ABSTRACT
Magnetic resonance (MR) has become a ubiquitous tool for noninvasive investi-
gation of brain changes in individuals suffering from psychiatric disease. How-
ever, it was only a short time ago that such studies became possible. The crucial
discovery of how to make images from MR occurred in early 1970s (1,2), with the
first human head scanned in 1978, requiring several hours to make a single image
(3). By the middle of the 1980s, technical advances in computing, scanner hard-
ware, and acquisition methodologies, made rapid imaging of the whole brain
feasible. By the late 1980s, the use of magnetic resonance imaging (MRI) as a
clinical and research tool had become widespread for measuring structure vol-
umes, tissue relaxation (described below), and brain chemistry. In the 1990s, blood
oxygenation level-dependent (BOLD) imaging was developed to measure brain
activation (4), a phenomenon known to occur since the turn of the century [for
review see (5)]. Though oxygenation changes could previously be measured by
single photon emission computed tomography (SPECT) or positron emission
tomography (PET) using radioactive tracers (6), the noninvasive nature of
BOLD MRI greatly facilitated the ability to make noninvasive examination
about the “thinking” brain, leading to widespread studies of cognition in health
and disease. In parallel with these improved data acquisition methods, analytic
methods have also made dramatic improvements. Whereas early structural
imaging studies were largely limited to tracing slice-by-slice areas using a
digitizer, multidimensional assessment of many sophisticated image parameters
can be rapidly obtained with modern computing and methods (e.g., regional
cortical volumes, metrics of shape, cortical thickness, activation locations to a
performed task, etc.).
