Parkinson's disease: Improved diagnosis using image processing

Abstract

An intensity-based texture segmentation approach for the detection of regions with abnormal texture characteristics in magnetic resonance imaging is presented. Our algorithm is tested over several images taken from The Parkinson's Progression Markers Initiative (PPMI-database), and the results suggest that this approach is suitable for the successful identification and extraction of regions of interest whose properties can be potentially related to signature features of Parkinson disease.