Artificial Intelligence Applied to Chest X-Ray Images for the Automatic Detection of COVID-19. A Thoughtful Evaluation Approach.

Julian D Arias-Londono,Laureano Moro-Velazquez,Jorge A Gomez-Garcia,Juan I Godino-Llorente

doi:10.1109/access.2020.3044858

Julian D Arias-Londono, Laureano Moro-Velazquez + Show 2 more

Open Access

https://doi.org/10.1109/access.2020.3044858

Copy DOI

Abstract

Current standard protocols used in the clinic for diagnosing COVID-19 include molecular or antigen tests, generally complemented by a plain chest X-Ray. The combined analysis aims to reduce the significant number of false negatives of these tests and provide complementary evidence about the presence and severity of the disease. However, the procedure is not free of errors, and the interpretation of the chest X-Ray is only restricted to radiologists due to its complexity. With the long term goal to provide new evidence for the diagnosis, this paper presents an evaluation of different methods based on a deep neural network. These are the first steps to develop an automatic COVID-19 diagnosis tool using chest X-Ray images to differentiate between controls, pneumonia, or COVID-19 groups. The paper describes the process followed to train a Convolutional Neural Network with a dataset of more than 79, 500 X-Ray images compiled from different sources, including more than 8, 500 COVID-19 examples. Three different experiments following three preprocessing schemes are carried out to evaluate and compare the developed models. The aim is to evaluate how preprocessing the data affects the results and improves its explainability. Likewise, a critical analysis of different variability issues that might compromise the system and its effects is performed. With the employed methodology, a 91.5% classification accuracy is obtained, with an 87.4% average recall for the worst but most explainable experiment, which requires a previous automatic segmentation of the lung region.

Highlights

COVID-19 pandemic has rapidly become one of the biggest health world challenges in recent years
Arias-Londoño et al.: Artificial Intelligence Applied to Chest X-Ray Images for the Automatic Detection of COVID-19 cases, the estimated period from the onset of the disease to death ranges from 6 to 41 days, being dependent on the patient’s age and the patient’s immune system status [3]
Arias-Londoño et al.: Artificial Intelligence Applied to Chest X-Ray Images for the Automatic Detection of COVID-19 TABLE 1

Summary

Introduction

COVID-19 pandemic has rapidly become one of the biggest health world challenges in recent years. In [14] authors developed a 121-layer Convolutional Neural Network (CNN) architecture, called Chexnet, which was trained with a dataset of 100, 000 XR images for the detection of different types of pneumonia. In [16], seven different deep CNN models were tested using a corpus of 50 controls and 25 COVID-19 patients. The best results were attained with the VGG19 and DenseNet models, obtaining F1-scores of 0.89 and 0.91 for controls and patients. The model was trained with a corpus gathered from different sources, consisting of 4, 575 XR images: 1, 525 of COVID-19 ( 912 come from a repository applying data augmentation), 1, 525 of pneumonia, and 1, 525 of controls. In [20], the VGG16 network was used for classification, employing a database of 132 COVID-19, 132 controls and 132 pneumonia images. Following a hold-out validation, about 100% accuracy was obtained identifying COVID-19, being lower on the other classes

Objectives

Methods

Results

Conclusion