Leveraging Handwriting Impairment as a Biomarker for Early Parkinson’s Disease Diagnosis

Parkinson’s Disease (PD) is a progressive neurodegenerative condition that significantly impacts motor function, leading to symptoms such as tremors, bradykinesia, and rigidity. However, diagnosing PD in its early stages remains challenging due to the absence of specific biomarkers, often resulting in delayed treatment and symptom management. Handwriting changes serve as a dependable marker of disease severity in PD patients, who exhibit significantly slower and lighter writing or drawing. In this paper, we aim to develop a cost-effective, non-invasive diagnostic tool utilizing handwriting samples, specifically spiral and wave drawings. We introduce a classification model incorporating a convolutional neural network (CNN) to extract relevant features from these drawings and distinguish individuals with Parkinson’s disease from healthy ones. Our study holds promising results, with the VGG-19 model achieving an accuracy of 91.33%, specificity of 90.45%, and sensitivity of 92.6%. These outcomes underscore the potential of our approach to facilitate early PD diagnosis and treatment, ultimately enhancing the quality of life for affected individuals. The use of handwriting as a unique biomarker underlines the importance of leveraging advanced AI techniques for accurate and accessible PD diagnosis.