Grunnleggende konsepter
Longitudinal changes in brain age are associated with future executive function performance in both Asian children and older adults, with distinct brain features contributing to age prediction in the two populations.
Sammendrag
The study investigated the generalizability of a pretrained deep learning brain age model to Singaporean elderly participants and children. The pretrained model performed well in the elderly participants, but required finetuning for the children.
In the elderly participants from the Epidemiology of Dementia in Singapore (EDIS) study, higher baseline brain age gap (BAG) was associated with poorer baseline cognitive performance, particularly in executive function. In the longitudinal Singapore Longitudinal Aging Brain Study (SLABS) dataset, the early rate of change in BAG was negatively associated with future decline in executive function, independent of baseline BAG.
In the children from the Growing Up in Singapore Towards healthy Outcomes (GUSTO) study, the early rate of change in BAG from 4.5 to 7.5 years old was positively associated with better future inhibitory control at 8.5 years old, independent of baseline BAG.
The model interpretability analysis revealed that the finetuned brain age models focused on distinct brain features in the elderly versus children. In the elderly, regions near the lateral ventricles, frontal/association areas, and subcortical regions were most salient. In children, the prominence of white matter regions, especially in posterior areas, increased compared to the elderly.
These findings suggest that longitudinal changes in brain age can capture ongoing processes of healthy brain aging and development, with distinct neural mechanisms underlying the associations with future executive function in the two populations.
Statistikk
The brain age gap (BAG) is calculated by subtracting chronological age from predicted brain age.
The early rate of change in BAG was calculated from a linear regression of BAG over time for each participant.
The long-term rate of change in cognitive performance was calculated from a linear regression of cognitive scores over time for each participant.
Sitater
"Longitudinal changes in brain age are associated with future executive function performance in both Asian children and older adults, with distinct brain features contributing to age prediction in the two populations."
"In the elderly participants, higher baseline brain age gap (BAG) was associated with poorer baseline cognitive performance, particularly in executive function."
"In the children, the early rate of change in BAG from 4.5 to 7.5 years old was positively associated with better future inhibitory control at 8.5 years old."