Research Blog 12 June 2023

5 ways artificial intelligence is supporting sight loss research

The spotlight is on AI's potential and pitfalls, with PM Rishi Sunak announcing that the UK will host a global safety summit. We explore the potential of AI in sight loss research and to support people with visual impairments.

The Guardian reports that the spotlight is on AI's potential and pitfalls, with PM Rishi Sunak announcing that the UK will host a global safety summit. The government believes that the UK is ideally placed to be at the forefront of AI technology.

The UK indeed has a long history of technological innovation. The area surrounding Cambridge earned the nickname Silicon Fen and is home to over 5,000 companies specialising in biotech, AI and fintech. UK technology companies such as Sinclair Research and Acorn Computers were founded in Cambridge. Across the UK, companies are at the cutting edge of technology.

Eye research is no exception. Researchers focussing on sight loss conditions are also tapping into AI to advance scientific understanding and deliver treatments. So what projects are currently underway, and what is the potential of AI in saving sight? We highlight five projects.

1. Using AI to improve diagnosis and treatment of Birdshot Uveitis

Artificial intelligence could enable earlier diagnosis and better monitoring of people with birdshot uveitis.

Fight for Sight is funding scientists from universities and hospitals in Birmingham and London who are using AI to extract ‘hidden’ information from the retinal scans of people with the rare condition.

The team aims to use the power of AI to help extract information that may be ‘hidden away’ from human examiners. Their goal is to create an automated system that can detect and interpret subtle patterns in the data, providing accurate measures to improve diagnosis and monitoring of the disease.

Birdshot uveitis is a rare, chronic (long-term) inflammatory eye disease that can lead to vision loss or blindness. Also known as birdshot chorioretinopathy, it takes its name from the painless, light-coloured spots that develop on the retina due to the condition. These spots are scattered in a birdshot pattern (the pattern from a shotgun barrel).

Developing accurate new tests that can diagnose the condition earlier could enable people to receive treatment sooner, helping to prevent unnecessary sight loss.

2. AI could save people from blindness

Researchers in the UK and Germany have used AI to develop a system called Eye2Gene, an AI system capable of identifying the genetic cause of inherited retinal diseases (IRDs) from retinal scans.

IRDs – single gene disorders affecting the retina – are challenging to diagnose as they can involve changes in one or many candidate genes. AI could provide more widespread, efficient testing.

The annual conference of the European Society of Human Genetics (June) saw Dr Nikolas Pontikos at the UCL Institute of Ophthalmology and Moorfields Eye Hospital, London, talk about Eye2Gene.

Eye2Gene is an AI system which can identify the genetic causes of IRDs from retinal scans. In the future, Eye2Gene could be easily incorporated into a standard retinal examination.

Donate now

3. AI for diabetic retinal screening

Diabetes is one of the leading causes of eye problems in the UK, with over 1,700 people living with diabetes facing serious difficulties with their sight every year. The World Health Organisation recommends screening for diabetic retinopathy. Early intervention can allow for sight-saving interventions.

A University of Liverpool spin-out company uses an artificial intelligence (AI) system to improve diabetic eye screening. Diabetes causes damage to small blood vessels in organs throughout the body, including the eye (diabetic retinopathy).

AI Sight Ltd will use an AI system conceived by academics at the University’s Department of Eye and Vision Science. The system identifies borderline and contentious cases and passes them on for human assessment locally or remotely, integrated within the system.

Fight for Sight is partnering with Moorfields Diabetes UK, the Macular Society and Moorfields Eye Charity to co-fund new research to reduce the risk of sight loss in people with diabetes.

4. AI Technology to improve patient care

Researchers at Moorfields Eye Hospital NHS Foundation Trust and the UCL Institute of Ophthalmology are working together to explore whether AI can improve clinical care for eye patients. In 2019, they announced a breakthrough because of their ongoing work with DeepMind Health.

Machine learning technology has been successfully trained on thousands of historic de-personalised eye scans to identify signs of eye disease and recommend how patients should be referred for care.

The research, published in the journal Nature, reported that the AI system could recommend the correct referral for over 50 eye diseases with 94% accuracy, which is a match for world-leading eye experts. The hope is that the technology could revolutionise how professionals carry out eye tests, allowing them to spot conditions earlier.

Speaking about the breakthrough Professor Sir Peng Tee Khaw, director of the NIHR Biomedical Research Centre at Moorfields Eye Hospital NHS Foundation Trust and UCL Institute of Ophthalmology, said: “The results of this pioneering research with DeepMind are very exciting and demonstrate the potential sight-saving impact AI could have for patients. I am in no doubt that AI has a vital role to play in the future of healthcare.”

5. AI app helps visually-impaired people apply make-up

In January 2023, Estée Lauder released an app that uses artificial intelligence to make it easier for people with visual impairments to apply their make-up more easily and explore creatively through brushes and colour.

The voice-enabled make-up artist (VMA) – Apple only – was co-created with visually-impaired people. It uses intelligent mirror technology developed with machine learning and works with the phone’s front-facing camera alongside voice instruction technology.

Investing in the future

It is important to highlight the speed at which these systems are being developed, and the need to balance this with testing and review.

While the government recognises a need to invest in AI, more needs to be invested in sight loss research. As our Time to Focus report revealed, in 2018, just over 1.5 per cent (£24 million) of the £1.4 billion that UK Research and Innovation, government and other public bodies invested in medical research was for eye research. That’s a total of just £9.60 for each person affected by sight loss in the UK.