COVID CT-Net: Predicting Covid-19 From Chest CT Images Using Attentional Convolutional Network

TL;DR

This paper introduces COVID CT-Net, a deep learning model using attentional convolutional networks to accurately detect COVID-19 from chest CT images, with visualization and annotated datasets to aid future research.

Contribution

The work presents a novel attentional convolutional network for COVID-19 detection from CT images, including manual annotations of infected regions for improved interpretability.

Findings

High sensitivity and specificity achieved.

Attention maps focus on infected lung regions.

Model outperforms baseline methods.

Abstract

The novel corona-virus disease (COVID-19) pandemic has caused a major outbreak in more than 200 countries around the world, leading to a severe impact on the health and life of many people globally. As of Aug 25th of 2020, more than 20 million people are infected, and more than 800,000 death are reported. Computed Tomography (CT) images can be used as a as an alternative to the time-consuming "reverse transcription polymerase chain reaction (RT-PCR)" test, to detect COVID-19. In this work we developed a deep learning framework to predict COVID-19 from CT images. We propose to use an attentional convolution network, which can focus on the infected areas of chest, enabling it to perform a more accurate prediction. We trained our model on a dataset of more than 2000 CT images, and report its performance in terms of various popular metrics, such as sensitivity, specificity, area under the curve, and also precision-recall curve, and achieve very promising results. We also provide a visualization of the attention maps of the model for several test images, and show that our model is attending to the infected regions as intended. In addition to developing a machine learning modeling framework, we also provide the manual annotation of the potentionally infected regions of chest, with the help of a board-certified radiologist, and make that publicly available for other researchers.