Source localization approach for functional DOT using MUSIC and FDR control

In this paper, we formulate diffuse optical tomography (DOT) problems as a source localization problem and propose a MUltiple SIgnal Classification (MUSIC) algorithm for functional brain imaging application. By providing MUSIC spectra for major chromophores such as oxy-hemoglobin (HbO) and deoxy-hem...

Full description

Saved in:
Bibliographic Details
Published in:Optics express 2012-03, Vol.20 (6), p.6267-6285
Main Authors: Jung, Jin Wook, Lee, Ok Kyun, Ye, Jong Chul
Format: Article
Language:English
Subjects:
Algorithms
Brain - physiology
Brain Mapping - methods
Computer Simulation
Humans
Image Interpretation, Computer-Assisted - methods
Models, Neurological
Reproducibility of Results
Sensitivity and Specificity
Tomography, Optical Coherence - methods
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, we formulate diffuse optical tomography (DOT) problems as a source localization problem and propose a MUltiple SIgnal Classification (MUSIC) algorithm for functional brain imaging application. By providing MUSIC spectra for major chromophores such as oxy-hemoglobin (HbO) and deoxy-hemoglobin (HbR), we are able to investigate the spatial distribution of brain activities. Moreover, the false discovery rate (FDR) algorithm can be applied to control the family-wise error in the MUSIC spectra. The minimum distance between the center of mass in DOT and the Montreal Neurological Institute (MNI) coordinates of target regions in experiments was between approximately 6 and 18 mm, and the displacement of the center of mass in DOT and fMRI ranged between 12 and 28 mm, which demonstrate the legitimacy of the DOT-based imaging. The proposed brain mapping method revealed its potential as an alternative algorithm to monitor the brain activation.
ISSN:1094-4087
1094-4087
DOI:10.1364/OE.20.006267