V-SVM: A VGG19-based Support Vector Machine for Early Detection of Parkinson’s Disease

Parkinson’s disease (PD) is a neurodegenerative disorder characterized by tremors, bradykinesia, and rigidity. Early and accurate diagnosis is crucial for timely intervention and improved patient outcomes. We explore the effectiveness of VGG19, a convolutional neural network architecture consisting of 19 layers, which include 16 convolutional layers and 3 fully connected layers for feature extraction coupled with Support Vector Machines (SVMs) for image classification. Our proposed model leverages the strengths of both techniques: CNNs excel at extracting spatial features from handwritten characters, while SVMs provide robust classification. We compare our approach to other established ML models which include SVM and CNN individually. Our results demonstrate that the VGG19-SVM combination achieves superior performance in identifying PD cases compared to other models, highlighting the potential of this approach for early and reliable PD detection using readily available handwriting data. This hybrid methodology holds promise for enhancing patient outcomes through timely intervention.