Scholarlink

<p>Parkinson’s disease is a progressive disorder that significantly affects motor functions, speech, and cognitive abilities. Studies indicate that speech impairments often emerge in the early stages of the disease, presenting a valuable opportunity for early diagnosis. This research proposes a novel approach by integrating Parrot optimization with a support vector machine to enhance the accuracy of Parkinson’s disease detection. The proposed model was rigorously evaluated using five speech datasets. Its performance was benchmarked against seven parameter optimization methods, including support vector machine based on Particle Swarm Optimization, Harris Hawks Optimization, Hunger Games Search, RIME Optimization, Weighted Mean of Vectors, Runge Kutta Optimization, and Moss Growth Optimization. Additionally, to enhance diagnostic accuracy, a novel feature selection method combining Least Absolute Shrinkage and Selection Operator and Information Gain was applied before the classification model. Results demonstrate the system’s superior classification capabilities, achieving an accuracy of 99%.</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

Parrot optimization-enhanced SVM model for Parkinson’s diagnosis

  • 期刊论文

  • 发表日期：

    2025年9月30日

查看全文
  • Mohamed Elkharadly,
  • Khaled Amin,
  • O. M. Abo-Seida,
  • Mina Ibrahim

摘要

Parkinson’s disease is a progressive disorder that significantly affects motor functions, speech, and cognitive abilities. Studies indicate that speech impairments often emerge in the early stages of the disease, presenting a valuable opportunity for early diagnosis. This research proposes a novel approach by integrating Parrot optimization with a support vector machine to enhance the accuracy of Parkinson’s disease detection. The proposed model was rigorously evaluated using five speech datasets. Its performance was benchmarked against seven parameter optimization methods, including support vector machine based on Particle Swarm Optimization, Harris Hawks Optimization, Hunger Games Search, RIME Optimization, Weighted Mean of Vectors, Runge Kutta Optimization, and Moss Growth Optimization. Additionally, to enhance diagnostic accuracy, a novel feature selection method combining Least Absolute Shrinkage and Selection Operator and Information Gain was applied before the classification model. Results demonstrate the system’s superior classification capabilities, achieving an accuracy of 99%.