Scholarlink

<p>Parkinson’s disease (PD) is a progressive neurodegenerative disorder, and early diagnosis remains a challenge for clinicians. While current methods can be complex, handwriting analysis offers a non-invasive, cost-effective, and highly accessible complementary tool for early detection. This paper introduces a novel system for early PD detection from offline hand-drawings, which is the first to leverage a combined SqueezeNet and TinySiamese network architecture. Our system is designed for minimal complexity and fast matching, which makes it well-suited for deployment in low-resource settings. We evaluated its performance on our newly created Iraqi hand-drawing dataset, publicly accessible at <a href="https://ieee-dataport.org/documents/online-offline-iraquian-hand-drawing-dataset-early-parkinsons-disease-detection">https://ieee-dataport.org/documents/online-offline-iraquian-hand-drawing-dataset-early-parkinsons-disease-detection</a>, comprising data from 30 healthy individuals and 30 PD patients. The system’s classification performance was rigorously evaluated using standard metrics such as accuracy, precision, recall, and F1 score. Our results demonstrate high performance on our dataset, with the system achieving up to 100% accuracy on spiral drawing tasks. Our system significantly improves upon a standalone SqueezeNet model, as evidenced by the increase in accuracy from 63.33% to 87.77% on the digit-writing task. Additionally, our system consistently outperforms other state-of-the-art methods, achieving an accuracy of 90.00% on both the spiral and wave drawing tasks in the Spiral-Wave dataset. These findings establish a new benchmark for automated, interpretable Parkinson’s screening.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Early Parkinson’s disease detection from offline hand-drawing based on SqueezeNet and TinySiamese network

  • 期刊论文

  • 发表日期：

    2026年3月16日

查看全文
  • Mohammed F. Allebawi,
  • Thameur Dhieb,
  • Islem Jarraya,
  • Mohamed Neji,
  • Nouha Farhat,
  • Khadija Moalla,
  • Tarek M. Hamdani,
  • Mariem Damak,
  • Chokri Mhiri,
  • Adel M. Alimi

摘要

Parkinson’s disease (PD) is a progressive neurodegenerative disorder, and early diagnosis remains a challenge for clinicians. While current methods can be complex, handwriting analysis offers a non-invasive, cost-effective, and highly accessible complementary tool for early detection. This paper introduces a novel system for early PD detection from offline hand-drawings, which is the first to leverage a combined SqueezeNet and TinySiamese network architecture. Our system is designed for minimal complexity and fast matching, which makes it well-suited for deployment in low-resource settings. We evaluated its performance on our newly created Iraqi hand-drawing dataset, publicly accessible at https://ieee-dataport.org/documents/online-offline-iraquian-hand-drawing-dataset-early-parkinsons-disease-detection, comprising data from 30 healthy individuals and 30 PD patients. The system’s classification performance was rigorously evaluated using standard metrics such as accuracy, precision, recall, and F1 score. Our results demonstrate high performance on our dataset, with the system achieving up to 100% accuracy on spiral drawing tasks. Our system significantly improves upon a standalone SqueezeNet model, as evidenced by the increase in accuracy from 63.33% to 87.77% on the digit-writing task. Additionally, our system consistently outperforms other state-of-the-art methods, achieving an accuracy of 90.00% on both the spiral and wave drawing tasks in the Spiral-Wave dataset. These findings establish a new benchmark for automated, interpretable Parkinson’s screening.