Covid-19 classification with deep neural network and belief functions

TL;DR

This paper introduces a belief function-enhanced deep neural network with semi-supervised training for automatic Covid-19 detection from CT images, improving reliability and explainability over traditional models.

Contribution

It presents a novel belief function-based CNN that combines deep features with belief degree maps for more reliable Covid-19 classification.

Findings

Achieved 81% accuracy in Covid-19 detection

F1 score of 0.812 indicating balanced precision and recall

AUC of 0.875 demonstrating strong discriminative ability

Abstract

Computed tomography (CT) image provides useful information for radiologists to diagnose Covid-19. However, visual analysis of CT scans is time-consuming. Thus, it is necessary to develop algorithms for automatic Covid-19 detection from CT images. In this paper, we propose a belief function-based convolutional neural network with semi-supervised training to detect Covid-19 cases. Our method first extracts deep features, maps them into belief degree maps and makes the final classification decision. Our results are more reliable and explainable than those of traditional deep learning-based classification models. Experimental results show that our approach is able to achieve a good performance with an accuracy of 0.81, an F1 of 0.812 and an AUC of 0.875.