Abstract
Detecting brain structural changes from magnetic resonance (MR) images can facilitate early diagnosis and treatment of Multiple Sclerosis (MS), a neurodegenerative disease of central nervous system. We develop a novel volume and shape of hippocampus-based feature with Support Vector Machine (SVM) to detect brain structural changes as potential biomarkers. This approach requires pre-processing which is influenced by artefacts such as image distortion. We represent hippocampus segmentation based on watershed bottom hat filtering algorithm and morphological operations. The extracted features are used as criteria to categorise image features into two classes, i.e. healthy and patient. Only healthy and patient features are used to predict the disease status of new brain images. The results proved that the proposed architecture has high contribute to computer-aided diagnosis of MS. Our empirical evaluation has a superior retrieval and diagnosis performance when compared to the performance of other works.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call