Myths and Truths in functional MRI: A Basic Guide for Practitioners

Abstract

Since the first successful demonstration of primary visual cortex “activation” by Sokoloff in 1961, activation studies based on regional blood flow alteration became one of the gold standards for neuroscientific investigation. The rapid advancement of non-invasive imaging techniques provided an appropriate stage for the human application of this original method. Following the tremendous success of positron emission tomography (PET) utilizing O¹⁵ labeled H₂O (H₂O¹⁵-PET), the magnetic resonance imaging (MRI) version, which is now referred to as functional MRI (fMRI), was introduced. fMRI is the method created based on the empirical observation that MRI exhibits spontaneous signal alteration associated with blood flow changes. It is now believed that the principle of fMRI is identical to H₂O¹⁵-PET, namely, that it is blood flow based in accordance with the well-known Munro-Kellie doctrine which predicts a decline in cerebral venous blood volume secondary to an increase in cerebral arterial blood volume. The method is inherently qualitative and does not provide quantitative information regarding flow alteration. With the growing use of fMRI, however, exponentially increasing well founded criticism questioning the validity of fMRI data has been raised. The majority of validation issues arise during the process of obtaining raw images, which eventually provides raw numerical data for post-processing statistical analysis. This article provides a distillation of the essential knowledge necessary for the non-MR physicist fMRI investigator.