{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,24]],"date-time":"2026-07-24T18:02:26Z","timestamp":1784916146415,"version":"3.55.0"},"reference-count":28,"publisher":"Public Library of Science (PLoS)","issue":"7","license":[{"start":{"date-parts":[[2026,7,24]],"date-time":"2026-07-24T00:00:00Z","timestamp":1784851200000},"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>Retinitis pigmentosa (RP) is a group of inherited retinal diseases that are caused by genetic defects that lead to progressive photoreceptor loss and eventual blindness. Early diagnosis would be helpful for effective management of the disease; however, many patients stay unaware of early symptoms. Meanwhile, fundus images are widely obtained during routine medical checkups but are underused for detecting RP. This study explores the effectiveness of finetuning deep learning models, pre-trained for general visual tasks, to identify RP from color fundus images. The dataset comprised 321 color fundus images from 201 Japanese subjects at Keio University Hospital, including 200 images from 107 patients with retinitis pigmentosa and 121 images from 94 non-retinitis pigmentosa subjects. Multiple images were available for some subjects. Using transfer learning, pretrained convolutional neural network models -VGG16, Resnet50, and InceptionV3- were finetuned to detect RP. As a result, Inception V3 achieved the best accuracy of 96.97%, which matches the average diagnostic accuracy of ophthalmologists. Gradient-weighted Class Activation Mapping (Grad-CAM) suggested that the model attended to clinically relevant fundus regions, including the peripheral retina and posterior pole, which may reflect features such as peripheral degenerative changes and retinal vascular attenuation. These findings support the potential interpretability of the finetuned model and suggest that deep learning may assist ophthalmologists in RP screening as a supportive tool.<\/jats:p>","DOI":"10.1371\/journal.pone.0354452","type":"journal-article","created":{"date-parts":[[2026,7,24]],"date-time":"2026-07-24T17:34:49Z","timestamp":1784914489000},"page":"e0354452","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Fundus image analysis of retinitis pigmentosa using artificial intelligence"],"prefix":"10.1371","volume":"21","author":[{"given":"Saki","family":"Ubukata","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kanato","family":"Masayoshi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yusaku","family":"Katada","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lizhu","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nobuhiro","family":"Ozawa","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mari","family":"Ibuki","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2892-9810","authenticated-orcid":true,"given":"Kazuno","family":"Negishi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5457-2720","authenticated-orcid":true,"given":"Toshihide","family":"Kurihara","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,7,24]]},"reference":[{"issue":"4","key":"pone.0354452.ref001","doi-asserted-by":"crossref","first-page":"260","DOI":"10.2174\/138920211795860071","article-title":"Good epidemiologic practice in retinitis pigmentosa: from phenotyping to biobanking","volume":"12","author":"M Chizzolini","year":"2011","journal-title":"Curr Genomics"},{"issue":"3","key":"pone.0354452.ref002","doi-asserted-by":"crossref","first-page":"147","DOI":"10.1097\/ICU.0000000000000660","article-title":"Gene therapy beyond luxturna: a new horizon of the treatment for inherited retinal disease","volume":"31","author":"DA Prado","year":"2020","journal-title":"Curr Opin Ophthalmol"},{"issue":"6","key":"pone.0354452.ref003","doi-asserted-by":"crossref","first-page":"671","DOI":"10.1080\/13816810.2018.1533027","article-title":"Gene therapy for RPE65-related retinal disease","volume":"39","author":"V Miraldi Utz","year":"2018","journal-title":"Ophthalmic Genet"},{"issue":"2","key":"pone.0354452.ref004","doi-asserted-by":"crossref","first-page":"167","DOI":"10.1136\/bjophthalmol-2018-313173","article-title":"Artificial intelligence and deep learning in ophthalmology","volume":"103","author":"DSW Ting","year":"2019","journal-title":"Br J Ophthalmol"},{"issue":"22","key":"pone.0354452.ref005","doi-asserted-by":"crossref","first-page":"2402","DOI":"10.1001\/jama.2016.17216","article-title":"Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs","volume":"316","author":"V Gulshan","year":"2016","journal-title":"JAMA"},{"key":"pone.0354452.ref006","doi-asserted-by":"crossref","first-page":"101644","DOI":"10.1016\/j.imu.2025.101644","article-title":"Glaucoma identification with retinal fundus images using deep learning: systematic review","volume":"56","author":"D Meedeniya","year":"2025","journal-title":"Informatics in Medicine Unlocked"},{"issue":"9","key":"pone.0354452.ref007","doi-asserted-by":"crossref","first-page":"1272","DOI":"10.1136\/bjophthalmol-2020-318544","article-title":"Prediction of causative genes in inherited retinal disorder from fundus photography and autofluorescence imaging using deep learning techniques","volume":"105","author":"Y Fujinami-Yokokawa","year":"2021","journal-title":"Br J Ophthalmol"},{"issue":"6","key":"pone.0354452.ref008","doi-asserted-by":"crossref","first-page":"967","DOI":"10.1038\/s42256-025-01040-8","article-title":"Next-generation phenotyping of inherited retinal diseases from multimodal imaging with Eye2Gene","volume":"7","author":"N Pontikos","year":"2025","journal-title":"Nat Mach Intell"},{"issue":"23","key":"pone.0354452.ref009","doi-asserted-by":"crossref","first-page":"5481","DOI":"10.3390\/jcm10235481","article-title":"Prediction of different eye diseases based on fundus photography via deep transfer learning","volume":"10","author":"C Guo","year":"2021","journal-title":"J Clin Med"},{"issue":"4","key":"pone.0354452.ref010","doi-asserted-by":"crossref","first-page":"948","DOI":"10.1007\/s10278-021-00479-6","article-title":"Artificial intelligence-assisted early detection of retinitis pigmentosa - the most common inherited retinal degeneration","volume":"34","author":"T-C Chen","year":"2021","journal-title":"J Digit Imaging"},{"issue":"4","key":"pone.0354452.ref011","doi-asserted-by":"crossref","first-page":"305","DOI":"10.1001\/jamaophthalmol.2022.6393","article-title":"Estimation of visual function using deep learning from ultra-widefield fundus images of eyes with retinitis pigmentosa","volume":"141","author":"D Nagasato","year":"2023","journal-title":"JAMA Ophthalmol"},{"issue":"10","key":"pone.0354452.ref012","doi-asserted-by":"crossref","first-page":"1484","DOI":"10.1136\/bjo-2021-320897","article-title":"Prediction of visual impairment in retinitis pigmentosa using deep learning and multimodal fundus images","volume":"107","author":"TYA Liu","year":"2023","journal-title":"Br J Ophthalmol"},{"key":"pone.0354452.ref013","author":"K Simonyan","year":"2014","journal-title":"Very deep convolutional networks for large-scale image recognition. arXiv"},{"key":"pone.0354452.ref014","author":"C Szegedy","year":"2015"},{"key":"pone.0354452.ref015","volume-title":"Resnet in resnet: generalizing residual architectures","author":"S Targ","year":"2016"},{"key":"pone.0354452.ref016","doi-asserted-by":"crossref","unstructured":"Deng J, Dong W, Socher R, Li L-J, Kai Li, Li Fei-Fei. ImageNet: a large-scale hierarchical image database. In: 2009 IEEE Conference on Computer Vision and Pattern Recognition, 2009. 248\u201355. https:\/\/doi.org\/10.1109\/cvpr.2009.5206848","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"pone.0354452.ref017","doi-asserted-by":"crossref","DOI":"10.7717\/peerj.6900","article-title":"Accuracy of a deep convolutional neural network in detection of retinitis pigmentosa on ultrawide-field images","volume":"7","author":"H Masumoto","year":"2019","journal-title":"PeerJ"},{"issue":"11","key":"pone.0354452.ref018","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0335419","article-title":"Broad-spectrum eye disease classification using a deep learning-based tailored software lens","volume":"20","author":"C Rieck","year":"2025","journal-title":"PLoS One"},{"key":"pone.0354452.ref019","doi-asserted-by":"crossref","first-page":"1630667","DOI":"10.3389\/fcell.2025.1630667","article-title":"Deep learning-based classification of multiple fundus diseases using ultra-widefield images","volume":"13","author":"M-M Duan","year":"2025","journal-title":"Front Cell Dev Biol"},{"issue":"1","key":"pone.0354452.ref020","doi-asserted-by":"crossref","first-page":"22523","DOI":"10.1038\/s41598-025-04648-3","article-title":"A deep learning model for diagnosis of inherited retinal diseases","volume":"15","author":"F Jafarbeglou","year":"2025","journal-title":"Sci Rep"},{"issue":"1","key":"pone.0354452.ref021","doi-asserted-by":"crossref","first-page":"43306","DOI":"10.1038\/s41598-025-27367-1","article-title":"A self-supervised learning method for detection of retinitis pigmentosa and Stargardt disease","volume":"15","author":"A Karimi","year":"2025","journal-title":"Sci Rep"},{"issue":"1","key":"pone.0354452.ref022","doi-asserted-by":"crossref","first-page":"137","DOI":"10.1038\/s41746-025-02311-9","article-title":"Leveraging large scale deep learning models for diagnosis and visual outcome prediction in retinitis pigmentosa","volume":"9","author":"T Nagai","year":"2026","journal-title":"NPJ Digit Med"},{"key":"pone.0354452.ref023","article-title":"Deep learning detection of retinitis pigmentosa inheritance forms through synthetic data expansion of a rare disease dataset","author":"EE Hwang","year":"2025","journal-title":"Res Sq"},{"issue":"4","key":"pone.0354452.ref024","doi-asserted-by":"crossref","DOI":"10.1088\/2057-1976\/ade9c7","article-title":"Artificial intelligence techniques in inherited retinal diseases: a review","volume":"11","author":"H Trinh","year":"2025","journal-title":"Biomed Phys Eng Express"},{"issue":"2","key":"pone.0354452.ref025","doi-asserted-by":"crossref","first-page":"255","DOI":"10.1016\/j.survophthal.2024.11.007","article-title":"Applications of artificial intelligence to inherited retinal diseases: a systematic review","volume":"70","author":"M Issa","year":"2025","journal-title":"Surv Ophthalmol"},{"issue":"1","key":"pone.0354452.ref026","doi-asserted-by":"crossref","first-page":"206","DOI":"10.1186\/s12886-026-04740-w","article-title":"Macular hole detection and segmentation on fundus photography using large multimodal generative models for synthetic augmentation","volume":"26","author":"TK Yoo","year":"2026","journal-title":"BMC Ophthalmol"},{"issue":"1","key":"pone.0354452.ref027","doi-asserted-by":"crossref","first-page":"4828","DOI":"10.1038\/s41467-021-25138-w","article-title":"Automatic detection of 39 fundus diseases and conditions in retinal photographs using deep neural networks","volume":"12","author":"L-P Cen","year":"2021","journal-title":"Nat Commun"},{"issue":"3","key":"pone.0354452.ref028","doi-asserted-by":"crossref","first-page":"237","DOI":"10.1007\/s10792-012-9678-2","article-title":"Grading severity in retinitis pigmentosa using clinical assessment, visual acuity, perimetry and optical coherence tomography","volume":"33","author":"HB Smith","year":"2013","journal-title":"Int Ophthalmol"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0354452","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,24]],"date-time":"2026-07-24T17:34:56Z","timestamp":1784914496000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0354452"}},"subtitle":[],"editor":[{"given":"Vahid","family":"Mansouri","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,7,24]]},"references-count":28,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2026,7,24]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0354452","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,24]]}}}