Scholarlink

<p>In recent years, progress monitoring and diagnosis devices for evaluation of Parkinson’s disease (PD) are becoming important. The early detection and rehabilitation of PD can improve the treatment consistency of patients from which may allow a fast decision of diagnosis. Broadly, the whole work is categorized in two sections, one is feature extraction and selection. This first section depicts how feature selection techniques can affects the accuracy of individual model. The second is early detection of disorder. For it, ANFIS + GWO model is proposed for early detection of Parkinson’s disease where parameters of ANFIS model are adjusted by exploiting grey wolf optimization (GWO). Different optimization methods are compared using different evaluation metrices like MSE, accuracy, computation time, iterations and number of epochs using a gait dataset of 166 people from Physionet repository. It demonstrate the superiority of ANFIS + GWO model against neural network (NN), adaptive neuro-fuzzy inference system (ANFIS), hybridization of ANFIS with genetic algorithm (GA), and partical swarm optimization (PSO). Here, maximum accuracy is attained ANFIS + GWO i.e. 99.8% that is compared with other optimization techniques used in recent literature for prediction of PD.</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

Optimized Learning Framework Approaches for Early Detection and Classification of Parkinson’s Disease

  • 期刊论文

  • 发表日期：

    2025年11月19日

查看全文
  • Pooja,
  • Karan Veer,
  • S. K. Pahuja

摘要

In recent years, progress monitoring and diagnosis devices for evaluation of Parkinson’s disease (PD) are becoming important. The early detection and rehabilitation of PD can improve the treatment consistency of patients from which may allow a fast decision of diagnosis. Broadly, the whole work is categorized in two sections, one is feature extraction and selection. This first section depicts how feature selection techniques can affects the accuracy of individual model. The second is early detection of disorder. For it, ANFIS + GWO model is proposed for early detection of Parkinson’s disease where parameters of ANFIS model are adjusted by exploiting grey wolf optimization (GWO). Different optimization methods are compared using different evaluation metrices like MSE, accuracy, computation time, iterations and number of epochs using a gait dataset of 166 people from Physionet repository. It demonstrate the superiority of ANFIS + GWO model against neural network (NN), adaptive neuro-fuzzy inference system (ANFIS), hybridization of ANFIS with genetic algorithm (GA), and partical swarm optimization (PSO). Here, maximum accuracy is attained ANFIS + GWO i.e. 99.8% that is compared with other optimization techniques used in recent literature for prediction of PD.