Document Detail

Dimensionality estimation for optimal detection of functional networks in BOLD fMRI data.
MedLine Citation:
PMID:  20858546     Owner:  NLM     Status:  MEDLINE    
Estimation of the intrinsic dimensionality of fMRI data is an important part of data analysis that helps to separate the signal of interest from noise. We have studied multiple methods of dimensionality estimation proposed in the literature and used these estimates to select a subset of principal components that was subsequently processed by linear discriminant analysis (LDA). Using simulated multivariate Gaussian data, we show that the dimensionality that optimizes signal detection (in terms of the receiver operating characteristic (ROC) metric) goes through a transition from many dimensions to a single dimension as a function of the signal-to-noise ratio. This transition happens when the loci of activation are organized into a spatial network and the variance of the networked, task-related signals is high enough for the signal to be easily detected in the data. We show that reproducibility of activation maps is a metric that captures this switch in intrinsic dimensionality. Except for reproducibility, all of the methods of dimensionality estimation we considered failed to capture this transition: optimization of Bayesian evidence, minimum description length, supervised and unsupervised LDA prediction, and Stein's unbiased risk estimator. This failure results in sub-optimal ROC performance of LDA in the presence of a spatially distributed network, and may have caused LDA to underperform in many of the reported comparisons in the literature. Using real fMRI data sets, including multi-subject group and within-subject longitudinal analysis we demonstrate the existence of these dimensionality transitions in real data.
Grigori Yourganov; Xu Chen; Ana S Lukic; Cheryl L Grady; Steven L Small; Miles N Wernick; Stephen C Strother
Related Documents :
12549736 - Instantaneous parameter estimation in cardiovascular time series by harmonic and time-f...
12007986 - The use of a neural network system in the identification of motor unit characteristics ...
16521026 - Singular perturbation analysis of camp signalling in dictyostelium discoideum aggregates.
21624376 - A generalization of hamilton's rule-love thy sibling how much?
3723446 - Assessment of compliance in hemodialysis adaptation.
20886036 - Disentangling ancient interactions: a new extinct passerine provides insights on charac...
Publication Detail:
Type:  Journal Article; Research Support, N.I.H., Extramural; Research Support, Non-U.S. Gov't     Date:  2010-09-19
Journal Detail:
Title:  NeuroImage     Volume:  56     ISSN:  1095-9572     ISO Abbreviation:  Neuroimage     Publication Date:  2011 May 
Date Detail:
Created Date:  2011-04-25     Completed Date:  2011-08-15     Revised Date:  2014-09-22    
Medline Journal Info:
Nlm Unique ID:  9215515     Medline TA:  Neuroimage     Country:  United States    
Other Details:
Languages:  eng     Pagination:  531-43     Citation Subset:  IM    
Copyright Information:
Copyright © 2010 Elsevier Inc. All rights reserved.
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Area Under Curve
Brain / physiology*
Brain Mapping / methods*
Image Processing, Computer-Assisted / methods*
Magnetic Resonance Imaging*
Middle Aged
Nerve Net / physiology*
ROC Curve
Reproducibility of Results
Young Adult
Grant Support
073204//PHS HHS; MOP14036//Canadian Institutes of Health Research; MOP84483//Canadian Institutes of Health Research; R43 MH073204/MH/NIMH NIH HHS; R43 MH073204-01/MH/NIMH NIH HHS

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine

Previous Document:  High-speed vascular dynamics of the hemodynamic response.
Next Document:  Phylogeny and biogeography of the Asian trogons (Aves: Trogoniformes) inferred from nuclear and mito...