Font Size: a A A

A Study Of Individual Brain Ageing Processes Based On Brain Network Representation Learning

Posted on:2022-08-17Degree:MasterType:Thesis
Country:ChinaCandidate:L Y MaFull Text:PDF
GTID:2480306563479854Subject:Computer Science and Technology
Abstract/Summary:PDF Full Text Request
The increasing aging of the population is a serious social problem in our country today,making aging research particularly necessary.One of the important changes that accompany aging is the degeneration of brain function with the consequent deterioration of behavioral abilities.However,current research on the full stages of brain aging in adults is not comprehensive and in-depth enough.In addition,the understanding of changes in individual differences in brain functional connectivity and behavioral abilities during adult development and aging is not sufficient.An effective model of healthy brain aging process can provide a theoretical basis for research on aging-related psychiatric disorders as well as degenerative brain diseases,and advance clinical diagnosis.To this end,this study was conducted to investigate individual brain aging processes based on learning of brain network representations.Based on resting-state functional magnetic resonance imaging(fMRI)data from a large dataset containing 543 healthy adults(aged 18-88 years),three aspects of this study were carried out,as follows:(1)An individual-level study of the relationship between functional brain networks and age.This work analyzed the relationship between functional connectivity and age in terms of static,dynamic and topological properties of functional connectivity,and further predicted individual age based on three types of functional connectivity characteristics.The results showed that the functional connectivity of human brain showed a change of"the intra-network connectivity decreases with age,the inter-network low-level connectivity decreases with age,and the high-level connectivity increases with age",and the age prediction accuracy based on brain functional connectivity is also relatively high(MAE=7.68 years,R~2=0.61).These results reveal the developmental trajectory of functional aging in the human brain from an individual perspective and validate the effectiveness of functional connectivity as a measure of functional aging in the human brain.(2)A study of individual differences in the development of brain functional connectivity at the cluster level.This study used 10 years as an age window,introduced the index of individual difference value(Ind Var),and analyzed the Ind Var of functional connectivity with age from three perspectives:voxel-level,network-level,and anatomical distance,respectively,to describe its spatial distribution and difference changes.The results showed that 1)the overall Ind Var of the brain increased significantly with age,where Ind Var was always high in the higher functional networks,but started lower in the primary functional networks and then increased with age;2)the strength and quantity distribution of short-,medium-,and long-range connections all showed"inverted U-shaped";3)the short-,medium-,and long-range connections were all negatively correlated with Ind Var.This study is the first to reveal individual differences in the developmental patterns of brain functional connectivity at the cluster level during brain aging.(3)Behavioral level brain cognition and motor ability aging study.Based on individual variability,aging was analyzed for motor ability and fluid intelligence,respectively.The results showed that 1)Ind Var of functional connectivity related to sensory-motor areas was associated with differences in behavioral abilities in a force-matching task,while the differences were more pronounced between genders;2)Ind Var of functional connectivity was significantly and positively correlated with differences in fluid intelligence,and individual variability in brain cognitive functions was manifested as significant differences in IQ.The novelty of this paper is that:(1)The changes of individual differences in functional connectivity during maturation and aging in adults are revealed for the first time,and two different models of individual differences in functional connectivity are proposed;(2)The increase of individual differences in functional connectivity accompanying brain aging confirms that the aging process and maturation process in the brain are two opposite patterns,showing a U-shaped trajectory from developmental maturation to aging.This finding partly explains the great variability in cognitive maintenance during aging and supports the"cognitive maintenance"hypothesis;(3)By examining individual differences in functional connectivity in the brain,we demonstrate that the functional variability in the brain can reflect behavioral performance.
Keywords/Search Tags:Functional magnetic resonance imaging(fMRI), Brain aging, Individual age prediction, Individual differences, Functional connectivity, Anatomical distance, Motor ability, Fluid intelligence
PDF Full Text Request
Related items