Related Experiment Video
Updated: Jul 20, 2026
Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images
Published on: January 7, 2019
Automatic volumetric measurement of segmented brain structures on magnetic resonance imaging
Kazunari Ishii1, Tsutomu Soma, Atsushi K Kono
1Department of Radiology and Nuclear Medicine, Hyogo Brain and Heart Center, 520 Saisho-Ko, Himeji, Hyogo 670-0981, Japan. kishii@hbhc.jp
Purpose:
The purpose of this study was to produce a fully automatic volumetric segmented brain image system (AVSIS) and to evaluate its utility for measuring hippocampal volumes and total intracranial volume (TIV).
Materials And Methods:
We developed a combination technique comprising an anatomical standardization technique to measure TIV, whole brain volume (WBV), and hippocampal volume obtained by magnetic resonance (MR) imaging. Altogether, 15 healthy volunteers and 15 patients with Alzheimer's disease (AD) underwent three-dimensional spoiled gradient echo (3D-SPGR) imaging. Three measurements were performed by manual volumetry as the gold standard, a previous semiautomatic method, and our new method, AVSIS.
Results:
WBV and hippocampal volume in the AD group were significantly smaller than those in the healthy volunteer group measured by the semiautomatic method, manual method, and AVSIS. Each volume measured by AVSIS or semiautomatic method correlated with that measured by the manual method. The correlation coefficients between TIVs, WBVs, or hippocampal volumes measured by AVSIS and the manual method were 0.910, 0.902, and 0.918, respectively; the correlation coefficients between TIVs, or WBVs, measured by the previous semiautomatic method and the manual method were 0.875, and 0.886, respectively.
Conclusion:
We developed a system for a fully automatic measurement of segmented brain structures and obtained good results.
