Scholarlink

<p>Heart arrhythmia is a life-threatening cardiological disorder that attacks due to imbalances in the heart's pulse rhythms. In this paper, we proposed a hybrid improved search ability-based integrated optimization algorithm named WSO-AOA: war search optimization and archimedes optimization algorithm. In this work, three publicly available ECG databases, namely MIT-BIH Arrhythmia, MIT-BIH normal sinus rhythm, and BIDMC Congestive Heart Failure Databases from the Physio Net health center server are used for whole experimental analysis. For conducting this research, 12 features are applied to select the signal, and independent component analysis (ICA) is used for feature selection to choose the main feature components. Features are extracted with wavelet packet transform (WPT) and classified with well-known machine learning classifiers such as support vector machine (SVM), least square SVM (LSSVM), adaptive neuro-fuzzy inference system (ANFIS) and extreme learning ANFIS (ELANFIS). The proposed WSO-AOA-ELANFIS methodology outperforms in terms of accuracy for ARR (99.8%), CHF (100%), NSR (100%), sensitivity for ARR (100.%), CHF (100%), NSR (100%), Specificity for ARR (99.68%), CHF (100%), NSR (100%), and similarly G-mean, selectivity, F1-score, AuC are calculated. Our proposed algorithm has the potential for integration with the Internet of Medical Things (IoMT) and can be further evaluated using other publicly available ECG datasets.</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

Automated Arrhythmia Detection Using War Strategy Optimization Enabled with Archimedes Optimization Algorithm and Rule-Based Classifiers

  • 期刊论文

  • 发表日期：

    2025年1月7日

查看全文
  • Prakash Chandra Sahoo,
  • Binod Kumar Pattnaik

摘要

Heart arrhythmia is a life-threatening cardiological disorder that attacks due to imbalances in the heart's pulse rhythms. In this paper, we proposed a hybrid improved search ability-based integrated optimization algorithm named WSO-AOA: war search optimization and archimedes optimization algorithm. In this work, three publicly available ECG databases, namely MIT-BIH Arrhythmia, MIT-BIH normal sinus rhythm, and BIDMC Congestive Heart Failure Databases from the Physio Net health center server are used for whole experimental analysis. For conducting this research, 12 features are applied to select the signal, and independent component analysis (ICA) is used for feature selection to choose the main feature components. Features are extracted with wavelet packet transform (WPT) and classified with well-known machine learning classifiers such as support vector machine (SVM), least square SVM (LSSVM), adaptive neuro-fuzzy inference system (ANFIS) and extreme learning ANFIS (ELANFIS). The proposed WSO-AOA-ELANFIS methodology outperforms in terms of accuracy for ARR (99.8%), CHF (100%), NSR (100%), sensitivity for ARR (100.%), CHF (100%), NSR (100%), Specificity for ARR (99.68%), CHF (100%), NSR (100%), and similarly G-mean, selectivity, F1-score, AuC are calculated. Our proposed algorithm has the potential for integration with the Internet of Medical Things (IoMT) and can be further evaluated using other publicly available ECG datasets.