Generative adversarial networks (GAN) for pre-dilation retinal photograph quality enhancement.

Jocelyn Hui Lin Goh^1,2,3, Mingrui Tan⁴, Xiaofeng Lei⁴, Shaohua Li⁴, Sarah Shwu Huey Tan¹, Haslina Hamzah^1,5, Gabriel Dawei Yang¹, Zhi Da Soh¹, Tyler Hyungtaek Rim^1,6,7, Gavin Siew Wei Tan^1,5,7, Rick Siow Mong Goh⁴, Yong Liu⁴, Ching-Yu Cheng^1,2,3,7, Yih Chung Tham^8,9,10,11

¹Singapore Eye Research Institute, Singapore National Eye Centre, Singapore, Singapore.

²Centre for Innovation and Precision Eye Health, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore.

³Department of Ophthalmology, Yong Loo Lin School of Medicine, National University of Singapore and National University Health System, Singapore, Singapore.

⁴Institute of High Performance Computing (IHPC), Agency for Science, Technology and Research (A*STAR), Singapore, Singapore.

⁵SNEC Ocular Reading Centre, Singapore National Eye Centre, Singapore, Singapore.

⁶Mediwhale Inc, Seoul, Korea.

⁷Ophthalmology and Visual Sciences Academic Clinical Program (Eye ACP), Duke-NUS Medical School Singapore, Singapore, Singapore.

⁸Singapore Eye Research Institute, Singapore National Eye Centre, Singapore, Singapore. thamyc@nus.edu.sg.

⁹Centre for Innovation and Precision Eye Health, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore. thamyc@nus.edu.sg.

¹⁰Department of Ophthalmology, Yong Loo Lin School of Medicine, National University of Singapore and National University Health System, Singapore, Singapore. thamyc@nus.edu.sg.

¹¹Ophthalmology and Visual Sciences Academic Clinical Program (Eye ACP), Duke-NUS Medical School Singapore, Singapore, Singapore. thamyc@nus.edu.sg.

PubMed DOI

Abstract

PURPOSE: Retinal image quality significantly affects the performance of diagnostic artificial intelligence (AI) models and is typically improved with pupil dilation in clinical settings. However, in real-world settings where dilation is not feasible, suboptimal image quality remains a challenge for AI deployment. In this study, we fine-tuned a CofeNet model to enhance the quality of undilated reti…

Create a free account to keep reading

Free members get 10 full research views every month across publications, clinical trials, FDA clearances, adverse events, and NIH grants. No credit card required.

Create Free Account Sign In

Want unlimited research access? See Pro plans

Data Accuracy Notice: Research intelligence on Health AI Central is aggregated from public sources (PubMed, ClinicalTrials.gov, FDA, NIH, CMS, and others) and refreshed nightly. Classifications and derived metrics are produced by automated methods described in our Methodology. We recommend verifying critical data points against the primary sources before making decisions.