Exploring the Efficacy of Modified Transfer Learning in Identifying Parkinson's Disease Through Drawn Image Patterns

2025-10-06

AI Summary

This study addresses the clinical bottleneck of costly, invasive procedures for early Parkinson’s disease (PD) diagnosis by proposing a non-invasive, low-cost screening method based on hand-drawn spiral and wave images. We introduce a novel three-stage deep learning framework that integrates a pretrained CNN backbone, learnable custom convolutional layers, and an attention mechanism to enhance discriminative feature learning; additionally, we incorporate data augmentation and ensemble hard voting to improve generalization and robustness. Experimental results show weighted F1-scores of 90.0% on spiral images and 96.7% on wave images, with an ensemble accuracy of 93.3%—significantly outperforming individual models. To our knowledge, this is the first systematic validation of hand-drawn dynamic trajectory images as viable PD biomarkers. The work establishes an interpretable, lightweight, and deployable paradigm for neurodegenerative disease screening.

Technology Category

• Machine Learning: Ensemble Methods
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Learning to Search

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Parkinson's disease (PD) is a progressive neurodegenerative condition characterized by the death of dopaminergic neurons, leading to various movement disorder symptoms. Early diagnosis of PD is crucial to prevent adverse effects, yet traditional diagnostic methods are often cumbersome and costly. In this study, a machine learning-based approach is proposed using hand-drawn spiral and wave images as potential biomarkers for PD detection. Our methodology leverages convolutional neural networks (CNNs), transfer learning, and attention mechanisms to improve model performance and resilience against overfitting. To enhance the diversity and richness of both spiral and wave categories, the training dataset undergoes augmentation to increase the number of images. The proposed architecture comprises three phases: utilizing pre-trained CNNs, incorporating custom convolutional layers, and ensemble voting. Employing hard voting further enhances performance by aggregating predictions from multiple models. Experimental results show promising accuracy rates. For spiral images, weighted average precision, recall, and F1-score are 90%, and for wave images, they are 96.67%. After combining the predictions through ensemble hard voting, the overall accuracy is 93.3%. These findings underscore the potential of machine learning in early PD diagnosis, offering a non-invasive and cost-effective solution to improve patient outcomes.

Problem

• Detecting Parkinson's disease using hand-drawn spiral and wave images
• Improving diagnostic accuracy through modified transfer learning methods
• Developing non-invasive machine learning models for early PD identification

Innovation

• Modified transfer learning with attention mechanisms
• Dataset augmentation for spiral and wave images
• Ensemble voting with pre-trained CNNs and custom layers