SEMI-Supervised learning of brain functional networks

Yuhui Du, Jing Sui, Qingbao Yu, Hao He, Vince D. Calhoun

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Identification of subject-specific brain functional networks of interest is of great importance in fMRI based brain network analysis. In this study, a novel method is proposed to identify subject-specific brain functional networks using a graph theory based semi-supervised learning technique by incorporating not only prior information of the network to be identified as similarly used in seed region based correlation analysis (SCA) but also background information, which leads to robust performance for fMRI data with low signal noise ratio (SNR). Comparison experiments on both simulated and real fMRI data demonstrate that our method is more robust and accurate for identification of known brain functional networks than SCA, blind independent component analysis (ICA), and clustering based methods including Ncut and Kmeans.

Original languageEnglish (US)
Title of host publication2014 IEEE 11th International Symposium on Biomedical Imaging, ISBI 2014
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1-4
Number of pages4
ISBN (Electronic)9781467319591
DOIs
StatePublished - Jul 29 2014
Externally publishedYes
Event2014 IEEE 11th International Symposium on Biomedical Imaging, ISBI 2014 - Beijing, China
Duration: Apr 29 2014May 2 2014

Publication series

Name2014 IEEE 11th International Symposium on Biomedical Imaging, ISBI 2014

Other

Other2014 IEEE 11th International Symposium on Biomedical Imaging, ISBI 2014
Country/TerritoryChina
CityBeijing
Period4/29/145/2/14

Keywords

  • Brain functional network
  • Clustering
  • Independent component analysis
  • Semi-supervised learning

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Fingerprint

Dive into the research topics of 'SEMI-Supervised learning of brain functional networks'. Together they form a unique fingerprint.

Cite this