Display Settings:

Format

Send to:

Choose Destination
See comment in PubMed Commons below
Psychiatry Res. 2000 Dec 4;100(2):97-126.

Cerebral cortex: a topographic segmentation method using magnetic resonance imaging.

Author information

  • 1Mental Health-Clinical Research Center, Department of Psychiatry, College of Medicine, University of Iowa Hospitals & Clinics, 2911 JPP, 200 Hawkins Drive, Iowa City, IA 52242-1057, USA.

Abstract

Remarkable developments in magnetic resonance imaging (MRI) technology provide a broad range of potential applications to explore in vivo morphological characteristics of the human cerebral cortex. MR-based parcellation methods of the cerebral cortex may clarify the structural anomalies in specific brain subregions that reflect underlying neuropathological processes in brain illnesses. The present study describes detailed guidelines for the parcellation of the cerebral cortex into 41 subregions. Our method conserves the topographic uniqueness of individual brains and is based on our ability to visualize the three orthogonal planes, the triangulated gray matter isosurface and the three-dimensional (3D) rendered brain simultaneously. Based upon topographic landmarks of individual sulci, every subregion was manually segmented on a set of serial coronal or transaxial slices consecutively. The reliability study indicated that the cerebral cortex could be parcelled reliably; intraclass correlation coefficients for each subregion ranged from 0.60 to 0.99. The validity of the method is supported by the fact that gyral subdivisions are similar to regions delineated in functional imaging studies conducted in our center. Ultimately, this method will permit us to detect subtle morphometric impairments or to find abnormal patterns of functional activation in circumscribed cortical subregions. The description of a thorough map of regional structural and functional cortical abnormalities will provide further insight into the role that different subregions play in the pathophysiology of brain illnesses.

PMID:
11114495
[PubMed - indexed for MEDLINE]
PubMed Commons home

PubMed Commons

0 comments
How to join PubMed Commons

    Supplemental Content

    Icon for Elsevier Science
    Loading ...
    Write to the Help Desk