COVID-19 detection in chest X-ray images using deep boosted hybrid learning

Saddam Hussain Khan, Anabia Sohail, Asifullah Khan, Mehdi Hassan, Yeon Soo Lee, Jamshed Alam, Abdul Basit, Saima Zubair

Abstract: Highlights•Two new frameworks, named as DHL and DBHL are proposed for COVID-19 detection in chest X-ray images.•DHL framework exploits the learning capacity of the developed COVID-RENets and SVM. COVID-RENet systematically learns the region homogeneity and boundaries features.•In the DBHL framework, rich information boosted representation is obtained by concatenating the feature space of the COVID-RENets.•The proposed frameworks significantly decrease false negatives as compared to existing deep CNNs.•A web predictor is developed for assisting the radiologist in making accurate COVID-19 decisions.