Deep Learning System Outperforms Clinicians in Identifying Optic Disc Abnormalities

dc.contributor.authorVasseneix C.
dc.contributor.authorNusinovici S.
dc.contributor.authorXu X.
dc.contributor.authorHwang J.M.
dc.contributor.authorHamann S.
dc.contributor.authorChen J.J.
dc.contributor.authorLoo J.L.
dc.contributor.authorMilea L.
dc.contributor.authorTan K.B.K.
dc.contributor.authorTing D.S.W.
dc.contributor.authorLiu Y.
dc.contributor.authorNewman N.J.
dc.contributor.authorBiousse V.
dc.contributor.authorWong T.Y.
dc.contributor.authorMilea D.
dc.contributor.authorNajjar R.P.
dc.contributor.authorGohier P.
dc.contributor.authorMiller N.
dc.contributor.authorVanikieti K.
dc.contributor.authorLa Morgia C.
dc.contributor.authorRougier M.B.
dc.contributor.authorAmbika S.
dc.contributor.authorFonseca P.
dc.contributor.authorLagrèze W.A.
dc.contributor.authorSanda N.
dc.contributor.authorChiquet C.
dc.contributor.authorYang H.
dc.contributor.authorChan C.K.M.
dc.contributor.authorCheung C.Y.
dc.contributor.authorChau T.T.H.
dc.contributor.authorJurkute N.
dc.contributor.authorYu-Wai-Man P.
dc.contributor.authorKho R.
dc.contributor.authorJonas J.B.
dc.contributor.authorVignal-Clermont C.
dc.contributor.authorKim D.H.
dc.contributor.authorYang H.K.
dc.contributor.authorAung T.
dc.contributor.authorSinghal S.
dc.contributor.authorTow S.
dc.contributor.authorNongpiur M.E.
dc.contributor.authorPerera S.
dc.contributor.authorNarayanaswamy A.
dc.contributor.authorThirugnanam U.N.
dc.contributor.authorFraser C.L.
dc.contributor.authorMejico L.J.
dc.contributor.authorFard M.A.
dc.contributor.otherMahidol University
dc.date.accessioned2023-06-12T17:16:17Z
dc.date.available2023-06-12T17:16:17Z
dc.date.issued2023-06-01
dc.description.abstractBackground: The examination of the optic nerve head (optic disc) is mandatory in patients with headache, hypertension, or any neurological symptoms, yet it is rarely or poorly performed in general clinics. We recently developed a brain and optic nerve study with artificial intelligence-deep learning system (BONSAI-DLS) capable of accurately detecting optic disc abnormalities including papilledema (swelling due to elevated intracranial pressure) on digital fundus photographs with a comparable classification performance to expert neuro-ophthalmologists, but its performance compared to first-line clinicians remains unknown. Methods: In this international, cross-sectional multicenter study, the DLS, trained on 14,341 fundus photographs, was tested on a retrospectively collected convenience sample of 800 photographs (400 normal optic discs, 201 papilledema and 199 other abnormalities) from 454 patients with a robust ground truth diagnosis provided by the referring expert neuro-ophthalmologists. The areas under the receiver-operating-characteristic curves were calculated for the BONSAI-DLS. Error rates, accuracy, sensitivity, and specificity of the algorithm were compared with those of 30 clinicians with or without ophthalmic training (6 general ophthalmologists, 6 optometrists, 6 neurologists, 6 internists, 6 emergency department [ED] physicians) who graded the same testing set of images. Results: With an error rate of 15.3%, the DLS outperformed all clinicians (average error rates 24.4%, 24.8%, 38.2%, 44.8%, 47.9% for general ophthalmologists, optometrists, neurologists, internists and ED physicians, respectively) in the overall classification of optic disc appearance. The DLS displayed significantly higher accuracies than 100%, 86.7% and 93.3% of clinicians (n = 30) for the classification of papilledema, normal, and other disc abnormalities, respectively. Conclusions: The performance of the BONSAI-DLS to classify optic discs on fundus photographs was superior to that of clinicians with or without ophthalmic training. A trained DLS may offer valuable diagnostic aid to clinicians from various clinical settings for the screening of optic disc abnormalities harboring potentially sight- or life-threatening neurological conditions.
dc.identifier.citationJournal of Neuro-Ophthalmology Vol.43 No.2 (2023) , 159-167
dc.identifier.doi10.1097/WNO.0000000000001800
dc.identifier.eissn15365166
dc.identifier.issn10708022
dc.identifier.pmid36719740
dc.identifier.scopus2-s2.0-85159737109
dc.identifier.urihttps://repository.li.mahidol.ac.th/handle/20.500.14594/83025
dc.rights.holderSCOPUS
dc.subjectMedicine
dc.titleDeep Learning System Outperforms Clinicians in Identifying Optic Disc Abnormalities
dc.typeArticle
mu.datasource.scopushttps://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85159737109&origin=inward
oaire.citation.endPage167
oaire.citation.issue2
oaire.citation.startPage159
oaire.citation.titleJournal of Neuro-Ophthalmology
oaire.citation.volume43
oairecerif.author.affiliationFarabi Eye Hospital
oairecerif.author.affiliationIstituto delle Scienze Neurologiche di Bologna
oairecerif.author.affiliationSeoul National University Bundang Hospital
oairecerif.author.affiliationDuke-NUS Medical School
oairecerif.author.affiliationUniversidade de Coimbra, Faculdade de Medicina
oairecerif.author.affiliationFondation Adolphe de Rothschild
oairecerif.author.affiliationAlma Mater Studiorum Università di Bologna
oairecerif.author.affiliationNUS Yong Loo Lin School of Medicine
oairecerif.author.affiliationMedical Research Foundation, Chennai
oairecerif.author.affiliationUniversität Freiburg
oairecerif.author.affiliationUniversity of California, Berkeley
oairecerif.author.affiliationSingapore Eye Research Institute
oairecerif.author.affiliationUniversite Catholique de Lille
oairecerif.author.affiliationSun Yat-Sen University
oairecerif.author.affiliationCentro Hospitalar e Universitário de Coimbra
oairecerif.author.affiliationSUNY Upstate Medical University
oairecerif.author.affiliationSingapore General Hospital
oairecerif.author.affiliationUniversität Heidelberg
oairecerif.author.affiliationMoorfields Eye Hospital NHS Foundation Trust
oairecerif.author.affiliationFaculty of Medicine Ramathibodi Hospital, Mahidol University
oairecerif.author.affiliationHong Kong Eye Hospital
oairecerif.author.affiliationSave Sight Institute
oairecerif.author.affiliationUCL Institute of Ophthalmology
oairecerif.author.affiliationHôpitaux Universitaires de Genève
oairecerif.author.affiliationCHU Angers
oairecerif.author.affiliationSingapore National Eye Centre
oairecerif.author.affiliationRigshospitalet
oairecerif.author.affiliationCentre Hospitalier Universitaire de Grenoble
oairecerif.author.affiliationMayo Clinic
oairecerif.author.affiliationA-Star, Institute of High Performance Computing
oairecerif.author.affiliationGroupe Hospitalier Pellegrin
oairecerif.author.affiliationChinese University of Hong Kong
oairecerif.author.affiliationEmory University School of Medicine
oairecerif.author.affiliationJohns Hopkins School of Medicine
oairecerif.author.affiliationAmerican Eye Center

Files

Collections