AI Transforming Retinal Diseases Diagnosis Via Eye Scans


by Hamna Bano

10 Best Activities



On 10th June 2023, at the annual meeting of the European Society of Human Genetics, a group leader from the UCL Institute of Ophthalmology and Moorfields Eye Hospital in London, UK, spoke about his team's creation of Eye2Gene, an AI system that can determine the genetic cause of Inherited Retinal Disease from retinal scans.

"Identifying the causative gene from a retinal scan is considered extremely challenging, even by experts. However, the AI is able to achieve this to a higher level of accuracy than most human experts” says the group leader Dr. Potikos.


To ease the process for them, the extensive database of data on IRDs maintained by Moorfields Hospital, which spans more than 30 years of research, was made available to the researchers. 

Moorfields is the biggest single-center dataset of individuals with both retinal and genetic data, having provided thorough retinal imaging and a genetic diagnosis to over 4000 cases.

In the future, “Eye2Gene could be easily incorporated into standard retinal examination, first as an assistant in specialist hospitals in order to get a second opinion, and eventually as a "synthetic expert" where such a person is not available. Ideally, Eye2Gene software would be embedded into the retinal imaging device" said Dr. Pontikos.

The system will need to pass regulatory approvals to show that it is safe and effective before it is used more widely. Future applications of artificial intelligence could make patient diagnosis and management and therapy more efficient, less invasive, and more broadly available. 

"We need further evaluation of Eye2Gene in order to assess its performance for different types of IRD patients from different ethnicities, different types of imaging devices, and in different types of settings, for example, primary vs secondary care. Clinical trials will be required before our system can be deployed in clinics as medical device software," added Dr. Pontikos.

Doctors are still surprised to see that even with specific Human Phenotype Ontology (HPO) terms being used, the Eye2Gene could do better than an HPO-specific approach.