References

1. Foundational & Cross-Cutting Reviews (AI in Ophthalmology)

• Ting DSW, et al. Artificial intelligence and deep learning in ophthalmology. Nature Biomedical Engineering, 2019. https://pubmed.ncbi.nlm.nih.gov/30361278 Foundational review covering AI applications, translation challenges, and clinical deployment issues.

Supporting / contextual

• Topol EJ.  High-performance medicine: the convergence of human and artificial intelligence.  Nature Medicine, 2019.  https://www.nature.com/articles/s41591-018-0300-7 Conceptual framing of AI as decision support rather than replacement.

2. Keratoconus & Corneal Disease (Detection, Screening, Limitations)

• Goodman D, Zhu AY.  Utility of artificial intelligence in the diagnosis and management of keratoconus: a systematic review Front Ophthalmol (Lausanne) 2024.  https://pmc.ncbi.nlm.nih.gov/articles/PMC11182163/ Comprehensive, high-quality systematic review covering detection, screening, limitations, and lack of generalisation.

Supporting / contextual

• Kuo IC, et al.  Detection of keratoconus with deep learning using corneal topography. Ophthalmology, 2020. https://pmc.ncbi.nlm.nih.gov/articles/PMC7533740/  Representative single-centre deep learning study for manifest keratoconus detection.
• Cochrane Eyes and Vision Group.  Artificial intelligence for diagnosing keratoconus.  https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD014911.pub2/full Cochrane Database (Corner / Commentary), 2023.  Highlights promise but emphasises methodological limitations and need for external validation.

3. Cataract Surgery — IOL Power Calculation (AI / ML Formulas)

• Darcy K, et al. Assessment of the accuracy of new and updated intraocular lens power calculation formulas in 10,930 eyes from the UK National Health Service. Journal of Cataract & Refractive Surgery, 2020.  https://pubmed.ncbi.nlm.nih.gov/32050225/  Peer-reviewed description and validation of the Kane formula and others
• Melles RB, et al.  Accuracy of intraocular lens calculation formulas. Ophthalmology, 2018. https://www.aaojournal.org/article/S0161-6420(18)30311-1/fulltext Large retrospective comparison of multiple IOL formulas; frequently cited benchmark.
• Debellemanière G, Saad A, Gatinel D. The PEARL-DGS Formula: The Development of an Open-Source Machine Learning–Based Thick IOL Calculation Formula.
  American Journal of Ophthalmology, 2021; https://www.researchgate.net/publication/381983735_The_PEARL-DGS_Formula. This study describes the methodology behind PEARL-DGS, a thick-lens, machine learning-augmented IOL power formula, and compares its performance (standard deviation of prediction errors) to contemporary formulas

Supporting / contextual

• Savini G, et al.  Recent advances in intraocular lens power calculation. Eye, 2020. https://www.nature.com/articles/s41433-020-0779-7 Critical review of modern formulas, limitations in extreme eyes and post-refractive surgery.
• Hill W.  Hill-RBF method for IOL power calculation.  (Referenced via peer-reviewed comparisons rather than original proprietary description.)

4. Refractive Surgery — Decision Support & Risk Prediction

• Yoo TK, et al. Adopting machine learning to automatically identify candidate patients for corneal refractive surgery  Digital Medicine, 2019. https://www.nature.com/articles/s41746-019-0135-8 Well-known study demonstrating ML-based decision support outperforming traditional screening metrics.

Supporting / contextual

• Choi JY, et al.  Artificial intelligence in refractive surgery. Annals of Eye Science, 2023. https://aes.amegroups.org/article/view/8271/html Narrative review contextualising AI use in refractive surgery planning and outcomes.
• Rampat R, et al. Corneal ectasia following LASIK. Eye, 2019. https://www.nature.com/articles/s41433-019-0407-8 Not AI-specific, but establishes the clinical problem AI attempts to address.

5. Other Anterior Segment Conditions

• Fang X, Deshmukh M, et al. Deep learning algorithms for automatic detection of pterygium using anterior segment photographs from slit-lamp and hand-held cameras. British Journal of Ophthalmology. 2022.  https://pubmed.ncbi.nlm.nih.gov/34244208/
• Ong ZZ, Sadek Y, et al.. Diagnostic performance of deep learning for infectious keratitis: a systematic review and meta-analysis.
  EClinicalMedicine. 2024;  https://pubmed.ncbi.nlm.nih.gov/39469534/ Strong multicentre work showing triage potential, especially for telemedicine.

6. Dataset Infrastructure & Data Limitations

• Niestrata M, et al. Global review of publicly available image datasets for the anterior segment of the eye. Journal of Cataract & Refractive Surgery, 2024.
  https://journals.lww.com/jcrs/Fulltext/2024/11000/Global_review_of_publicly_available_image_datasets.14.aspx
  Systematic mapping of existing datasets and access limitations.
• Niestrata M, et al. Building on the Foundations: Creating an AI-Assisted instead of a Manual Approach to Building a Library of Publicly Accessible Anterior Segment Image Datasets. ESCRS Digital Health SIG commissioned project, 2026. Coming Soon