Scholarlink

<p>Parkinson’s disease (PD) represents a significant and progressive neurodegenerative challenge, primarily affecting motor control and speech capabilities. Early and accurate detection is crucial for effective intervention and management. Vocal analysis offers a promising, non-invasive diagnostic approach, as vocal impairments often manifest in early PD stages. This study introduces an innovative approach, harnessing Dragonfly Optimization (DO) to optimize a Support Vector Machine (SVM) model, thus improving PD detection. The DO-SVM model was validated on four speech datasets. To thoroughly assess the efficacy of the proposed model, DO-SVM was compared against six parameter optimization methods, including SVM based on Particle Swarm Optimization (PSO), Grey Wolf Optimization (GWO), Salp Swarm Algorithm (SSA), Sine Cosine Algorithm (SCA), Multi-Verse Optimization (MVO), and Henry Gas Solubility Optimization (HGSO). Furthermore, to improve diagnostic accuracy, a hybrid feature selection method based on Principal Component Analysis (PCA) and Information Gain (IG) was employed before the DO-SVM method, therefore the PCAIG-DO-SVM was proposed. The results confirm the superior effectiveness of the proposed approach, with a classification accuracy of 96.47% obtained via ten-fold cross-validation.</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

Dragonfly optimization boosted support vector machine for Parkinson’s diagnosis

  • 期刊论文

  • 发表日期：

    2026年6月17日

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

摘要

Parkinson’s disease (PD) represents a significant and progressive neurodegenerative challenge, primarily affecting motor control and speech capabilities. Early and accurate detection is crucial for effective intervention and management. Vocal analysis offers a promising, non-invasive diagnostic approach, as vocal impairments often manifest in early PD stages. This study introduces an innovative approach, harnessing Dragonfly Optimization (DO) to optimize a Support Vector Machine (SVM) model, thus improving PD detection. The DO-SVM model was validated on four speech datasets. To thoroughly assess the efficacy of the proposed model, DO-SVM was compared against six parameter optimization methods, including SVM based on Particle Swarm Optimization (PSO), Grey Wolf Optimization (GWO), Salp Swarm Algorithm (SSA), Sine Cosine Algorithm (SCA), Multi-Verse Optimization (MVO), and Henry Gas Solubility Optimization (HGSO). Furthermore, to improve diagnostic accuracy, a hybrid feature selection method based on Principal Component Analysis (PCA) and Information Gain (IG) was employed before the DO-SVM method, therefore the PCAIG-DO-SVM was proposed. The results confirm the superior effectiveness of the proposed approach, with a classification accuracy of 96.47% obtained via ten-fold cross-validation.