SpiralDrawNet: CNN-Driven Early Detection of Parkinson's Disease using Spiral Drawings for Accessible Healthcare Solution

Millions of people worldwide suffer with Parkinson's disease (PD), a neurological condition. For PD to be effectively managed and treated, early identification is essential. This study investigates the use of convolutional neural networks (CNNs) in combination with spiral drawings, a straightforward and non-invasive technique, for Parkinson's disease identification. Spiral drawings from both PD patients and non-PD patients make up the research dataset. Digital images are created by processing these sketches. These spiral designs are used to create and train a CNN architecture that divides them into PD and non-PD categories. To increase the model's resilience, a variety of data augmentation strategies are used. Experimental results demonstrate promising performance in PD detection, with the CNN achieving high accuracy, sensitivity, and specificity. The study highlights the potential of utilizing spiral drawings as a cost-effective and accessible tool for early PD screening. Additionally, the CNN-based approach shows the feasibility of automated PD diagnosis, offering a scalable solution for healthcare professionals.