RT Journal Article
SR Electronic
T1 The absolute lobar and pixel analysis of COVID-19 pneumonia in CT images
JF European Respiratory Journal
JO Eur Respir J
FD European Respiratory Society
SP PA3241
DO 10.1183/13993003.congress-2021.PA3241
VO 58
IS suppl 65
A1 Elif Soya
A1 Nur Ekenel
A1 Serkan Kocoglu
A1 Recep Savas
A1 Minsuok Kim
A1 Ozkan Doganay
YR 2021
UL http://erj.ersjournals.com/content/58/suppl_65/PA3241.abstract
AB Introduction:Artificial Intelligence (AI) based algorithms demonstrated a higher incidence of COVID-19 pneumonia in the lower lobes, however these algorithms focused on the cluster analysis of pneumonia volumes which may overlooked the pixel-by-pixel CT attenuation changes over the whole lungs particularly the upper lobes since the COVID-19 pneumonia clusters widely developed at peripheral regions at the lower lobes.Objectives:To develop a technique for quantitative measurement of pneumonia related changes in CT attenuation values over the entire lung in pixel-by-pixel bases rather than only clustered focal pneumonia volumes.Methods:Total n=100 patients confirmed COVID-19 with RT-PCR (n=50) and age matched healthy cohort (n=50) were analyzed using the proposed technique in this study and a commercially available AI software for comparison. The pixel-by-pixel CT attenuation volumes were calculated by excluding pulmonary airways, vessels, and fissure as shown in Fig. 1.Results:The %pneumonia from the upper lung lobes in COVID-19 cohort was 6.0±5.2% and %Lung-content from the healthy cohort was 1.3±0.7% and statistically different (p<0.01) which showed a sensitivity 3.9 times higher (p<0.01) than a commercially available AI algorithm.Conclusion:Using the proposed novel technique, %pneumonia could be calculated not only in the clusters but also upper lung lobes with an improved sensitivity by a factor of 3.9 compared to AI based analysis.FootnotesCite this article as: European Respiratory Journal 2021; 58: Suppl. 65, PA3241.This abstract was presented at the 2021 ERS International Congress, in session “Prediction of exacerbations in patients with COPD”.This is an ERS International Congress abstract. No full-text version is available. Further material to accompany this abstract may be available at www.ers-education.org (ERS member access only).