Predicting functional decline in mild cognitive impairment: integrating functional neural network, individual reserve, and myelin water imaging
Description: Objectives: (1) To unveil the complex interaction between mild cognitive impairment (MCI), functional decline, individual reserve (IR), gross and microstructural white matter integrity and functional neural networks using novel structural magnetic resonance myelin water imaging (MWI) in combination with individual reserve brain mapping; and (2) to accurately identify older adults with MCI at greater risk for functional decline using advanced machine learning algorithms.
Hypotheses: (1) IR will moderate the associations between white matter integrity and cognitive and physical function; and (2) features extracted from MWI and IR networks will accurately identify individuals with MCI at greater risk of decline.
Design and Participants: A 12-month longitudinal study comprised of 94 community-dwelling older adults 65 years and older with MCI.
Instruments: Neuropsychological and physical batteries will be administered with MRI including: MWI, T1w Imaging, resting-state functional MRI, and T2w Imaging.
Data Analysis and Expected Results: IR will be calculated via residual variance modelling. IR brain mapping will be performed using the computed IR and resting-state fMRI. Gross white matter damage will be examined via estimating white matter hyperintensity (WMH) volume and microstructural damage via myelin water fraction. WMH will be estimated by an automated machine learning pipeline and myelin water fraction will be extracted from 17 white matter tracts and non-tract regions of interest. IR is expected to significantly moderate the relationship between gross and microstructural white matter damage and cognitive and physical function. Machine learning model will accurately identify older adults with MCI at greater risk for decline.
Funder: Health and Medical Research Fund
Amount: HKD 1,019,000
Funding type: grant
Funder scheme: HMRF - Research Fellowship Scheme
Start date: 2024-01-01
Investigator: Chun Liang HSU
Institution: Department of Rehabilitation Sciences, The Hong Kong Polytechnic University
Funded outputs count: 0
Funder award ID: 10240177
OpenAlex ID: G7375603985