Optimizing Multiple Sclerosis Detection through ResNet Feature Fusion and SVM Classifier

Multiple sclerosis (MS) is a chronic inflammatory disease that affects the central nervous system (CNS) and is characterised by inflammation, demyelination, and persistent nerve damage known as sclerosis. Accurate diagnosis and monitoring are challenging due to the ambiguity of MS symptoms, despite their importance. MRI is essential for the diagnosis of MS by identifying and evaluating lesions that demonstrate disease activity. This study introduces hybrid framework that uses high level features of pre-trained CNN models (ResNet18, ResNet50, and ResNet101) to distinguish between MS and control classes, utilising axial and sagittal MRI scans. The F1 score was 90.53%, and the classification accuracy was 88.10%. This was achieved by combining features from ResNet18 and ResNet50 with a SVM classifier. This paradigm represents a significant advancement in the diagnosis and classification of multiple sclerosis, and it has the potential to have a broad impact on neuroinflammatory disorders.