{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,30]],"date-time":"2025-10-30T17:41:40Z","timestamp":1761846100814,"version":"build-2065373602"},"reference-count":30,"publisher":"Public Library of Science (PLoS)","issue":"10","license":[{"start":{"date-parts":[[2025,10,30]],"date-time":"2025-10-30T00:00:00Z","timestamp":1761782400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>Novel respiratory diseases can have a devastating impact on healthcare systems, which underlines the importance of developing methods to improve the prevention, diagnosis, and prognosis of such diseases. Developing computer-aided diagnosis tools that determine infection severity can aid healthcare professionals in deciding treatment strategies and preventing cross-infection. In such manner, lung infection severity classification from chest CTs is crucial for deciding treatment plans and interventions needed to block illness progression in individual patients and reduce cross infection. However, current techniques face performance, generalizability, and explainability challenges for automated infection severity classification methods due to the high spatial complexity of 3D volumes. Significant efforts have been focused on segmentation approaches to quantify lung infection and assess infection severity, but such segmentation methods require expensive data annotation and clinical expertise. In this work, we propose a weakly-supervised classification approach to distinguish between different levels of infection, while providing clinicians with explainable results. To mimic clinical practice, the different stages in our approach focus on low-level infection patterns as well as high-level infection coverage in lung CT scans. We then fuse the high-level features with the positionally encoded low-level features to provide volume-level infection classification. Testing on the SARS-CoV-2 (COVID-19) multi-centre multi-region datasets, our approach shows promising performance gains compared to existing state-of-the-art methods, where we achieve state-of-the-art severity classification performance. Furthermore, we show significant performance gains on cross-site train\/test splits. Finally, we quantitatively and qualitatively demonstrate the explainability of our weakly-supervised approach, where we can achieve substantial infection coverage.<\/jats:p>","DOI":"10.1371\/journal.pone.0334431","type":"journal-article","created":{"date-parts":[[2025,10,30]],"date-time":"2025-10-30T17:36:36Z","timestamp":1761845796000},"page":"e0334431","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Weakly-supervised explainable infection severity classification from chest CT scans"],"prefix":"10.1371","volume":"20","author":[{"ORCID":"https:\/\/orcid.org\/0009-0008-8802-7107","authenticated-orcid":true,"given":"Ibrahim","family":"Almakky","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Mohammad","family":"Yaqub","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"340","published-online":{"date-parts":[[2025,10,30]]},"reference":[{"key":"pone.0334431.ref001","doi-asserted-by":"crossref","first-page":"107184","DOI":"10.1016\/j.asoc.2021.107184","article-title":"CoVNet-19: a deep learning model for the detection and analysis of COVID-19 patients","volume":"104","author":"P Kedia","year":"2021","journal-title":"Appl Soft Comput."},{"key":"pone.0334431.ref002","doi-asserted-by":"crossref","first-page":"114054","DOI":"10.1016\/j.eswa.2020.114054","article-title":"Deep learning approaches for COVID-19 detection based on chest X-ray images","volume":"164","author":"AM Ismael","year":"2021","journal-title":"Expert Syst Appl."},{"issue":"2","key":"pone.0334431.ref003","doi-asserted-by":"crossref","first-page":"1010","DOI":"10.1007\/s10489-020-01867-1","article-title":"COVID-19 detection and disease progression visualization: deep learning on chest X-rays for classification and coarse localization","volume":"51","author":"T Zebin","year":"2021","journal-title":"Appl Intell (Dordr)."},{"key":"pone.0334431.ref004","doi-asserted-by":"crossref","unstructured":"Ridzuan M, Bawazir A, Gollini NI, Almakky I, Yaqub M. Self-supervision and multi-task learning: challenges in fine-grained COVID-19 multi-class classification from chest X-rays. In: Yang G, Aviles-Rivero A, Roberts M, Sch\u00f6nlieb CB, editors. Medical Image Understanding and Analysis. Lecture Notes in Computer Science. Springer International Publishing. p. 234\u201350.","DOI":"10.1007\/978-3-031-12053-4_18"},{"key":"pone.0334431.ref005","unstructured":"Panel CTG. Coronavirus disease 2019 (COVID-19) treatment guidelines. p. 360."},{"key":"pone.0334431.ref006","doi-asserted-by":"crossref","first-page":"108941","DOI":"10.1016\/j.ejrad.2020.108941","article-title":"CT manifestations of coronavirus disease-2019: a retrospective analysis of 73 cases by disease severity","volume":"126","author":"K-C Liu","year":"2020","journal-title":"Eur J Radiol."},{"issue":"12","key":"pone.0334431.ref007","doi-asserted-by":"crossref","first-page":"1197","DOI":"10.1038\/s41551-020-00633-5","article-title":"Open resource of clinical data from patients with pneumonia for the prediction of COVID-19 outcomes via deep learning","volume":"4","author":"W Ning","year":"2020","journal-title":"Nat Biomed Eng."},{"issue":"8","key":"pone.0334431.ref008","doi-asserted-by":"crossref","first-page":"2584","DOI":"10.1109\/TMI.2020.2996256","article-title":"Accurate screening of COVID-19 Using attention-based deep 3D multiple instance learning","volume":"39","author":"Z Han","year":"2020","journal-title":"IEEE Trans Med Imaging."},{"issue":"1","key":"pone.0334431.ref009","doi-asserted-by":"crossref","first-page":"19549","DOI":"10.1038\/s41598-020-76550-z","article-title":"COVID-Net: a tailored deep convolutional neural network design for detection of COVID-19 cases from chest X-ray images","volume":"10","author":"L Wang","year":"2020","journal-title":"Sci Rep."},{"key":"pone.0334431.ref010","doi-asserted-by":"crossref","first-page":"151972","DOI":"10.1109\/ACCESS.2021.3125324","article-title":"Self-supervised deep convolutional neural network for chest X-ray classification","volume":"9","author":"M Gazda","year":"2021","journal-title":"IEEE Access."},{"issue":"2","key":"pone.0334431.ref011","doi-asserted-by":"crossref","first-page":"854","DOI":"10.1007\/s10489-020-01829-7","article-title":"Classification of COVID-19 in chest X-ray images using DeTraC deep convolutional neural network","volume":"51","author":"A Abbas","year":"2021","journal-title":"Appl Intell (Dordr)."},{"key":"pone.0334431.ref012","doi-asserted-by":"crossref","first-page":"105532","DOI":"10.1016\/j.cmpb.2020.105532","article-title":"COVID-19 identification in chest X-ray images on flat and hierarchical classification scenarios","volume":"194","author":"RM Pereira","year":"2020","journal-title":"Comput Methods Programs Biomed."},{"issue":"1","key":"pone.0334431.ref013","doi-asserted-by":"crossref","first-page":"202","DOI":"10.1148\/radiol.2020200230","article-title":"CT imaging features of 2019 novel coronavirus (2019-nCoV)","volume":"295","author":"M Chung","year":"2020","journal-title":"Radiology."},{"key":"pone.0334431.ref014","doi-asserted-by":"crossref","first-page":"103795","DOI":"10.1016\/j.compbiomed.2020.103795","article-title":"Application of deep learning technique to manage COVID-19 in routine clinical practice using CT images: results of 10 convolutional neural networks","volume":"121","author":"AA Ardakani","year":"2020","journal-title":"Comput Biol Med."},{"key":"pone.0334431.ref015","unstructured":"Gozes O, Frid-Adar M, Greenspan H, Browning PD, Zhang H, Ji W. Rapid AI Development Cycle for the Coronavirus (COVID-19) pandemic: initial results for automated detection & patient monitoring using deep learning CT image analysis. 2020."},{"key":"pone.0334431.ref016","first-page":"1100110","article-title":"xViTCOS: explainable vision transformer based COVID-19 screening using radiography","volume":"10","author":"AK Mondal","year":"2021","journal-title":"IEEE J Transl Eng Health Med."},{"key":"pone.0334431.ref017","doi-asserted-by":"crossref","unstructured":"Wang L, Wang J, Zhu L, Fu H, Li P, Cheng G. Dual multi-scale mean teacher network for semi-supervised infection segmentation in chest CT volume for COVID-19. arXiv preprint 2022. http:\/\/arxiv.org\/abs\/2211.05548","DOI":"10.1109\/TCYB.2022.3223528"},{"key":"pone.0334431.ref018","doi-asserted-by":"crossref","first-page":"3113","DOI":"10.1109\/TIP.2021.3058783","article-title":"JCS: an explainable COVID-19 diagnosis system by joint classification and segmentation","volume":"30","author":"Y-H Wu","year":"2021","journal-title":"IEEE Trans Image Process."},{"issue":"6","key":"pone.0334431.ref019","doi-asserted-by":"crossref","first-page":"4846","DOI":"10.1609\/aaai.v35i6.16617","article-title":"MiniSeg: an extremely minimum network for efficient COVID-19 segmentation","volume":"35","author":"Y Qiu","year":"2021","journal-title":"AAAI."},{"key":"pone.0334431.ref020","article-title":"A collaborative online AI engine for CT-based COVID-19 diagnosis","author":"Y Xu","year":"2020","journal-title":"medRxiv."},{"issue":"3","key":"pone.0334431.ref021","doi-asserted-by":"crossref","first-page":"199","DOI":"10.1038\/s42256-021-00307-0","article-title":"Common pitfalls and recommendations for using machine learning to detect and prognosticate for COVID-19 using chest radiographs and CT scans","volume":"3","author":"M Roberts","year":"2021","journal-title":"Nat Mach Intell."},{"issue":"5","key":"pone.0334431.ref022","doi-asserted-by":"crossref","first-page":"494","DOI":"10.1038\/s42256-022-00483-7","article-title":"An interpretable deep learning workflow for discovering subvisual abnormalities in CT scans of COVID-19 inpatients and survivors","volume":"4","author":"L Zhou","year":"2022","journal-title":"Nat Mach Intell."},{"issue":"10","key":"pone.0334431.ref023","doi-asserted-by":"crossref","first-page":"2344","DOI":"10.3390\/math11102344","article-title":"ELCT-YOLO: an efficient one-stage model for automatic lung tumor detection based on CT images","volume":"11","author":"Z Ji","year":"2023","journal-title":"Mathematics."},{"key":"pone.0334431.ref024","doi-asserted-by":"crossref","unstructured":"Zhou Y, Zhu Y, Ye Q, Qiu Q, Jiao J. Weakly supervised instance segmentation using class peak response. In: 2018 IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 2018. p. 3791\u2013800. https:\/\/doi.org\/10.1109\/cvpr.2018.00399","DOI":"10.1109\/CVPR.2018.00399"},{"issue":"10","key":"pone.0334431.ref025","doi-asserted-by":"crossref","first-page":"3475","DOI":"10.3390\/s21103475","article-title":"WS-RCNN: Learning to Score Proposals for Weakly Supervised Instance Segmentation","volume":"21","author":"J-R Ou","year":"2021","journal-title":"Sensors (Basel)."},{"key":"pone.0334431.ref026","unstructured":"Dosovitskiy A, Beyer L, Kolesnikov A, Weissenborn D, Zhai X, Unterthiner T. An image is worth 16x16 words: transformers for image recognition at scale. arXiv. 2020."},{"key":"pone.0334431.ref027","doi-asserted-by":"crossref","unstructured":"Morozov SP, Andreychenko AE, Pavlov NA, Vladzymyrskyy AV, Ledikhova NV, Gombolevskiy VA, et al. MosMedData: chest CT scans with COVID-19 related findings dataset. Cold Spring Harbor Laboratory. 2020. https:\/\/doi.org\/10.1101\/2020.05.20.20100362","DOI":"10.1101\/2020.05.20.20100362"},{"key":"pone.0334431.ref028","unstructured":"China NHC. Protocol on Prevention and Control of COVID-19 (Edition 6). 2023."},{"key":"pone.0334431.ref029","unstructured":"Hara K, Kataoka H, Satoh Y. Can spatiotemporal 3D CNNs retrace the history of 2D CNNs and ImageNet?arXiv preprint http:\/\/arxiv.org\/abs\/1711.09577"},{"key":"pone.0334431.ref030","unstructured":"Kingma DP, Ba J. Adam: a method for stochastic optimization. arXiv preprint http:\/\/arxiv.org\/abs\/1412.6980"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0334431","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,30]],"date-time":"2025-10-30T17:36:42Z","timestamp":1761845802000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0334431"}},"subtitle":[],"editor":[{"given":"Asadullah","family":"Shaikh","sequence":"first","affiliation":[],"role":[{"role":"editor","vocabulary":"crossref"}]}],"short-title":[],"issued":{"date-parts":[[2025,10,30]]},"references-count":30,"journal-issue":{"issue":"10","published-online":{"date-parts":[[2025,10,30]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0334431","relation":{},"ISSN":["1932-6203"],"issn-type":[{"type":"electronic","value":"1932-6203"}],"subject":[],"published":{"date-parts":[[2025,10,30]]}}}