COVID-MTL: Multitask learning with Shift3D and random-weighted loss for COVID-19 diagnosis and severity assessment. (April 2022)

• Record Type: Journal Article
• Authors: Bao, Guoqing; Chen, Huai; Liu, Tongliang; Gong, Guanzhong; Yin, Yong; Wang, Lisheng; Wang, Xiuying
• Is Part Of: Pattern recognition. Volume 124(2022)
• Publication Date: 2022-04
• Subjects: COVID-19 -- Multitask learning -- 3D CNNs -- Diagnosis -- Severity assessment -- Deep learning -- Computer tomography; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2021.108499
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: Automated and simultaneous detection and severity assessment of COVID-19 using CT. Tackle under segmentation of ground-glass opacities in COVID-19 chest CT scans. 3D CNNs boosted by shifting low-level feature representations of volumetric inputs. COVID-19 multitask learning performance improved by random-weighted loss function. Identified features significantly related to positivity and severity of COVID-19. Abstract: There is an urgent need for automated methods to assist accurate and effective assessment of COVID-19. Radiology and nucleic acid test (NAT) are complementary COVID-19 diagnosis methods. In this paper, we present an end-to-end multitask learning (MTL) framework (COVID-MTL) that is capable of automated and simultaneous detection (against both radiology and NAT) and severity assessment of COVID-19. COVID-MTL learns different COVID-19 tasks in parallel through our novel random-weighted loss function, which assigns learning weights under Dirichlet distribution to prevent task dominance; our new 3D real-time augmentation algorithm (Shift3D) introduces space variances for 3D CNN components by shifting low-level feature representations of volumetric inputs in three dimensions; thereby, the MTL framework is able to accelerate convergence and improve joint learning performance compared to single-task models. By only using chest CT scans, COVID-MTL was trained on 930 CT scans and tested on separate 399 cases. COVID-MTL achieved AUCs of 0.939 and 0.846, and accuracies of 90.23% and 79.20% for detection of COVID-19 against radiology and NAT, respectively, which outperformed the state-of-the-art models. Meanwhile, COVID-MTL yielded AUC of 0.800 ± 0.020 and 0.813 ± 0.021 (with transfer learning) for classifying control/suspected, mild/regular, and severe/critically-ill cases. To decipher the recognition mechanism, we also identified high-throughput lung features that were significantly related ( P < 0.001) to the positivity and severity of COVID-19.