Analysis of functional magnetic resonance imaging data using self‐organizing mapping with spatial connectivity

Commonly used methods in analyzing functional magnetic resonance imaging (fMRI) data, such as the Student's t‐test and cross‐correlation analysis, are model‐based approaches. Although these methods are easy to implement and are effective in analyzing data obtained with simple paradigms, they are not applicable in situations in which patterns of neuronal response are complicated and when fMRI response is unknown. In this work, Kohonen's self‐organizing mapping (SOM), which is a model‐free approach, is adapted for analyzing fMRI data. Because spatial connectivity is an important function in the identification of activation sites in functional brain imaging, it is incorporated into the SOM algorithm. Receiver operating characteristic analysis on simulated data shows that the new algorithm achieves measurable improvement over the standard algorithm. The applicability of the new algorithm is demonstrated on experimental data. Magn Reson Med 41:939–946, 1999. © 1999 Wiley‐Liss, Inc.

[1]  Ravi S. Menon,et al.  Functional imaging of human motor cortex at high magnetic field. , 1993, Journal of neurophysiology.

[2]  B. J. Casey,et al.  Activation of the prefrontal cortex in a nonspatial working memory task with functional MRI , 1994, Human brain mapping.

[3]  R. Turner,et al.  Dynamic magnetic resonance imaging of human brain activity during primary sensory stimulation. , 1992, Proceedings of the National Academy of Sciences of the United States of America.

[4]  T H Le,et al.  Methods for assessing accuracy and reliability in functional MRI , 1997, NMR in biomedicine.

[5]  Jack L. Lancaster,et al.  Clustered pixels analysis for functional MRI activation studies of the human brain , 1995 .

[6]  Ravi S. Menon,et al.  Intrinsic signal changes accompanying sensory stimulation: functional brain mapping with magnetic resonance imaging. , 1992, Proceedings of the National Academy of Sciences of the United States of America.

[7]  V M Haughton,et al.  Functional magnetic resonance imaging of somatosensory stimulation. , 1994, Neurosurgery.

[8]  Azriel Rosenfeld,et al.  Blob Detection by Relaxation , 1981, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[9]  Jonathan D. Cohen,et al.  Improved Assessment of Significant Activation in Functional Magnetic Resonance Imaging (fMRI): Use of a Cluster‐Size Threshold , 1995, Magnetic resonance in medicine.

[10]  M. Raichle,et al.  Focal physiological uncoupling of cerebral blood flow and oxidative metabolism during somatosensory stimulation in human subjects. , 1986, Proceedings of the National Academy of Sciences of the United States of America.

[11]  R. S. Hinks,et al.  Time course EPI of human brain function during task activation , 1992, Magnetic resonance in medicine.

[12]  C. Metz.  ROC Methodology in Radiologic Imaging , 1986, Investigative radiology.

[13]  Bernd Fritzke,et al.  Growing cell structures--A self-organizing network for unsupervised and supervised learning , 1994, Neural Networks.

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

[15]  X. Hu,et al.  4 T-fMRI study of nonspatial shifting of selective attention: cerebellar and parietal contributions. , 1998, Journal of neurophysiology.

[16]  D. Tank,et al.  Brain magnetic resonance imaging with contrast dependent on blood oxygenation. , 1990, Proceedings of the National Academy of Sciences of the United States of America.

[17]  Robert A. Lordo,et al.  Learning from Data: Concepts, Theory, and Methods , 2001, Technometrics.

[18]  X. Hu,et al.  Evaluation of the early response in fMRI in individual subjects using short stimulus duration , 1997, Magnetic resonance in medicine.

[19]  P. Boesiger,et al.  Detection of unknown activation patterns by a fuzzy clustering technique and comparison with conventional post processing methods in fMRI , 1996, NeuroImage.

[20]  E C Wong,et al.  Processing strategies for time‐course data sets in functional mri of the human brain , 1993, Magnetic resonance in medicine.

[21]  J. Binder,et al.  Functional magnetic resonance imaging of human auditory cortex , 1994, Annals of neurology.

[22]  G. Radda,et al.  Oxygenation dependence of the transverse relaxation time of water protons in whole blood at high field. , 1982, Biochimica et biophysica acta.

[23]  Karl J. Friston,et al.  Assessing the significance of focal activations using their spatial extent , 1994, Human brain mapping.

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

[25]  G. McCarthy,et al.  Echo-planar magnetic resonance imaging studies of frontal cortex activation during word generation in humans. , 1993, Proceedings of the National Academy of Sciences of the United States of America.

[26]  A. Georgopoulos,et al.  Sequential activity in human motor areas during a delayed cued finger movement task studied by time‐resolved fMRI , 1997, Neuroreport.

[27]  K Ugurbil,et al.  Functional magnetic resonance imaging of Broca's area during internal speech. , 1993, Neuroreport.

[28]  A. Georgopoulos,et al.  Time‐resolved fMRI of mental rotation , 1997, Neuroreport.

[29]  J Xiong,et al.  Assessment and optimization of functional MRI analyses , 1996, Human brain mapping.

[30]  Karl J. Friston,et al.  Analysis of functional MRI time‐series , 1994, Human Brain Mapping.