Texture Analysis of Reaction Diffusion Level Set Evolution Of Multiple Sclerosis Lesions In Brain MR Images

Multiple sclerosis (MS) is a chronic idiopathic disease that results in multiple areas of inflammatory demyelination in the Human Central Nervous System (CNS) and it causes various troubles with mobility and upper limb function, bladder, vision, speech, swelling, and cognition. Hence, we need early detection and accurate identification of pathological changes of disease progression for MS patients. In this work, automated segmentation method of Reaction-Diffusion Level Set Evolution (RDLSE) is employed for T2-Weighted (T2W) Magnetic Resonance images. Axial view of MR images of Various MS Patients based on Extended Disability Status Scale (EDSS) of the University Medical Centre Ljubljana are used in this analysis. The segmented output image accuracy is validated with Ground Truth images. Texture analyses are employed to Segmented MS region in MR images to improve the accuracy and identification of differences in brain tissue structure. Results show RDLSE methods are able to segment the small MS lesions even in the presence of heterogeneous intensity values and segmented output image feature value show the MS lesion load for various EDSS values in MS patient T2W MR Images.