An Intelligent System for Automatic Skull Stripping of Human Brain MRI

Abstract

Skull strapping is a critical step in medical image analysis, involving the segmentation of the brain from the surrounding skull structure. Our work describes the development of an intelligent system that automates the skull strapping process for human brain mag- netic resonance (MR) images. Automating this process has the potential to significantly save time and effort for clinicians and researchers. To achieve accurate and efficient skull strapping, our proposed system trains the Unet model which is CNN architecture on a large dataset of annotated MR images that can effectively learn the intricate features and patterns associated with the skull and brain structures. The obtained results prove the efficiency of the proposed algorithm