Standard Article

Magnetic Resonance Imaging, Functional

Biomedical Spectroscopy

  1. Wolfgang Richter

Published Online: 15 SEP 2006

DOI: 10.1002/9780470027318.a0109

Encyclopedia of Analytical Chemistry

Encyclopedia of Analytical Chemistry

How to Cite

Richter, W. 2006. Magnetic Resonance Imaging, Functional. Encyclopedia of Analytical Chemistry. .

Author Information

  1. National Research Council, Winnipeg, Canada

Publication History

  1. Published Online: 15 SEP 2006


Functional magnetic resonance imaging (FMRI) is an analytical method for measuring brain activity while it occurs. FMRI was first demonstrated in 1992, but it has since become the most popular neuroimaging method. Its temporal resolution is of the order of seconds and hence superior to positron emission tomography (PET). Its spatial resolution is on the order of millimeters which makes it superior to both PET and electrophysiological methods such as electroencephalography (EEG). Furthermore, FMRI is noninvasive in the sense that no external contrast agent has to be used. FMRI contrast is based on the intrinisic blood oxygenation changes that occur at the site of brain activity in response to a specific task. The exact mechanism that links activity and signal change is currently not well understood and is an area of active research.

FMRI is subject to many experimental difficulties, however. A vexing problem is that of physiological (heartbeat and breathing) and gross motion. Gross motion is often coupled to the presentation of the stimulus and hence especially prone to producing artefactual activation. The analysis of the experimental data is not a standard procedure at present. While past research has generally used paradigmatic methods of analysis (hypothesis testing), nonparadigmatic (data driven) methods like fuzzy clustering analysis (FCA) or independent component analysis (ICA) have become important tools.

A more complete understanding of the physiological mechanisms leading to the activation signal, and a better grasp of the proper statistical treatment of the data, are likely to increase the power of FMRI even further.