Does lens opacity matter? The effect of cataract on deep learning based cardiovascular disease risk scores from fundus photos.

Jee Eun Lee¹, Jungkyung Cho², Yong-Seok Lee³, Simon Nusinovici⁴, Sahil Thakur⁵, Junseok Park², Dongjin Nam⁶, Jaewon Seo⁷, Wonyoung Seo⁷, Yong Yu Tan⁷, Zhi Da Soh⁴, Can Can Xue⁴, Ching-Yu Cheng⁸, Sunjin Hwang⁹

¹Department of Ophthalmology, Kim's Eye Hospital, Seoul, South Korea.

²Seoul National University College of Medicine, Seoul, South Korea.

³Yonsei University College of Medicine, Seoul, South Korea.

⁴Ophthalmology and Visual Sciences Academic Clinical Program (Eye ACP), Duke-NUS Medical School, Singapore.

⁵Singapore Eye Research Institute, Singapore National Eye Centre, Singapore, Republic of Singapore.

⁶Department of Internal Medicine, Severance Hospital, Yonsei University College of Medicine, Seoul, South Korea.

⁷Mediwhale Inc., Seoul, South Korea.

⁸Ophthalmology and Visual Sciences Academic Clinical Program (Eye ACP), Duke-NUS Medical School, Singapore. Electronic address: chingyu.cheng@duke-nus.edu.sg.

⁹Department of Ophthalmology, Hanyang University Guri Hospital, Guri, South Korea. Electronic address: sunjin1989@hanyang.ac.kr.

PubMed DOI

Abstract

PURPOSE: To investigate the effect of cataracts on a deep learning (DL) model for cardiovascular disease (CVD) risk prediction. METHODS: This retrospective, dual-cohort study analyzed fundus images at baseline, 1, and 6-months post-cataract surgery from a longitudinal cohort (patients who underwent cataract surgery at Hanyang University Guri Hospital [HUGH]) and a cross-sectional replication coho…

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.