Truncated inception net: COVID-19 outbreak screening using chest X-rays

TLDR

The COVID‑19 pandemic has spread rapidly worldwide, and AI‑driven imaging tools such as CT scans and chest X‑rays are widely used to detect outbreaks. This study proposes a deep‑learning Convolutional Neural Network, Truncated Inception Net, to screen COVID‑19 positive chest X‑rays from non‑COVID and healthy cases. The model was validated on six datasets comprising COVID‑19, pneumonia, tuberculosis, and healthy chest X‑rays. It achieved 99.96 % accuracy (AUC 1.0) versus pneumonia and healthy cases and 99.92 % accuracy (AUC 0.99) versus pneumonia, tuberculosis, and healthy cases, outperforming existing AI screening tools.

Abstract

Since December 2019, the Coronavirus Disease (COVID-19) pandemic has caused world-wide turmoil in a short period of time, and the infection, caused by SARS-CoV-2, is spreading rapidly. AI-driven tools are used to identify Coronavirus outbreaks as well as forecast their nature of spread, where imaging techniques are widely used, such as CT scans and chest X-rays (CXRs). In this paper, motivated by the fact that X-ray imaging systems are more prevalent and cheaper than CT scan systems, a deep learning-based Convolutional Neural Network (CNN) model, which we call Truncated Inception Net, is proposed to screen COVID-19 positive CXRs from other non-COVID and/or healthy cases. To validate our proposal, six different types of datasets were employed by taking the following CXRs: COVID-19 positive, Pneumonia positive, Tuberculosis positive, and healthy cases into account. The proposed model achieved an accuracy of 99.96% (AUC of 1.0) in classifying COVID-19 positive cases from combined Pneumonia and healthy cases. Similarly, it achieved an accuracy of 99.92% (AUC of 0.99) in classifying COVID-19 positive cases from combined Pneumonia, Tuberculosis, and healthy CXRs. To the best of our knowledge, as of now, the achieved results outperform the existing AI-driven tools for screening COVID-19 using the acquired CXRs, and proves the viability of using the proposed Truncated Inception Net as a screening tool.