Subscribe to Newsletter
Subspecialties Basic & Translational Research, Imaging & Diagnostics

Preparing for the Brave New World

Rishikesh Gandhewar (Credit: Headshot supplied by author)

“The AI doctor will see you now.” Well, perhaps not quite yet. Although AI is set to change the future of ophthalmology, further research is still required to navigate the various trials the technology presents and, in turn, mitigate the potential risks. To envisage the future of AI, we need to identify the barriers that still surround the subject – and find solutions.

Machine learning and deep learning are among the most powerful forms of AI in ophthalmology. Both technologies work by leveraging detailed, non-invasive ophthalmic imaging to develop discriminative AI that is able to classify, predict, and represent complex data. These technologies were pioneered through the Moorfields Eye Hospital and Google DeepMind collaboration, which created an AI system able to rival expert performance in retinal diagnoses derived from OCT scans (1). Such technologies offer an innovative solution to managing the immense ophthalmic disease burden, and offer hopes of streamlining what is currently the busiest outpatient department, allowing ophthalmologists to focus their attention on more complex cases.

Alongside this, insights gained through machine learning are valuable to novel drug discovery and to achieving personalized approaches for conditions such as glaucoma. We are also learning more about the potential to integrate ophthalmic data with screening for systemic diseases, such as Alzheimer’s disease and atherosclerosis (2). Such approaches have been of interest to tech titans like Spotify’s Daniel Ek, who is developing personalized, non-invasive body scanning (3).

Despite proof-of-concept, however, integrating these approaches into practice remains challenging for several reasons. For one, AI is not perfect – medical errors are inevitable, leading to potential complications. And the ethical and (more practical) legal frameworks around this approach are still formative. Consequently, AI remains assistive, with the final responsibility remaining with the clinician, thus limiting AI’s capabilities. Moreover, an algorithm is only as good as its data. Data that is skewed towards specific demographics will propagate bias and may not be universally applicable. This is particularly obstructive when considering the significant opportunity for integrating machine learning within developing regions, where preventable eye disease is rife (4).

But what is the true potential of AI? The various manifestations of AI hold the capability to change almost all facets of care. Natural language processing (NLP) is one such manifestation that has the potential to enhance medical consultations. Combining this with generative AI and online avatars could see the AI-doctor graduate and conduct entirely independent consultations. The obvious limitation here would be the lack of human interaction we deem so intrinsic to maintaining a patient-centered practice.

Robotics is another promising area, holding great potential for delivering precise microsurgery in ophthalmology by mitigating the risk of factors such as physiological tremors. But the cost and utility of robotics requires further optimization and risks creating further economic-derived inequality.

Finally, the field of computer vision, alongside augmented and virtual reality (AR and VR), holds a real opportunity to improve care and quality of life. This becomes more relevant as the metaverse nears its advent and headsets become more accessible. Current uses may include improved planning and training for surgeons to better visualize anatomy and simulate procedures.

There are a plethora of applications for AI in ophthalmology. It is clear that AI has the capacity and the capability to truly revolutionize the field of ophthalmology, ensuring better patient and doctor experiences. As we learn, build, and integrate AI’s many capabilities, an ethical and measured approach is necessary to traverse this “brave new world.” But I think I speak for many when I say I am excited to see where AI-driven technologies will take the field of ophthalmology in the near future.

Receive content, products, events as well as relevant industry updates from The Ophthalmologist and its sponsors.

When you click “Subscribe” we will email you a link, which you must click to verify the email address above and activate your subscription. If you do not receive this email, please contact us at [email protected].
If you wish to unsubscribe, you can update your preferences at any point.

  1. J D Fauw et al., “Clinically applicable deep learning for diagnosis and referral in retinal disease,” Nat Med, 24, 1342 (2018). PMID: 30104768.
  2. S K Wagner et al., “Insights into Systemic Disease through Retinal Imaging-Based Oculomics,” Transl Vis Sci Technol, [Online ahead of print] (2020). PMID: 32704412.
  3. Tech Crunch, “Neko, Daniel Ek’s next play, is another spin on preventative healthcare” (2023). Available at:
  4. W H Dean et al., “Ophthalmology training in sub-Saharan Africa: a scoping review,” Eye (Lond), 35, 1066 (2021). PMID: 33323984.
About the Author
Rishikesh Gandhewar

Rishikesh Gandhewar is an Academic Foundation Year 2 Doctor at Imperial College NHS Trust, Imperial College London.

Product Profiles

Access our product directory to see the latest products and services from our industry partners

Most Popular
Register to The Ophthalmologist

Register to access our FREE online portfolio, request the magazine in print and manage your preferences.

You will benefit from:
  • Unlimited access to ALL articles
  • News, interviews & opinions from leading industry experts
  • Receive print (and PDF) copies of The Ophthalmologist magazine



The Ophthalmologist website is intended solely for the eyes of healthcare professionals. Please confirm below: