American Scientist

To the Editors:

I enjoyed reading the recent article entitled "The Structure of the Human Brain" (May–June), by John S. Allen, Joel Bruss and Hanna Damasio, in which the important role of magnetic resonance imaging (MRI) for determining brain structure is emphasized. Unfortunately, the brief paragraph describing the basic physics of MRI is incorrect and incomplete.

Indeed, a powerful magnet is used to align the "spins," in this case hydrogen protons. This magnet is, however, never "turned off" in order to let the nuclei "fall back to a normal state," as stated in the article. Rather, the spins are manipulated by applying a very small radio-frequency (RF) field oscillating in a plane perpendicular to the strong static field. When a brief pulse of the small oscillating field is turned off, it leaves the spins in an excited state, from which they return to equilibrium, releasing energy in the form of RF waves.

A further stated misconception is that the "frequency of these waves provides a measure of the local hydrogen concentration, which varies according to tissue type, such as bone or fat." Although there is a small frequency difference between protons within fat and water molecules, this chemical shift is of no consequence in brain imaging per se, where all the relevant signal arises from water protons.

Finally, it should be noted that the hydrogen concentration itself is only one factor contributing to the rich image contrast exploited by researchers using MRI to study the structure of the brain, other critical factors being the water proton relaxation times.

Robert V. Mulkern
Children’s Hospital
Harvard Medical School