Occlusion Detection for COVID-19 Identification: A Review

Abstract

Coronavirus has significantly entered the globe and has affected everyone's life in some form or another. The identification of the coronavirus is as important as prevention. An early and precise diagnosis will have significant-good benefits for us. The Convolutional Neural Network can be implemented for virus detection, and content-adaptive progressive occlusion analysis could be used to handle occlusion. To improve precision, it is also possible to add many other classification methods accordingly, which are mentioned in this paper.

A Survey on Deep Learning Techniques for COVID-19 Detection using Medical Imaging

“…This affects both the training and reliability of deep learning models. In [1] the authors review the effect of occlusion in images and how they can affect classification and mention techniques to alleviate the effects of occlusions in these images. [2] sheds light toward the difficulty of prediction of COVID-19 fro m chest CT scans due to boundary blurring and proposes a data driven framework to counter the same thus making an unbiased learning model.…”