A Review of Deep Learning and Machine Learning Approaches in COVID-19 Detection

Abstract

COVID-19 resulted from SARS-CoV-2 and has disastrous effects on human. It has obliged researchers worldwide to exploit approaches for an accurate and reliable diagnostic tools for it early detection. However, the pandemic nature of this disease has made the traditional diagnostic methods ineffective. Artificial Intelligence (AI) researchers have come up with a number of promising algorithms to attain most effective and rapid classification system that can alternate the tedious and time consuming traditional diagnostic techniques. These algorithms used either Computed Tomography (CT) images, X-ray images or both for COVID-19 classification. The most recent exploited Deep Learning (DL) and Machine Learning (ML) approaches along with feature extraction techniques mostly on CT images are reviewed in this paper. The overall accuracy of these techniques ranged from 86.1% to 99.7^% which indicate that they are applicable in COVID-19 detection. This paper will assist researchers in future development of these techniques for COVID-19 diagnosis.

Ähnliche Arbeiten