Medical Imaging for Neurodegenerative Diseases (MIND) | Medical Physics and Complex Systems

Abstract: Until recently the segmentation of the hippocampus, i.e. its identification and separation from surrounding brain structures, was performed mainly manually or with semi-automated techniques, followed by manual editing. This is obviously timeconsuming and subject to investigator variability, so a number of automated segmentation methods have been developed. In this project a fully automated pattern recognition system for accurate and reproducible segmentation of the hippocampus in structural Magnetic Resonance Imaging (MRI) is presented. The system has been validated on T1-weighted structural brain MR images. The proposed approach can be useful for large-scale research studies, in the first instance on Alzheimer’s disease, where the hippocampal volume is an important biomarker, but also on other brain disorders in which the hippocampus plays a relevant pathogenetic role.