Diffusion map for clustering fMRI spatial maps extracted by Indipendent Component Analysis

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperConference articleResearchpeer-review

  • Tuomo Sipola, Univ Jyvaskyla, University of Jyvaskyla, Dept Math Informat Technol
  • ,
  • Fengyu Cong, Univ Jyvaskyla, University of Jyvaskyla, Dept Math Informat Technol
  • ,
  • Tapani Ristaniemi, Univ Jyvaskyla, University of Jyvaskyla, Dept Math Informat Technol
  • ,
  • Vinoo Alluri, Univ Jyvaskyla, University of Jyvaskyla, Dept Math Informat Technol
  • ,
  • Petri Toiviainen
  • ,
  • Elvira Brattico
  • Asoke K. Nandi, Univ Jyvaskyla, University of Jyvaskyla, Dept Math Informat Technol

Functional magnetic resonance imaging (fMRI) produces data about activity inside the brain, from which spatial maps can be extracted by independent component analysis (ICA). In datasets, there are n spatial maps that contain p voxels. The number of voxels is very high compared to the number of analyzed spatial maps. Clustering of the spatial maps is usually based on correlation matrices. This usually works well, although such a similarity matrix inherently can explain only a certain amount of the total variance contained in the high-dimensional data where n is relatively small but p is large. For high-dimensional space, it is reasonable to perform dimensionality reduction before clustering. In this research, we used the recently developed diffusion map for dimensionality reduction in conjunction with spectral clustering. This research revealed that the diffusion map based clustering worked as well as the more traditional methods, and produced more compact clusters when needed.

Original languageEnglish
JournalIEEE International Workshop on Machine Learning for Signal Processing
Number of pages6
ISSN2161-0363
DOIs
Publication statusPublished - 2013
Event23rd IEEE International Workshop on Machine Learning for Signal Processing (MLSP) - , Denmark
Duration: 22 Sep 201325 Sep 2013

Conference

Conference23rd IEEE International Workshop on Machine Learning for Signal Processing (MLSP)
CountryDenmark
Period22/09/201325/09/2013

    Research areas

  • clustering, diffusion map, dimensionality reduction, functional magnetic resonance imaging (fMRI), independent component analysis, spatial maps, DYNAMICAL-SYSTEMS

See relations at Aarhus University Citationformats

ID: 90546893