Functional connectome fingerprinting: Identifying individuals and predicting cognitive function via deep learning

Biao Cai, Gemeng Zhang, Aiying Zhang, Li Xiao, Wenxing Hu, Julia M. Stephen, Tony W. Wilson, Vince D. Calhoun, Yu Ping Wang

Research output: Contribution to journalArticlepeer-review

Abstract

The dynamic characteristics of functional network connectivity have been widely acknowledged and studied. Both shared and unique information has been show to be present in the connectomes. However, very little has been known about whether and how this common pattern can predict the individual variability of the brain, i.e.”brain fingerprinting”, which attempts to reliably identify a particular individual from a pool of subjects. In this paper, we propose to enhance the individual uniqueness based on an autoencoder network. More specifically, we rely on the hypothesis that the common neural activities shared across individuals may lessen the individual discrimination. By reducing contributions from shared activities, inter-subject variability can be enhanced. Results show that that refined connectomes utilizing an autoencoder with sparse dictionary learning can successfully distinguish one individual from the remaining participants with reasonably high accuracy (up to 99.5% for the rest-rest pair). Furthermore, high-level cognitive behavior (e.g., fluid intelligence, executive function, and language comprehension) can also be better predicted using the refined functional connectivity profiles. As expected, high-order association cortices contributed more to both individual discrimination and behavior prediction. The proposed approach provides a promising way to enhance and leverage the individualized characteristics of brain networks.

Original languageEnglish (US)
JournalUnknown Journal
StatePublished - Jun 17 2020
Externally publishedYes

Keywords

  • Autoencoder network
  • Common connectivity patterns
  • High-level cognition prediction
  • Index Terms—Functional connectivity
  • Individual identification
  • Refined connectomes

ASJC Scopus subject areas

  • General

Fingerprint Dive into the research topics of 'Functional connectome fingerprinting: Identifying individuals and predicting cognitive function via deep learning'. Together they form a unique fingerprint.

Cite this