Automatic segmentation of gray matter multiple sclerosis lesions on DIR images

Although Multiple Sclerosis (MS) is traditionally considered a chronic inflammatory-demyelinating disease of the white matter (WM) of central nervous system, several studies have recently demonstrated that a large amount of the distinctive MS lesions are located in the grey matter (GM) and especially in the cortex [1,2]. Albeit such GM lesions have been therefore demonstrated to play a major role in the physical and cognitive disability and in the disease progression [3,4], their manually detection is usually time consuming, error prone and operator dependent. We present a technique to automatically segment gray matter lesions on double inversion recovery (DIR) sequences.