Detecting low‐frequency functional connectivity in fMRI using a self‐organizing map (SOM) algorithm

Low‐frequency oscillations (<0.08 Hz) have been detected in functional MRI studies, and appear to be synchronized between functionally related areas. A current challenge is to detect these patterns without using an external reference. Self‐organizing maps (SOMs) offer a way to automatically group data without requiring a user‐biased reference function or region of interest. Resting state functional MRI data was classified using a self‐organizing map (SOM). Functional connectivity between the left and right motor cortices was detected in five subjects, and was comparable to results from a reference‐based approach. SOMs are shown to be an attractive option in detecting functional connectivity using a model‐free approach. Hum. Brain Mapping 20:220–226, 2003. © 2003 Wiley‐Liss, Inc.

[1]  Karl J. Friston,et al.  Functional Connectivity: The Principal-Component Analysis of Large (PET) Data Sets , 1993, Journal of cerebral blood flow and metabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism.

[2]  B. Biswal,et al.  Functional connectivity in the motor cortex of resting human brain using echo‐planar mri , 1995, Magnetic resonance in medicine.

[3]  X Hu,et al.  Retrospective estimation and correction of physiological fluctuation in functional MRI , 1995, Magnetic resonance in medicine.

[4]  M. Lowe,et al.  Functional Connectivity in Single and Multislice Echoplanar Imaging Using Resting-State Fluctuations , 1998, NeuroImage.

[5]  Scott T. Grafton,et al.  Automated image registration: I. General methods and intrasubject, intramodality validation. , 1998, Journal of computer assisted tomography.

[6]  P. Boesiger,et al.  A new correlation‐based fuzzy logic clustering algorithm for FMRI , 1998, Magnetic resonance in medicine.

[7]  X Hu,et al.  Analysis of functional magnetic resonance imaging data using self‐organizing mapping with spatial connectivity , 1999, Magnetic resonance in medicine.

[8]  J Hennig,et al.  Neural network‐based analysis of MR time series , 1999, Magnetic resonance in medicine.

[9]  D. P. Russell,et al.  Treatment of baseline drifts in fMRI time series analysis. , 1999, Journal of computer assisted tomography.

[10]  V. Haughton,et al.  Mapping functionally related regions of brain with functional connectivity MR imaging. , 2000, AJNR. American journal of neuroradiology.

[11]  B. Biswal,et al.  Cocaine administration decreases functional connectivity in human primary visual and motor cortex as detected by functional MRI , 2000, Magnetic resonance in medicine.

[12]  T. Lund,et al.  Physiological noise reduction in fMRI using vessel time-series as covariates in a general linear model , 2001, NeuroImage.

[13]  J. Lurito,et al.  Multiple sclerosis: low-frequency temporal blood oxygen level-dependent fluctuations indicate reduced functional connectivity initial results. , 2002, Radiology.

[14]  P. Skudlarski,et al.  Detection of functional connectivity using temporal correlations in MR images , 2002, Human brain mapping.

[15]  S. J. Peltier,et al.  T2 * Dependence of Low Frequency Functional Connectivity , 2002, NeuroImage.