A deep learning framework identifies dimensional representations of Alzheimer’s Disease from brain structure

Zhijian Yang, Ilya M. Nasrallah, Haochang Shou, Junhao Wen, Jimit Doshi, Mohamad Habes, Guray Erus, Ahmed Abdulkadir, Susan M. Resnick, Marilyn S. Albert, Paul Maruff, Jurgen Fripp, John C. Morris, David A. Wolk and Christos Davatzikos ()
Additional contact information
Zhijian Yang: University of Pennsylvania
Ilya M. Nasrallah: University of Pennsylvania
Haochang Shou: University of Pennsylvania
Junhao Wen: University of Pennsylvania
Jimit Doshi: University of Pennsylvania
Mohamad Habes: University of Pennsylvania
Guray Erus: University of Pennsylvania
Ahmed Abdulkadir: University of Pennsylvania
Susan M. Resnick: National Institute on Aging
Marilyn S. Albert: Johns Hopkins University School of Medicine
Paul Maruff: University of Melbourne
Jurgen Fripp: Australian e-Health Research Centre CSIRO
John C. Morris: Washington University in St. Louis
David A. Wolk: University of Pennsylvania
Christos Davatzikos: University of Pennsylvania

Nature Communications, 2021, vol. 12, issue 1, 1-15

Abstract: Abstract Heterogeneity of brain diseases is a challenge for precision diagnosis/prognosis. We describe and validate Smile-GAN (SeMI-supervised cLustEring-Generative Adversarial Network), a semi-supervised deep-clustering method, which examines neuroanatomical heterogeneity contrasted against normal brain structure, to identify disease subtypes through neuroimaging signatures. When applied to regional volumes derived from T1-weighted MRI (two studies; 2,832 participants; 8,146 scans) including cognitively normal individuals and those with cognitive impairment and dementia, Smile-GAN identified four patterns or axes of neurodegeneration. Applying this framework to longitudinal data revealed two distinct progression pathways. Measures of expression of these patterns predicted the pathway and rate of future neurodegeneration. Pattern expression offered complementary performance to amyloid/tau in predicting clinical progression. These deep-learning derived biomarkers offer potential for precision diagnostics and targeted clinical trial recruitment.

Date: 2021
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
https://www.nature.com/articles/s41467-021-26703-z Abstract (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:nat:natcom:v:12:y:2021:i:1:d:10.1038_s41467-021-26703-z

Ordering information: This journal article can be ordered from
https://www.nature.com/ncomms/

DOI: 10.1038/s41467-021-26703-z

Access Statistics for this article

Nature Communications is currently edited by Nathalie Le Bot, Enda Bergin and Fiona Gillespie

More articles in Nature Communications from Nature
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().