Automated Gyrus Labeling Using Knowledge-based Fuzzy Inference Systems

Automated labeling of the cerebral gyri on the cerebral surface is a fundamental work for estimating the regional atrophy of the cerebrum. Estimating regional brain atrophy will help us to diagnose the cerebral diseases. This article proposes a fully automated method for labeling the cerebral gyri using 3D human brain magnetic resonance (MR) images. The proposed method is composed of two steps; (1) initializing a surface model, and (2) deforming the surface model. They are based on fuzzy pattern matching and fuzzy inference systems based on anatomical knowledge. This article also introduces a knowledge extraction system for constructing the systems. Comparison between the automatically labeled gyri and the manually labeled gyri showed that the automatically labeled gyri overlap with the manually labeled gyri well.