Show simple item record

dc.contributor.authorBai, Xiang
dc.contributor.authorWang, Hanchen
dc.contributor.authorMa, Liya
dc.contributor.authorXu, Yongchao
dc.contributor.authorGan, Jiefeng
dc.contributor.authorFan, Ziwei
dc.contributor.authorYang, Fan
dc.contributor.authorMa, Ke
dc.contributor.authorYang, Jiehua
dc.contributor.authorBai, Song
dc.contributor.authorShu, Chang
dc.contributor.authorZou, Xinyu
dc.contributor.authorHuang, Renhao
dc.contributor.authorZhang, Changzheng
dc.contributor.authorLiu, Xiaowu
dc.contributor.authorTu, Dandan
dc.contributor.authorXu, Chuou
dc.contributor.authorZhang, Wenqing
dc.contributor.authorWang, Xi
dc.contributor.authorChen, Anguo
dc.contributor.authorZeng, Yu
dc.contributor.authorYang, Dehua
dc.contributor.authorWang, Ming-Wei
dc.contributor.authorHolalkere, Nagaraj
dc.contributor.authorHalin, Neil J
dc.contributor.authorKamel, Ihab R
dc.contributor.authorWu, Jia
dc.contributor.authorPeng, Xuehua
dc.contributor.authorWang, Xiang
dc.contributor.authorShao, Jianbo
dc.contributor.authorMongkolwat, Pattanasak
dc.contributor.authorZhang, Jianjun
dc.contributor.authorLiu, Weiyang
dc.contributor.authorRoberts, Michael
dc.contributor.authorTeng, Zhongzhao
dc.contributor.authorBeer, Lucian
dc.contributor.authorSanchez, Lorena Escudero
dc.contributor.authorSala, Evis
dc.contributor.authorRubin, Daniel
dc.contributor.authorWeller, Adrian
dc.contributor.authorLasenby, Joan
dc.contributor.authorZheng, Chuangsheng
dc.contributor.authorWang, Jianming
dc.contributor.authorLi, Zhen
dc.contributor.authorSchönlieb, Carola-Bibiane
dc.contributor.authorXia, Tian
dc.date.accessioned2022-01-05T16:30:31Z
dc.date.available2022-01-05T16:30:31Z
dc.date.issued2021-11-18
dc.date.submitted2021-03-02
dc.identifier.issn2522-5839
dc.identifier.others42256-021-00421-z
dc.identifier.other421
dc.identifier.urihttps://www.repository.cam.ac.uk/handle/1810/332056
dc.description.abstractArtificial intelligence (AI) provides a promising substitution for streamlining COVID-19 diagnoses. However, concerns surrounding security and trustworthiness impede the collection of large-scale representative medical data, posing a considerable challenge for training a well-generalised model in clinical practices. To address this, we launch the Unified CT-COVID AI Diagnostic Initiative (UCADI), where the AI model can be distributedly trained and independently executed at each host institution under a federated learning framework (FL) without data sharing. Here we show that our FL model outperformed all the local models by a large yield (test sensitivity /specificity in China: 0.973/0.951, in the UK: 0.730/0.942), achieving comparable performance with a panel of professional radiologists. We further evaluated the model on the hold-out (collected from another two hospitals leaving out the FL) and heterogeneous (acquired with contrast materials) data, provided visual explanations for decisions made by the model, and analysed the trade-offs between the model performance and the communication costs in the federated training process. Our study is based on 9,573 chest computed tomography scans (CTs) from 3,336 patients collected from 23 hospitals located in China and the UK. Collectively, our work advanced the prospects of utilising federated learning for privacy-preserving AI in digital health.
dc.languageen
dc.publisherSpringer Science and Business Media LLC
dc.subjectArticle
dc.subject/4000/4008
dc.subject/692/700
dc.subject/639/705/1042
dc.subject/631/326/596/4130
dc.subjectarticle
dc.titleAdvancing COVID-19 Diagnosis with Privacy-Preserving Collaboration in Artificial Intelligence.
dc.typeArticle
dc.date.updated2022-01-05T16:30:30Z
prism.endingPage1089
prism.issueIdentifier12
prism.publicationNameArXiv
prism.startingPage1081
prism.volume3
dc.identifier.doi10.17863/CAM.79503
dcterms.dateAccepted2021-10-27
rioxxterms.versionofrecord10.1038/s42256-021-00421-z
rioxxterms.versionVoR
rioxxterms.licenseref.urihttp://creativecommons.org/licenses/by/4.0/
dc.contributor.orcidRoberts, Michael [0000-0002-3484-5031]
dc.contributor.orcidSala, Evis [0000-0002-5518-9360]
dc.contributor.orcidWeller, Adrian [0000-0003-1915-7158]
dc.contributor.orcidLasenby, Joan [0000-0002-0571-0218]
dc.identifier.eissn2331-8422
pubs.funder-project-idEngineering and Physical Sciences Research Council (EP/N014588/1)
pubs.funder-project-idEuropean Commission Horizon 2020 (H2020) Marie Sk?odowska-Curie actions (777826)
pubs.funder-project-idEPSRC (EP/S026045/1)
pubs.funder-project-idEPSRC (EP/T017961/1)
pubs.funder-project-idEPSRC (EP/T003553/1)
cam.issuedOnline2021-12-15


Files in this item

Thumbnail
Thumbnail
Thumbnail

This item appears in the following Collection(s)

Show simple item record