Manuscript received June 20, 2025; revised July 15, 2025; accepted September 3, 2025; published 16, 2026.

Abstract—This paper presents the design of a pre-diagnostic system for Parkinson’s Disease (PD) based on the analysis of hand-drawn spiral patterns using machine learning techniques. Parkinson’s disease is a progressive neurodegenerative disorder whose early motor manifestations—such as micrographia and tremors—can be reflected in fine motor tasks like handwriting. Although handwriting disturbances are not part of the core diagnostic criteria, they are frequently observed in early stages and are recognized by the Movement Disorder Society as supportive markers. In this study, grayscale spiral drawings are preprocessed and binarized using Otsu’s thresholding method. From each image, 100 equidistant pixel coordinates are extracted where the trace is present, forming structured feature vectors. These coordinates are then used to train and evaluate several machine learning classifiers, including Random Forest, k-Nearest Neighbors, and Support Vector Machines. The proposed method prioritizes simplicity, explainability, and computational efficiency, offering a lightweight yet effective tool for early Parkinson’s detection. Experimental results demonstrate the model’s potential as a clinically relevant and accessible diagnostic support system.

Keywords—Parkinson disease, spiral drawing, random forest, Otsu’s thresholding, feature extraction

Cite: Axell Albano Gutiérrez Ramírez, Antonio Alarcón Paredes, and Cornelio Yáñez Márquez, "Detection of Parkinson’s Disease Using Spiral Drawings and Machine Learning Approaches," Journal of Image and Graphics, Vol. 14, No. 1, pp. 58-64, 2026.

Copyright © 2026 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC-BY-4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.