TY - JOUR
T1 - Deep learning identifies brain structures that predict cognition and explain heterogeneity in cognitive aging
AU - Saboo, Krishnakant V.
AU - Hu, Chang
AU - Varatharajah, Yogatheesan
AU - Przybelski, Scott A.
AU - Reid, Robert I.
AU - Schwarz, Christopher G.
AU - Graff-Radford, Jonathan
AU - Knopman, David S.
AU - Machulda, Mary M.
AU - Mielke, Michelle M.
AU - Petersen, Ronald C.
AU - Arnold, Paul M.
AU - Worrell, Gregory A.
AU - Jones, David T.
AU - Jack, Clifford R.
AU - Iyer, Ravishankar K.
AU - Vemuri, Prashanthi
N1 - We thank Anu Aggarwal and Jenny Applequist for their constructive comments. This material is based upon work supported by the National Science Foundation under Grant Nos. CNS-1337732 and CNS-1624790 ( CCBGM ); by NIH grants U01 AG006786 , R01 NS097495 , R01 AG056366 , P50 AG016574 , R37 AG011378 , R01 AG041851 , and R01 AG034676 ( Rochester Epidemiology Project, PI: Rocca ), by a Gerald and Henrietta Rauenhorst Foundation grant , and by Mayo/Illinois Alliance Fellowships for Technology-Based Healthcare Research .
PY - 2022/5/1
Y1 - 2022/5/1
N2 - Specific brain structures (gray matter regions and white matter tracts) play a dominant role in determining cognitive decline and explain the heterogeneity in cognitive aging. Identification of these structures is crucial for screening of older adults at risk of cognitive decline. Using deep learning models augmented with a model-interpretation technique on data from 1432 Mayo Clinic Study of Aging participants, we identified a subset of brain structures that were most predictive of individualized cognitive trajectories and indicative of cognitively resilient vs. vulnerable individuals. Specifically, these structures explained why some participants were resilient to the deleterious effects of elevated brain amyloid and poor vascular health. Of these, medial temporal lobe and fornix, reflective of age and pathology-related degeneration, and corpus callosum, reflective of inter-hemispheric disconnection, accounted for 60% of the heterogeneity explained by the most predictive structures. Our results are valuable for identifying cognitively vulnerable individuals and for developing interventions for cognitive decline.
AB - Specific brain structures (gray matter regions and white matter tracts) play a dominant role in determining cognitive decline and explain the heterogeneity in cognitive aging. Identification of these structures is crucial for screening of older adults at risk of cognitive decline. Using deep learning models augmented with a model-interpretation technique on data from 1432 Mayo Clinic Study of Aging participants, we identified a subset of brain structures that were most predictive of individualized cognitive trajectories and indicative of cognitively resilient vs. vulnerable individuals. Specifically, these structures explained why some participants were resilient to the deleterious effects of elevated brain amyloid and poor vascular health. Of these, medial temporal lobe and fornix, reflective of age and pathology-related degeneration, and corpus callosum, reflective of inter-hemispheric disconnection, accounted for 60% of the heterogeneity explained by the most predictive structures. Our results are valuable for identifying cognitively vulnerable individuals and for developing interventions for cognitive decline.
KW - Brain reserve
KW - Cognitive aging
KW - Cognitive heterogeneity
KW - Deep learning
UR - http://www.scopus.com/inward/record.url?scp=85125262412&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85125262412&partnerID=8YFLogxK
U2 - 10.1016/j.neuroimage.2022.119020
DO - 10.1016/j.neuroimage.2022.119020
M3 - Article
C2 - 35196565
AN - SCOPUS:85125262412
SN - 1053-8119
VL - 251
JO - NeuroImage
JF - NeuroImage
M1 - 119020
ER -