Litcius/Paper detail

Detection of Corona virus Disease Using a Novel Machine Learning Approach

Ayodeji Olalekan Salau

20212021 International Conference on Decision Aid Sciences and Application (DASA)23 citationsDOIOpen Access PDF

Abstract

In recent times, Coronavirus Disease 2019 (COVID-19) has become a growing concern which has taken the world by surprise. Early detection of this virus can be used to save millions of lives. In this study, a Support Vector Machine (SVM) method is proposed for the identification and classification of COVID-19 as an early diagnostic method to help clinicians and doctors to accurate distinguish COVID-19 from SARS-CoV-2. A discrete wavelet transform (DWT) algorithm was used to extract features, while SVM was used to classify the extracted features. For the performance evaluation, metrics such as sensitivity (Sens), specificity (Spec), accuracy (Acc), and F -score metrics were used. A detection rate of 98.2 % was achieved using the proposed SVM method. Finally, the performance of the SVM method is compared to that of current methods, and it is discovered that the SVM method outperforms them.

Topics & Concepts

Support vector machineArtificial intelligenceComputer scienceCoronavirus disease 2019 (COVID-19)Pattern recognition (psychology)Identification (biology)Machine learningSurpriseDiscrete wavelet transformWavelet transformData miningWaveletDiseaseMedicineSocial psychologyBiologyBotanyInfectious disease (medical specialty)PathologyPsychologyCOVID-19 diagnosis using AIAnomaly Detection Techniques and ApplicationsImage Processing Techniques and Applications