An Optimized Machine Learning Framework for Automatic Detection of COVID-19 Using CT-Scan and X-Ray Images

Application of artificial intelligence (AI) techniques for accurate COVID-19 diagnosis on X-ray and CT-Scan imaging modalities is crucial due to the difficulties with the RT-PCR test. Researchers hold in high regard the development of computer-aided diagnosis Systems based on AI by utilizing X-ray images and CT scans for accurate diagnosis of COVID-19. In this work, an optimization-based machine learning frame work (OMLF) with different feature extraction technique is introduced to diagnosis of the COVID-19 in X-ray and CT scan images. The feature extraction techniques are to extract the features that help distinguish the COVID & non-COVID images. The extracted features are considered for denoting the images as vectors. These image vectors are trained with six different machine learning algorithms. The hyper parameters of are optimized using six algorithms in this work, among all the Random Forest (RF) with Fire Fly Optimization Algorithm fairly diagnoses COVID-19 from the CT &X-ray image directory.