Inteligência artificial para diagnóstico e encaminhamento de catarata usando dados reais
DOI:
https://doi.org/10.59681/2175-4411.v17.2025.1441Palavras-chave:
Inteligência Artificial, Diagnóstico, CatarataResumo
O uso de inteligência artificial (IA) para aplicações oftalmológicas têm mostrado resultados promissores, entretanto, seu desempenho depende do grupo populacional amostrado e precisa ser avaliado em cenários reais. Propomos o uso de IA para diagnóstico de catarata e encaminhamento para especialista usando uma base de dados real, composta por 2642 imagens oculares, de um serviço de telemedicina no sul do Brasil. Nossa solução adota um modelo composto para aprimorar o desempenho dos classificadores. Os melhores resultados mostram acurácia de 90,6% para diagnóstico, com área correspondente sob a curva de operação característica do receptor (ROC AUC) de 96,7%. A acurácia para encaminhamento de catarata para cirurgia foi de 86,5%, com ROC AUC de 94,3%. Os dados obtidos apontam que o uso de um modelo composto treinado com uma base clínica heterogênea real permitiu que nossa solução atingisse desempenho superior a outros trabalhos da literatura quando avaliados com dados do mundo real.
Downloads
Referências
Gutierrez L, Lim JS, Foo LL, Ng WY, Yip M, Lim GYS, et al. Application of artificial intelligence in cataract management: current and future directions. Eye Vis. 2022;9(1):1–11. DOI: https://doi.org/10.1186/s40662-021-00273-z
Gunasekeran DV, Wong TY. Artificial intelligence in ophthalmology in 2020: A technology on the cusp for translation and implementation. Asia-Pacific J Ophthalmol. 2020;9(2):61–6. DOI: https://doi.org/10.1097/01.APO.0000656984.56467.2c
Ting DSJ, Foo VHX, Yang LWY, Sia JT, Ang M, Lin H, et al. Artificial intelligence for anterior segment diseases: Emerging applications in ophthalmology. British Journal of Ophthalmology. 2021. DOI: https://doi.org/10.1136/bjophthalmol-2019-315651
Lin H, Liu L, Wu X., Artificial Intelligence for Cataract Management. Artificial Intelligence in Ophthalmology. 2021:203–6. DOI: https://doi.org/10.1007/978-3-030-78601-4_16
Tong Y, Lu W, Yu Y, Shen Y. Application of machine learning in ophthalmic imaging modalities. Eye Vis. 2020;7(1). DOI: https://doi.org/10.1186/s40662-020-00183-6
Zhang XQ, Hu Y, Xiao, ZJ, Fang JS, Higashita R, Liu J, Machine learning for cataract classification/grading on ophthalmic imaging modalities: A survey. Mach. Intell. Res. 2022;19(3):184–208. DOI: https://doi.org/10.1007/s11633-022-1329-0
Orfao J, van der Haar D. A Comparison of Computer Vision Methods for the Combined Detection of Glaucoma, Diabetic Retinopathy and Cataracts. In: Lecture Notes in Computer Science. 2021. DOI: https://doi.org/10.1007/978-3-030-80432-9_3
Alam M, Hallak JA. AI-automated referral for patients with visual impairment. Lancet Digit Heal [Internet]. 2021;3(1):e2–3 DOI: https://doi.org/10.1016/S2589-7500(20)30286-7
Wu X, Xu D, Ma T, Li ZH, Ye Z, Wang F, et al. Artificial Intelligence Model for Antiinterference Cataract Automatic Diagnosis: A Diagnostic Accuracy Study. Front Cell Dev Biol. 2022;10(July):1–11. DOI: https://doi.org/10.3389/fcell.2022.906042
Wu X, Huang Y, Liu Z, Lai W, Long E, Zhang K, et al. Universal artificial intelligence platform for collaborative management of cataracts. British Journal of Ophthalmology. 2019;103(11):1553–60. DOI: https://doi.org/10.1136/bjophthalmol-2019-314729
Pratap T, Dhulipalla VR, Kokil P. Computer-aided Cataract Diagnosis with Fundus Retinal Images under Noisy Conditions. IEEE International Conference on Interdisciplinary Approaches in Technology and Management for Social Innovation (IATMSI) 2024:1–18. DOI: https://doi.org/10.1109/IATMSI60426.2024.10502479
Kassoff A, Kassoff J, Mehu M, Buehler JA, Eglow M, Kaufman F, et al. The Age-Related Eye Disease Study (AREDS) system for classifying cataracts from photographs: AREDS Report No. 4. Am J Ophthalmol. 2001;131(2):167–75. DOI: https://doi.org/10.1016/S0002-9394(00)00732-7
Maaliw RR, Alon AS, Lagman AC, Garcia MB, Abante MV, Belleza RC, et al.. Cataract Detection and Grading Using Ensemble Neural Networks and Transfer Learning. IEEE Annual Information Technology, Electronics and Mobile Communication Conference. 2022:1-9. DOI: https://doi.org/10.1109/IEMCON56893.2022.9946550
Tham YC, Anees A, Zhang L, Goh JHL, Rim TH, Nusinovici S, et al. Referral for disease-related visual impairment using retinal photograph-based deep learning: a proof-of-concept, model development study. Lancet Digit Heal [Internet]. 2021;3(1):e29–40. DOI: https://doi.org/10.1016/S2589-7500(20)30271-5
Jiang L, Huang D, Liu M, Yang W. Beyond Synthetic Noise: Deep Learning on Controlled Noisy Labels. In: 37th International Conference on Machine Learning, ICML 2020. 2020.
Yahata E, Winnikow EP, Suyama R, Simoes PW. Explainability in Machine Learning Predictive Models in Breast Cancer. Journal of Health Informatics, 2022;15(Especial). DOI: https://doi.org/10.59681/2175-4411.v15.iEspecial.2023.1090
Fawcett T. An introduction to ROC analysis. Pattern Recognit Lett. 2006;27(8):861–74. DOI: https://doi.org/10.1016/j.patrec.2005.10.010
Chauhan K, Kashish, Dagar K, Yadav RK. Cataract detection from eye fundus image using an ensemble of transfer learning models. Intern Conference on Advance Computing and Innovative Technologies in Engineering. 2022:2194-98. DOI: https://doi.org/10.1109/ICACITE53722.2022.9823638
Chollet F. Deep Learning with Python Manning. 2018.
Gao X, Wong DWK, Ng TT, Cheung CYL, Cheng CY, Wong TY. Automatic grading of cortical and PSC cataracts using retroillumination lens images. Lect Notes Comput Sci . 2013;7725 LNCS(PART 2):256–67. DOI: https://doi.org/10.1007/978-3-642-37444-9_20
de Araujo AL, Moreira TC, Rados DRV, Gross PB, Bastos CGM, Katz N, et al. The use of telemedicine to support Brazilian primary care physicians in managing eye conditions: The Teleoftalmo project. PLoS One. 2020;15(4):1–12. DOI: https://doi.org/10.1371/journal.pone.0231034
Carneiro T, Da Nobrega RVM, Nepomuceno T, Bian G Bin, De Albuquerque VHC, Filho PPR. Performance Analysis of Google Colaboratory as a Tool for Accelerating Deep Learning Applications. IEEE Access. 2018;6:61677–85. DOI: https://doi.org/10.1109/ACCESS.2018.2874767
Ramlan LA, Zaki WMDW, Mutalib HÁ, Hussain A, Mustapha A. Cataract Detection using Pupil Patch Classification and Ruled-based System in Anterior Segment Photographed Images. Symposium on Computer Applications & Industrial Electronics. 2023: 124-9. DOI: https://doi.org/10.1109/ISCAIE57739.2023.10165004
Kaur N, Gupta G. Cataract Disease Diagnosis Using SURF Features and Pre-Trained Variants of an EfficientNet Model: Comparative Analysis. Inter Conf Comp, Autom and Know Manag (ICCAKM) 2023:1-6. DOI: https://doi.org/10.1109/ICCAKM58659.2023.10449495
Shimizu, E, Tanji M, Nakayama S, Ishikawa T, Agata N, Yokoiwa R , Nishimura H, et al. AI-based diagnosis of nuclear cataract from slit-lamp videos. Sci Rep 13, 22046 (2023). DOI: https://doi.org/10.1038/s41598-023-49563-7
Downloads
Publicado
Como Citar
Edição
Seção
Licença
Copyright (c) 2025 Marcelo Negreiros, Ronaldo Husemann, Valter Roesler, Aline L. de Araujo, Dimitris Rucks Varvaki Rados , Felipe C. Cabral, Natan Katz
Este trabalho está licenciado sob uma licença Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
A submissão de um artigo ao Journal of Health Informatics é entendida como exclusiva e que não está sendo considerada para publicação em outra revista (Declaração de Exclusividade). A permissão dos autores para a publicação de seu artigo no J. Health Inform. implica na exclusiva autorização concedida aos editores para incluí-lo na revista. Ao submeter um artigo, ao autor será solicitada a permissão de um Termo de Transferência de Direitos de Publicação. Uma mensagem eletrônica será enviada ao autor correspondente confirmando o recibo do manuscrito e o aceite da Declaração de Direito de Publicação.
