Detection of Neuro Disorder Movement Using Deep Learning

Movement-related neurological disorders often lead to disruptions in motor coordination, which can be visually represented through hand-drawn patterns. This research presents a deep learning-based system designed to detect such neuro-motor irregularities by analyzing spiral and wave drawings. These simple sketches, commonly used in clinical screening, reflect the neuromuscular control of individuals and can indicate the presence of early-stage disorders. The proposed method utilizes Convolutional Neural Networks (CNNs) for automatic feature extraction and classification, eliminating the need for manual analysis or handcrafted features. The model is trained on a curated dataset containing spiral and wave images from both healthy individuals and those with motor dysfunctions. Performance evaluation metrics such as accuracy, precision, recall, and F1-score are used to validate the model’s effectiveness. The results demonstrate that the deep learning approach can successfully distinguish between normal and impaired motor patterns with high reliability. Though Parkinson’s disease is one of the key conditions examined, the model’s architecture is adaptable for broader applications in neuro-disorder diagnosis. This approach paves the way for non- invasive, cost-effective, and scalable diagnostic tools to support early intervention and clinical decision-making.