Scholarlink

<p>Epilepsy is one of the world’s most common neurological diseases. The study of seizure prediction holds significant importance in enhancing the quality of life for individuals afflicted with epilepsy and has garnered increasing attentio n in recent times. Traditionally, Patients’ eyes are visualized by brain scientists and medical seizure monitors the EEG of people who have had a seizure to detect epileptic activity. However, this traditional detection method is complex, time-consuming and difficult to achieve better accuracy with reduced time complexity. To overcome these difficulties, a Radial Basis Function neural network is proposed to recognize epileptic seizure patients from healthy subjects. The EEG brain signal was first considered as input, then pre-processed the input signal using a fractional bandpass filter and Morphological Component Analysis (MCA) to eliminate unwanted noise from the input signal. The performance of this filter is improved by means of optimal selection of fractional orders such as <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\:\alpha\:\)</EquationSource> </InlineEquation> and <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(\:\beta\:\)</EquationSource> </InlineEquation> in the range of 0 to 1. In order to select the optimal fractional order value dingo optimization algorithm is used. After pre-processing, features are extracted using discrete wavelet transform (DWT) and empirical mode decomposition (EMD) in order to calculate the feature coefficients. The features are then fed into the proposed Radial Basis Function Neural Network (RBFNN) classifier as an input. Finally, the classifier predicts whether the epilepsy disease affected the patients or not. According to the simulation analysis, the proposed method achieves 98% precision, 2% error, 98% accuracy, 97% specificity and so on. It shows that the proposed approach outperforms previous methods. According to this proposed model, the prognosis for epileptic disease may be improved, leading to a significant enhancement of the quality lives for epileptic patients.</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

Epileptic patient prediction using Dingo optimization based fractional band pass filter and radial basis function neural network

  • 期刊论文

  • 发表日期：

    2026年3月9日

查看全文
  • Neeraj Nagwanshi,
  • Anjali Potnis

摘要

Epilepsy is one of the world’s most common neurological diseases. The study of seizure prediction holds significant importance in enhancing the quality of life for individuals afflicted with epilepsy and has garnered increasing attentio n in recent times. Traditionally, Patients’ eyes are visualized by brain scientists and medical seizure monitors the EEG of people who have had a seizure to detect epileptic activity. However, this traditional detection method is complex, time-consuming and difficult to achieve better accuracy with reduced time complexity. To overcome these difficulties, a Radial Basis Function neural network is proposed to recognize epileptic seizure patients from healthy subjects. The EEG brain signal was first considered as input, then pre-processed the input signal using a fractional bandpass filter and Morphological Component Analysis (MCA) to eliminate unwanted noise from the input signal. The performance of this filter is improved by means of optimal selection of fractional orders such as \(\:\alpha\:\) and \(\:\beta\:\) in the range of 0 to 1. In order to select the optimal fractional order value dingo optimization algorithm is used. After pre-processing, features are extracted using discrete wavelet transform (DWT) and empirical mode decomposition (EMD) in order to calculate the feature coefficients. The features are then fed into the proposed Radial Basis Function Neural Network (RBFNN) classifier as an input. Finally, the classifier predicts whether the epilepsy disease affected the patients or not. According to the simulation analysis, the proposed method achieves 98% precision, 2% error, 98% accuracy, 97% specificity and so on. It shows that the proposed approach outperforms previous methods. According to this proposed model, the prognosis for epileptic disease may be improved, leading to a significant enhancement of the quality lives for epileptic patients.