Peter Awad, Mark Awad, John Awad
Introduction
Artificial intelligence (AI) has evolved from an emerging research interest to an increasingly important component of modern ophthalmic practice (2). Machine learning algorithms have demonstrated impressive performance in detecting diabetic retinopathy, glaucoma, age-related macular degeneration (AMD) and other ocular diseases using retinal imaging and optical coherence tomography (OCT) (1,5). Several AI systems have now received regulatory approval and are being incorporated into screening programmes and clinical workflows (4).
While advances in AI promise earlier diagnosis, improved efficiency and greater access to specialist care, their successful implementation depends not only on technological performance but also on clinicians’ ability to understand their capabilities and limitations (6). The widespread adoption of AI therefore creates a new educational responsibility: future ophthalmologists must develop sufficient AI literacy to safely evaluate and integrate these technologies into patient care (3,7).
This commentary argues that AI literacy should be recognised as a fundamental professional competency within ophthalmology training rather than an optional specialist interest.
AI Is Becoming Part of Routine Ophthalmic Practice
Ophthalmology is uniquely positioned for AI implementation because clinical decision-making relies heavily on digital imaging. High-quality fundus photography, OCT imaging, visual field analysis and slit lamp imaging provide structured datasets that are particularly well suited to machine learning applications (2).
Evidence accumulated over the past decade suggests that AI systems can achieve diagnostic performance comparable to expert clinicians in the detection of diabetic retinopathy (1). Subsequent studies have demonstrated similar promise in the diagnosis of other retinal diseases using OCT and other imaging modalities (5). Beyond diagnosis, AI applications continue to expand into disease progression prediction, surgical planning, automated image segmentation, workflow optimisation and administrative support.
As AI becomes increasingly embedded within ophthalmology, clinicians must be able to critically evaluate AI-generated outputs rather than simply accept them at face value. Factors including dataset bias, image quality and limitations in external validation can all influence algorithm performance and affect clinical decision-making (6,8). Understanding these limitations is therefore essential to ensuring the safe and effective use of AI in patient care (7).
The growing adoption of AI represents not only a technological advancement but also an educational challenge. As with evidence-based medicine or imaging interpretation, ophthalmologists require a foundational understanding of AI to use these technologies responsibly and maintain professional accountability in an increasingly digital healthcare system.
Defining AI Literacy
AI literacy extends beyond the ability to operate AI-powered software; rather it encompasses the knowledge and skills required to understand how AI systems are developed, recognise their capability and limitations, critically evaluate their outputs and incorporate them appropriately into clinical decision-making (3,7).
Importantly, AI literacy does not require ophthalmologists to become data scientists or software engineers. Instead, clinicians should possess a working understanding of key concepts such as machine learning, algorithm validation, bias and the appropriate interpretation of AI-generated recommendations (6,8). This knowledge enables clinicians to identify situations in which AI may enhance decision-making while recognising circumstances in which algorithmic output should be interpreted with caution.
As AI becomes increasingly integrated into routine ophthalmic practice, clinical responsibility remains with the clinician regardless of whether AI contributes to the decision-making process (7). Consequently, AI literacy should be viewed not as a technical skill but as a professional competency that supports safe, ethical and evidence-based patient care. Developing these competencies will enable ophthalmologists to use AI confidently while maintaining accountability for clinical decision-making in an evolving digital healthcare system (3,9).
AI Literacy as a Patient Safety Issue
The integration of AI into ophthalmic practise has important implications for patient safety. Although AI systems can improve diagnostic accuracy and efficiency, they remain susceptible to limitations including dataset bias, poor image quality and reduced performance in populations that differ from those used during algorithm development (6,8). AI should therefore be regarded as a decision support tool rather than a replacement for clinical judgement.
AI literacy enables ophthalmologists to critically evaluate AI-generated recommendations, recognise situations in which algorithmic outputs may be unreliable and communicate the role and limitations of AI to patients (7). As responsibility for clinical decisions ultimately remains with the treating clinician, maintaining patient safety will depend not only on the accuracy of AI systems but also on clinicians possessing the knowledge to use them appropriately (7,9). Developing AI literacy should therefore be recognised as an essential component of safe, ethical and accountable clinical practice.
Preparing Future Ophthalmologists
As AI becomes increasingly integrated into ophthalmic practice, AI literacy should be regarded as a core component of postgraduate training rather than an optional area of specialist interest. Future ophthalmologists will require sufficient understanding to critically evaluate AI tools, recognise their limitations and incorporate them appropriately into clinical practice (3,7).
Training should focus on practical competencies rather than technical expertise. Trainees should understand the principles of machine learning, recognise common sources of algorithmic bias, interpret measures of diagnostic performance and appreciate the importance of external validation in clinical governance (6,8). Equally important is the ability to communicate the role of AI to patients and to maintain professional accountability when AI informs clinical decisions (7,9).
Embedding these competencies during training will better prepare future ophthalmologists to adopt emerging technologies safely and responsibly while ensuring that clinical judgement remains central to patient care.
Integrating AI into Existing Curricula
AI literacy does not require the creation of a separate training programme. Instead, its principles can be incorporated into existing ophthalmology curricula through teaching on evidence-based medicine, clinical governance, image interpretation and research methodology (3,6). Practical learning opportunities, including critical appraisal of AI studies, case-based discussions and supervised use of AI-assisted clinical tools, would allow trainees to develop competencies within routine clinical education.
Professional organisations, training bodies and medical schools will play an important role in defining learning outcomes and ensuring that educational programmes evolve alongside advances in AI (9). By embedding AI literacy within existing educational frameworks, future ophthalmologists can be equipped to adopt emerging technology safely and effectively without adding a substantial burden to already demanding training programmes.
Challenges and Considerations
Integrating AI literacy into ophthalmology training will present several challenges. Curricula are already very demanding, and the rapid pace of AI development means educational content will require regular review and updating (3). Furthermore, not all trainees or training centres will have equal access to AI-enabled technologies, creating the potential for variation in learning (9).
Despite these challenges, delaying AI education is unlikely to be sustainable. As AI becomes increasingly embedded within clinical practice, ensuring that all ophthalmologists possess a foundational understanding of its capabilities and limitations will be essential to maintaining safe, equitable and high-quality patient care (6,7).
Conclusion
AI is rapidly becoming an integral component of ophthalmic practice, creating new expectations of the knowledge and skills required of future ophthalmologists. While AI has the potential to improve diagnostic accuracy, efficiency and patient care, its safe implementation depends on clinicians who can critically evaluate its capabilities and limitations (7,10).
AI literacy should therefore be recognised as a core professional competency within ophthalmology training rather than an optional area of specialist interest. Embedding AI literacy within existing educational frameworks will help ensure that future ophthalmologists are equipped to adopt emerging technologies safely, responsibly and in the best interests of their patients.
References
- Gulshan, V., Peng, L., Coram, M., Stumpe, M. C., Wu, D., Narayanaswamy, A., Venugopalan, S., Widner, K., Madams, T., Cuadros, J., Kim, R., Raman, R., Nelson, P. C., Mega, J. L., & Webster, D. R. (2016). Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs. JAMA, 316(22), 2402–2410. https://doi.org/10.1001/jama.2016.17216
- Ting, D. S. W., Pasquale, L. R., Peng, L., Campbell, J. P., Lee, A. Y., Raman, R., Tan, G. S. W., Schmetterer, L., Keane, P. A., & Wong, T. Y. (2019). Artificial intelligence and deep learning in ophthalmology. The British journal of ophthalmology, 103(2), 167–175. https://doi.org/10.1136/bjophthalmol-2018-313173
- Topol, E.J. (2019) Deep Medicine: How Artificial Intelligence Can Make Healthcare Human Again. Basic Books.
- Abràmoff, Michael & Lavin, Philip & Birch, Michele & Shah, Nilay & Folk, James. (2018). Pivotal trial of an autonomous AI-based diagnostic system for detection of diabetic retinopathy in primary care offices. npj Digital Medicine. 1. 10.1038/s41746-018-0040-6.
- De Fauw, J., Ledsam, J. R., Romera-Paredes, B., Nikolov, S., Tomasev, N., Blackwell, S., Askham, H., Glorot, X., O’Donoghue, B., Visentin, D., van den Driessche, G., Lakshminarayanan, B., Meyer, C., Mackinder, F., Bouton, S., Ayoub, K., Chopra, R., King, D., Karthikesalingam, A., Hughes, C. O., … Ronneberger, O. (2018). Clinically applicable deep learning for diagnosis and referral in retinal disease. Nature medicine, 24(9), 1342–1350. https://doi.org/10.1038/s41591-018-0107-6
- Kelly, C. J., Karthikesalingam, A., Suleyman, M., Corrado, G., & King, D. (2019). Key challenges for delivering clinical impact with artificial intelligence. BMC medicine, 17(1), 195. https://doi.org/10.1186/s12916-019-1426-2
- Char, D. S., Shah, N. H., & Magnus, D. (2018). Implementing Machine Learning in Health Care – Addressing Ethical Challenges. The New England journal of medicine, 378(11), 981–983. https://doi.org/10.1056/NEJMp1714229
- Liu, X., Cruz Rivera, S., Moher, D., Calvert, M. J., Denniston, A. K., & SPIRIT-AI and CONSORT-AI Working Group (2020). Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI extension. The Lancet. Digital health, 2(10), e537–e548. https://doi.org/10.1016/S2589-7500(20)30218-1
- World Health Organisation. Ethics and Governance of Artificial Intelligence for Health. Geneva: World Health Organisation; 2021
- Ken Masters (2023) Ethical use of Artificial Intelligence in Health Professions Education: AMEE Guide No. 158, Medical Teacher, 45:6, 574-584, DOI: 10.1080/0142159X.2023.2186203 https://doi.org/10.1080/0142159X.2023.2186203