networks and functional network connectivity between male and female of the same age,and compare these differences between different age groups;Develop a toolkit that can analyze large-scale brain function imaging data and extract brain function networks and related measures from it.The analysis of functional magnetic resonance imaging(f MRI)data at rest has shown great potential in understanding brain function and identifying various mental and cognitive states in the brain.However,due to the lack of comparable standards between different data sets,different individuals,and constantly changing brain states,it is not possible to capture the differences and corresponding relationships between different subjects.Using a reliable brain function network template as a guide to estimate the brain function network of individual subjects can overcome the above shortcomings.The first task of this paper is to obtain reliable and stable brain function network templates based on large sample data.This work is based on the resting f MRI data of 1080 healthy youth samples from the Human Connector Project(HCP),using the split-merged independent component analysis(ICA)method to cluster and obtain group level brain functional networks under multiple different model orders,Then,threshold values are set to filter these networks to obtain reliable brain function network templates.Gender differences in the brain are well known,but current research has not fully revealed the impact of gender on brain functional networks.At the same time,as age increases,the changing trend of this difference also needs to be further explored.In order to explore the differences in brain functional networks between male and female and the changes brought about by aging,the second work of this article is based on the Neuro Mark method,using data from two age groups of the UK Biobank to explore the differences in brain functional networks and connections between male and female of the same age group,as well as the differences in functional networks and connections between different age groups.The results show that compared to male,female aged 49 to 50 years old have more developed subcortical and auditory areas in male,while female aged 69 to 70 years old do not have such changes between them,indicating that the differences in functional networks between male and female tend to flatten out with age.For functional network connectivity,the functional connectivity regions with significant differences between female and male aged 49 to 50 years are concentrated in the subcortical,cerebellar,and sensorimotor regions,while for female and male aged 69 to 70 years,the functional connectivity regions with significant differences are only present in a few networks in the subcortical,visual,and sensory regions,and the differences in functional network connectivity gradually weaken with age.As a data driven approach,ICA is widely used to extract brain functional networks.In order to facilitate the use of ICA to analyze brain function networks and brain function network connections,the third task of this article is to develop a MATLAB software package called Intelligent Analysis of Brain Connectivity(IABC).IABC integrates group information guided ICA(GIG-ICA),Neuro Mark,and SMART based ICA methods to estimate reliable brain function networks and related neuroimaging measures for individual subjects.After inputting f MRI data organized by multiple subjects in a specific way and clicking several buttons to set parameters,IABC will automatically calculate and output the brain functional network,relevant time series,and functional network connection matrix for each subject.These neuroimaging measures are expected to provide reliable clues for researchers to understand brain function and brain diseases.