2020/04/05 by Karim Hammoudi, Halim Benhabiles, Hammoudi, Karim +11 · 1 citation
Computer Science · Medicine · #AI in cancer detection #COVID-19 diagnosis using AI #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #FOS: Electrical engineering #I.2.6 #I.4.9 #Image and Video Processing (eess.IV) #J.3 #Machine Learning (cs.LG) #Radiomics and Machine Learning in Medical Imaging #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2004.03399
openalex publication_date 2020/04/05 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Coronavirus disease 2019 (COVID-19) is an infectious disease with first\nsymptoms similar to the flu. COVID-19 appeared first in China and very quickly\nspreads to the rest of the world, causing then the 2019-20 coronavirus\npandemic. In many cases, this disease causes pneumonia. Since pulmonary\ninfections can be observed through radiography images, this paper investigates\ndeep learning methods for automatically analyzing query chest X-ray images with\nthe hope to bring precision tools to health professionals towards screening the\nCOVID-19 and diagnosing confirmed patients. In this context, training datasets,\ndeep learning architectures and analysis strategies have been experimented from\npublicly open sets of chest X-ray images. Tailored deep learning models are\nproposed to detect pneumonia infection cases, notably viral cases. It is\nassumed that viral pneumonia cases detected during an epidemic COVID-19 context\nhave a high probability to presume COVID-19 infections. Moreover, easy-to-apply\nhealth indicators are proposed for estimating infection status and predicting\npatient status from the detected pneumonia cases. Experimental results show\npossibilities of training deep learning models over publicly open sets of chest\nX-ray images towards screening viral pneumonia. Chest X-ray test images of\nCOVID-19 infected patients are successfully diagnosed through detection models\nretained for their performances. The efficiency of proposed health indicators\nis highlighted through simulated scenarios of patients presenting infections\nand health problems by combining real and synthetic health data.\n