© FUJIFILM Corporation

A Deep Learning-based CT Classification System for COVID-19

2022.09.29Journals

Coronavirus disease 2019 (COVID-19) is typically confirmed by reverse-transcription polymerase chain reaction (RT-PCR) testing. However, RT-PCR has some problems of its accuracy and required time. In this paper, authors aimed to develop and externally validate a novel machine learning model that can classify CT image as COVID-19 or non-COVID-19. They used 2,928 images from a wide variety of case-control type data sources for the development. In external validation, proposed model exhibited a high sensitivity in external validation datasets. The model may help physicians to rule out COVID-19 in a timely manner at emergency departments. Further studies are warranted to improve model specificity.

Figure 1. Proposed approach [1] (Lung areas and lung disease areas are extracted respectively by the lung segmentation model and lung disease segmentation model. This information is used for learning of classify image findings as COVID-19 or non-COVID-19.)
Figure 2. Sensitivity analysis for external validation datasets in consideration of initial RT-PCR findings as the reference standard [1]
Figure 3. ROC curves for external validation datasets in consideration of initial RT-PCR findings as the reference standard [1]
[1] Kataoka, Yuki, et al. “Development and external validation of a deep learning-based computed tomography classification system for COVID-19.” Annals of Clinical Epidemiology (2022): 22014.

DOI: https://doi.org/10.37737/ace.22014

Index