Scholarlink

This research introduces a multimodal method for detecting Parkinson’s disease by analyzing handwriting. We incorporate Dynamic Time Warping (DTW), Convolutional Neural Networks (CNNs), and Random Forest classification to evaluate spiral drawing patterns from both healthy individuals and those with Parkinson’s. DTW measures handwriting differences by comparing sequences of contours, while CNNs identify deep features from the images. The DTW-derived distances and the probability scores from CNNs are combined into a Random Forest classifier to enhance predictive accuracy. Experimental findings indicate that this integrated approach effectively differentiates handwriting characteristics of Parkinson’s patients, achieving high accuracy. This study underscores the potential for merging statistical methods and deep learning approaches for the early diagnosis of Parkinson’s disease through non-invasive handwriting examination.

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

Integrating Temporal and Spatial Features for Parkinson’s Disease Diagnosis from Handwriting Images

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Tania Dutta,
  • Srinjoy Chanda,
  • Daizy Deb

摘要

This research introduces a multimodal method for detecting Parkinson’s disease by analyzing handwriting. We incorporate Dynamic Time Warping (DTW), Convolutional Neural Networks (CNNs), and Random Forest classification to evaluate spiral drawing patterns from both healthy individuals and those with Parkinson’s. DTW measures handwriting differences by comparing sequences of contours, while CNNs identify deep features from the images. The DTW-derived distances and the probability scores from CNNs are combined into a Random Forest classifier to enhance predictive accuracy. Experimental findings indicate that this integrated approach effectively differentiates handwriting characteristics of Parkinson’s patients, achieving high accuracy. This study underscores the potential for merging statistical methods and deep learning approaches for the early diagnosis of Parkinson’s disease through non-invasive handwriting examination.