PMID- 35318171 OWN - NLM STAT- MEDLINE DCOM- 20220519 LR - 20221216 IS - 1879-0534 (Electronic) IS - 0010-4825 (Print) IS - 0010-4825 (Linking) VI - 145 DP - 2022 Jun TI - Machine learning for detecting COVID-19 from cough sounds: An ensemble-based MCDM method. PG - 105405 LID - S0010-4825(22)00197-4 [pii] LID - 10.1016/j.compbiomed.2022.105405 [doi] AB - This research aims to analyze the performance of state-of-the-art machine learning techniques for classifying COVID-19 from cough sounds and to identify the model(s) that consistently perform well across different cough datasets. Different performance evaluation metrics (precision, sensitivity, specificity, AUC, accuracy, etc.) make selecting the best performance model difficult. To address this issue, in this paper, we propose an ensemble-based multi-criteria decision making (MCDM) method for selecting top performance machine learning technique(s) for COVID-19 cough classification. We use four cough datasets, namely Cambridge, Coswara, Virufy, and NoCoCoDa to verify the proposed method. At first, our proposed method uses the audio features of cough samples and then applies machine learning (ML) techniques to classify them as COVID-19 or non-COVID-19. Then, we consider a multi-criteria decision-making (MCDM) method that combines ensemble technologies (i.e., soft and hard) to select the best model. In MCDM, we use the technique for order preference by similarity to ideal solution (TOPSIS) for ranking purposes, while entropy is applied to calculate evaluation criteria weights. In addition, we apply the feature reduction process through recursive feature elimination with cross-validation under different estimators. The results of our empirical evaluations show that the proposed method outperforms the state-of-the-art models. We see that when the proposed method is used for analysis using the Extra-Trees classifier, it has achieved promising results (AUC: 0.95, Precision: 1, Recall: 0.97). CI - Copyright (c) 2022 Elsevier Ltd. All rights reserved. FAU - Chowdhury, Nihad Karim AU - Chowdhury NK AD - Department of Computer Science and Engineering, University of Chittagong, Bangladesh. Electronic address: nihad@cu.ac.bd. FAU - Kabir, Muhammad Ashad AU - Kabir MA AD - Data Science Research Unit, School of Computing, Mathematics and Engineering, Charles Sturt University, NSW, Australia. Electronic address: akabir@csu.edu.au. FAU - Rahman, Md Muhtadir AU - Rahman MM AD - Department of Computer Science and Engineering, University of Chittagong, Bangladesh. Electronic address: muhtadir.cse@std.cu.ac.bd. FAU - Islam, Sheikh Mohammed Shariful AU - Islam SMS AD - Institute for Physical Activity and Nutrition, Deakin University, Geelong, VIC, 3216, Australia. Electronic address: shariful.islam@deakin.edu.au. LA - eng PT - Journal Article DEP - 20220317 PL - United States TA - Comput Biol Med JT - Computers in biology and medicine JID - 1250250 SB - IM MH - Algorithms MH - *COVID-19/diagnosis MH - Cough/diagnosis MH - Humans MH - Machine Learning MH - Sound PMC - PMC8926945 OTO - NOTNLM OT - COVID-19 OT - Classification OT - Cough OT - Ensemble OT - Entropy OT - MCDM OT - Machine learning OT - TOPSIS COIS- All authors declare that there is no conflict of interest in this work. EDAT- 2022/03/24 06:00 MHDA- 2022/05/20 06:00 PMCR- 2022/03/17 CRDT- 2022/03/23 05:43 PHST- 2021/11/20 00:00 [received] PHST- 2022/03/10 00:00 [revised] PHST- 2022/03/11 00:00 [accepted] PHST- 2022/03/24 06:00 [pubmed] PHST- 2022/05/20 06:00 [medline] PHST- 2022/03/23 05:43 [entrez] PHST- 2022/03/17 00:00 [pmc-release] AID - S0010-4825(22)00197-4 [pii] AID - 105405 [pii] AID - 10.1016/j.compbiomed.2022.105405 [doi] PST - ppublish SO - Comput Biol Med. 2022 Jun;145:105405. doi: 10.1016/j.compbiomed.2022.105405. Epub 2022 Mar 17.