186 268

Cited 6 times in

DeepPDT-Net: predicting the outcome of photodynamic therapy for chronic central serous chorioretinopathy using two-stage multimodal transfer learning

DC Field Value Language
dc.contributor.author김민-
dc.contributor.author김성수-
dc.contributor.author변석호-
dc.contributor.author최은영-
dc.contributor.author이승규-
dc.date.accessioned2022-12-22T05:13:58Z-
dc.date.available2022-12-22T05:13:58Z-
dc.date.issued2022-11-
dc.identifier.urihttps://ir.ymlib.yonsei.ac.kr/handle/22282913/192338-
dc.description.abstractCentral serous chorioretinopathy (CSC), characterized by serous detachment of the macular retina, can cause permanent vision loss in the chronic course. Chronic CSC is generally treated with photodynamic therapy (PDT), which is costly and quite invasive, and the results are unpredictable. In a retrospective case-control study design, we developed a two-stage deep learning model to predict 1-year outcome of PDT using initial multimodal clinical data. The training dataset included 166 eyes with chronic CSC and an additional learning dataset containing 745 healthy control eyes. A pre-trained ResNet50-based convolutional neural network was first trained with normal fundus photographs (FPs) to detect CSC and then adapted to predict CSC treatability through transfer learning. The domain-specific ResNet50 successfully predicted treatable and refractory CSC (accuracy, 83.9%). Then other multimodal clinical data were integrated with the FP deep features using XGBoost.The final combined model (DeepPDT-Net) outperformed the domain-specific ResNet50 (accuracy, 88.0%). The FP deep features had the greatest impact on DeepPDT-Net performance, followed by central foveal thickness and age. In conclusion, DeepPDT-Net could solve the PDT outcome prediction task challenging even to retinal specialists. This two-stage strategy, adopting transfer learning and concatenating multimodal data, can overcome the clinical prediction obstacles arising from insufficient datasets.-
dc.description.statementOfResponsibilityopen-
dc.languageEnglish-
dc.publisherNature Publishing Group-
dc.relation.isPartOfSCIENTIFIC REPORTS-
dc.rightsCC BY-NC-ND 2.0 KR-
dc.subject.MESHCase-Control Studies-
dc.subject.MESHCentral Serous Chorioretinopathy* / diagnostic imaging-
dc.subject.MESHCentral Serous Chorioretinopathy* / drug therapy-
dc.subject.MESHChronic Disease-
dc.subject.MESHFluorescein Angiography-
dc.subject.MESHHumans-
dc.subject.MESHMachine Learning-
dc.subject.MESHPhotochemotherapy* / methods-
dc.subject.MESHPhotosensitizing Agents / therapeutic use-
dc.subject.MESHPorphyrins*-
dc.subject.MESHRetina / diagnostic imaging-
dc.subject.MESHRetrospective Studies-
dc.subject.MESHTomography, Optical Coherence-
dc.subject.MESHVerteporfin / therapeutic use-
dc.subject.MESHVisual Acuity-
dc.titleDeepPDT-Net: predicting the outcome of photodynamic therapy for chronic central serous chorioretinopathy using two-stage multimodal transfer learning-
dc.typeArticle-
dc.contributor.collegeCollege of Medicine (의과대학)-
dc.contributor.departmentDept. of Ophthalmology (안과학교실)-
dc.contributor.googleauthorTae Keun Yoo-
dc.contributor.googleauthorSeo Hee Kim-
dc.contributor.googleauthorMin Kim-
dc.contributor.googleauthorChristopher Seungkyu Lee-
dc.contributor.googleauthorSuk Ho Byeon-
dc.contributor.googleauthorSung Soo Kim-
dc.contributor.googleauthorJinyoung Yeo-
dc.contributor.googleauthorEun Young Choi-
dc.identifier.doi10.1038/s41598-022-22984-6-
dc.contributor.localIdA00455-
dc.contributor.localIdA00571-
dc.contributor.localIdA01849-
dc.contributor.localIdA05056-
dc.contributor.localIdA02913-
dc.relation.journalcodeJ02646-
dc.identifier.eissn2045-2322-
dc.identifier.pmid36333442-
dc.contributor.alternativeNameKim, Min-
dc.contributor.affiliatedAuthor김민-
dc.contributor.affiliatedAuthor김성수-
dc.contributor.affiliatedAuthor변석호-
dc.contributor.affiliatedAuthor최은영-
dc.contributor.affiliatedAuthor이승규-
dc.citation.volume12-
dc.citation.number1-
dc.citation.startPage18689-
dc.identifier.bibliographicCitationSCIENTIFIC REPORTS, Vol.12(1) : 18689, 2022-11-
Appears in Collections:
1. College of Medicine (의과대학) > Dept. of Ophthalmology (안과학교실) > 1. Journal Papers

qrcode

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.