Skip to main content
  • Artificial Intelligence

    Artificial intelligence (AI) has been defined as the scientific understanding of the mechanisms underlying thought and intelligent behavior embodied in machines. AI has been incorporated into various industries including healthcare, business, education, and manufacturing. However, AI is relatively new to medicine and ophthalmology, but has had a prominent role in ophthalmology, given the enormous interdependence on digital imaging for clinical decision-making.

    Committee on Artificial Intelligence

    The Academy restructured Information Technology activities within its Committee structure to emphasize the importance and growing relevance of the field. The Committee on Artificial Intelligence was created in 2021 to identify areas of focus for the Academy, to provide resident ophthalmologists and practicing ophthalmologists with pertinent knowledge and resources, and to coordinate activities on AI across the Academy. The chair and members of the Committee are listed here:
    J. Peter Campbell, MD, MPH; Michael Abramoff, MD, PhD; Sally Baxter, MD; Pearse Keane, MD; Judy Kim, MD; Jayashree Kalpathy-Cramer, PhD; Pietro Perona, PhD; Rishi Singh, MD; and Daniel Ting, MD. 


    An important focus of the Committee is providing education and tools for ophthalmologists to learn more about AI and have opportunities to work with datasets for AI. A new partnership between the American College of Radiology (ACR) and American Academy of Ophthalmology will advance AI in ophthalmology. This will expand the use of ACR's Data Science Institute (DSI)'s groundbreaking AI-LABTM platform to include eye care. Use cases and data from the Academy will help ophthalmologists to access real world examples and encourage AI development. Visit ACR AI-LAB (

    Formation of Task Force on Artificial Intelligence

    The Academy created the Task Force on Artificial Intelligence in October 2019 to review the state of the art in the field, to identify unresolved issues and future challenges for the field of ophthalmology, and create a blueprint for what will become a future committee. The chair and members of the task force are listed here: Michael F. Chiang, MD, MA, Michael D. Abramoff, MD, PhD, Peter Campbell, MD, MPH, Pearse Keane, MD, MSc, Aaron Y. Lee, MD, MSc, Louis R. Pasquale, MD, Pietro Perona, PhD, Michael X. Repka, MD, MBA, Rishi Singh, MD, and Daniel Shu Wei Ting, MD, PhD.

    The Task Force's goals included the following:

    • Review state of the art in the field.
    • Identify unresolved issues & future challenges for the field of ophthalmology.
    • Provide guidance for ophthalmologists, industry, and policy makers about the field.
    • Produce a white paper summarizing the above

    State of the Art

    The Task Force reviewed the state of the art in the field, concurred that several AI technologies have been proven in terms of sensitivity and specificity for automated or assisted diagnosis and decision making, and identified seminal articles to provide a fundamental base of knowledge for the ophthalmologist. These are listed in the bibliography below. The Task Force also concluded that AI is here to stay, and despite challenges and barriers, AI would play a significant role in medical practice, and in particular, ophthalmology, and has the potential to enhance quality to care and access to care. To address this major development, the Academy Board of Trustees created a Committee made of subject matter experts to guide the Academy and its members through the related technical and policy aspects its use and dissemination in ophthalmology. This is similar to the role of the Committee for Medical Information Technology (CMIT) that helps to guide the Academy's response to federal health information technology activities, to coordinate different Committee activities, and to educate Academy members. The chair and members of the Committee are as follows: J. Peter Campbell, MD, MPH, Chair; Michael D. Abramoff, MD, PhD, Sally Baxter, MD, Peter Campbell, MD, MPH, Pearse Keane, MD, MSc, Jayashree Kalpathy-Cramer, PhD, Judy Kim, MD, Louis R. Pasquale, MD, Pietro Perona, PhD, Michael X. Repka, MD, MBA, Rishi Singh, MD, and Daniel Shu Wei Ting, MD, PhD.


    The Task Force also developed a series of papers to help ophthalmologists understand the real-world implications of AI for clinical practice, ranging from topics on medical education, privacy, ethics, real-world adoption and big data requirements. A synopsis of these articles is provided below.

    Protecting Data Privacy in the Age of AI-enabled Ophthalmology

    The advent of artificial intelligence and big data usher in important concerns about data privacy and protections. Principles to guide consideration of these issues include respect for persons, beneficence and justice; non-maleficence; and respect for autonomy. The difficulty with large datasets is that it may not be possible to remove all potentially identifiable information. Another issue is the exchange of data, and use of data by large corporations. On the other hand, if data is not ever able to be shared because of protections, then this could limit the benefits to patient care. New solutions for data privacy will need to be considered to balance the benefits and risks of AI.

    Big Data Requirements for Artificial Intelligence

    With the advent of standards for health information exchange and scalable methods such as cloud-based storage and computing architecture, big datasets in imaging and other data are able to be created in ophthalmology more rapidly and cheaper than ever before. This expansive framework permits artificial intelligence to be able to use and train on large datasets. To optimize big data in AI, several standards are required, including for data labels, sharing AI model architecture, accessible code and Application Programming interfaces.

    Current Challenges and Barriers to Real-World Artificial Intelligence Adoption for the Healthcare System, Provider, and Patient

    Currently, physicians encounter challenges associated with the real-world implementation of artificial intelligence. These include evaluation of different AI platforms and companies or consideration of in-house development of AI platforms, evolution of reimbursement models, the learning curve for physicians, response of patients and impact on the physician-patient relationship. All of these issues will require thoughtful consideration by different sectors of health care and the public to achieve high quality outcomes for patients.

    Emerging Ethical Considerations in the Use of Artificial Intelligence

    This article explores 3 unique considerations that arise with the use of AI, providing hypothetical examples that can be related to ophthalmic practice. The first consideration is the explainability of AI, which is limited and thus reduces the ability to perform a full root cause analysis of failure. The second consideration is the issue of responsibility and liability if AI is used for off-label diagnoses and a missed diagnosis occurs. A third consideration is the implications of scale of use, because AI could be used for broad public health screening. All these considerations and additional issues associated with the use of AI will require weighing benefits and risks, and defining what is acceptable or not.

    Bringing Ophthalmic Graduate Medical Education into the 20s with Information Technology

    At the current time, ophthalmology training programs do not incorporate Artificial Intelligence. However, AI is being incorporated into medicine and ophthalmologists should understand how to use AI in their daily clinical practice. Thus, training programs should develop a formal AI curriculum, covering topics such as big data and information technology, imaging, telemedicine, limitations, impact on the role of the physician, and integrating AI into practice, while preserving the physician-patient relationship.

    Reporting Guidelines for Artificial Intelligence in Medical Research

    There is an abundance of reports about artificial intelligence algorithms in clinical practice. These are difficult for practicing ophthalmologists to interpret because of their explosion in articles. Thus there is a need to standardize the reporting of these articles about AI to provide readers important information needed for interpretation and review. This article describes the recently published AI-extensions to the CONSORT and SPIRIT guidelines.

    Additional Activities

    The Task Force also coordinated with the Task Force on Telemedicine and Committee on Medical Information Technology on a statement, "Use of Artificial Intelligence in the Diagnosis of Diabetic Retinopathy and Other Conditions," that was also reviewed by the Board of Trustees in June 2020. The statement recommended: "The Academy recognizes the potential of autonomous diagnostic tests in increasing access to health care services, enhancing patient involvement in their health care decision making, improving efficiency, and reducing health care costs. The Academy recommends that ophthalmologists and policy makers evaluate these technologies as they would any other diagnostic tests, ensuring they are validated and have appropriate regulatory approval, before deployment."

    In summary, the Task Force on AI has outlined the importance of AI to ophthalmology and the relevance for ophthalmologists to understand the technical as well as nontechnical issues. This will be a critical tool in the ophthalmologist's armamentarium to provide optimal patient care. By creating a Committee on AI to expand upon the Task Force's initiative, the Academy will be assuming a critical leadership role in defining the educational, ethical, research, advocacy and public health issues for medicine and society as this tool is being adopted into daily clinical practice.

    Task Force on AI and CMIT Articles:

    1. Tom E, Keane PA, Blazes M, et al. Protecting data privacy in the age of AI-enabled ophthalmology. Transl Vis Sci 2020; 9:36.
    2. Wang SY, Pershing S, Lee AY et al. Big data requirements for artificial intelligence. Curr Open Ophthalmol 2020; 31:318-323.
    3. Singh RP, Hom GL, Abramoff MD et al. Current challenges and barriers to real-world artificial intelligence adoption for the healthcare system, provider and the patient. Transl Vis Sci Technol 2020; 9:45.
    4. Campbell JP, Lee AY, Abramoff M et al. Reporting guidelines for artificial intelligence in medical research. Ophthalmology 2020; 127:1596-1599.
    5. Cole E, Valikodath NG, Maa A et al. Bringing ophthalmic graduate medical education into the 2020s with information technology. Ophthalmology 2021; 128:349-35.
    6. Valikodath NG, Cole E, Ting DSW et al. Impact of artificial intelligence on medical education in ophthalmology. Transl Vis Sci Technol 2021; 10:14.

    Bibliography of Seminal Articles:

    1. Abramoff MD, Cunningham B, Patel B et al, on behalf of the Collaborative Community on Ophthalmic Imaging Executive Committee and Foundational Principles of Ophthalmic Imaging and Algorithmic Interpretation Working Group. Foundational Considerations for Artificial Intelligence Utilizing Ophthalmic Images. Ophthalmology. 2021:   
    2. Korot E, Wagner SK, Faes L et al. Will AI Replace Ophthalmologists? Transl Vis Sci Technol. 2020; 9: 
    3. Lee CS, Lee AY. How Artificial Intelligence Can Transform Randomized Clinical Trials. Transl Vis Sci Technol. 2020; 9: 
    4. Ting DSW, Lee AY, Wong TY. An Ophthalmologist's Guide to Deciphering Studies in Artificial Intelligence. Ophthalmology. 2019; 126:1474-1479.
    5. Coyner AS, Campbell JP, Chiang MF. Demystifying the Jargon: The Bridge Between Ophthalmology and Artificial Intelligence. Ophthalmology Retina. 2019; 3:291-3.
    6. Ting DSW, Peng L, Varadarajan AV et al. Deep Learning in Ophthalmology: The Technical and Clinical Considerations. Prog Retin Eye Res 2019; 
    7. Emanuel EJ, Wachter RM. Artificial Intelligence in Health Care Will the Clue Match the Hype? JAMA. 2019; 321:2281-2.
    8. Ting DSW, Pasquale LR, Peng L et al. Artificial Intelligence and Deep Learning in Ophthalmology. Br J Ophthalmol. 2019; 103:167-175.
    9. Naylor CD. On the Prospects for a (Deep) Learning Health Care System. JAMA. 2018; 320:1099-1110.
    10. Zhu L, Zheng WJ. Informatics, Data Science, and Artificial Intelligence. JAMA. 2018; 320:1103-4.
    11. Stead W.W. Clinical Implications and Challenges of Artificial Intelligence and Deep Learning. JAMA. 2018; 320:1107-1108.
    12. Carin L, Pencina MJ. On Deep Learning for Medical Image Analysis. JAMA. 2018; 320:1192-93.
    13. Baxter SL, Lee AY. Gaps in Standards for Integrating Artificial Intelligence Technologies into Ophthalmic Practice. Curr Opin Ophthalmol 2021; 32:431-438.