Artificial intelligence analysis of chest X-ray and
computed tomographic images for COVID-19 patient management: abridged secondary publication
J Cai1, M Ying1, WC Chan1, MF Wong2, HC Tsang3, FM Kong4, YX Wang5, J Qin6, G Ren1
Department of Medicine and Therapeutics, The Chinese University of Hong Kong, Hong Kong SAR, China
- The artificial intelligence (AI)-enhanced chest X-ray (CXR) model increased lung signal and suppressed rib signal, enabling more accurate classification of COVID-19 conditions.
- The AI-based segmentation model delineated lung contours and infected regions in computed tomographic and CXR images with high Dice scores.
- The multi-view approach significantly improved the area under the curve compared with the use of a single feature type.
- The developed computer programme supports COVID-19 diagnosis and patient management.