High-order concept discovery in functional brain imaging

Michael Barnathan, Vasileios Megalooikonomou, Christos Faloutsos, Feroze B. Mohamed, Scott Faro

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

Abstract

Many spatiotemporal medical image datasets exhibit "high-order" structure, in which many independent variables exist (e.g. space and time) or features are not scalar at all. We analyze these datasets as tensors (high-order generalizations of matrices), preprocessing our dataset using wavelets to improve efficiency and performing latent concept discovery using parallel factor analysis. Both our method and naive tensor approaches discovered concepts representing handedness in an 11 subject motor task fMRI dataset. However, our method compressed the dataset by 98% and completed in 2 hours vs. 8 days, suggesting that a wavelet and tensor approach gains the benefits of high-order analysis while preserving the efficiency of low-order techniques.

Original languageEnglish (US)
Title of host publication2010 7th IEEE International Symposium on Biomedical Imaging
Subtitle of host publicationFrom Nano to Macro, ISBI 2010 - Proceedings
Pages664-667
Number of pages4
DOIs
StatePublished - 2010
Externally publishedYes
Event7th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2010 - Rotterdam, Netherlands
Duration: Apr 14 2010Apr 17 2010

Publication series

Name2010 7th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2010 - Proceedings

Other

Other7th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2010
Country/TerritoryNetherlands
CityRotterdam
Period4/14/104/17/10

Keywords

  • Concept discovery
  • Latent semantic analysis
  • Parallel factor analysis
  • Tensors
  • Wavelets
  • fMRI

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Fingerprint

Dive into the research topics of 'High-order concept discovery in functional brain imaging'. Together they form a unique fingerprint.

Cite this