Multimodal retinal imaging technology specialist, Optos, shares insights into how AI tools are transforming practice

AI is set to prove vital against a revolving backdrop of ageing patients, economic challenges, increased eye disorders, and systemic disease.

Professor Paulo E Stanga, a consultant ophthalmologist, retinal surgeon, director of the Retina Clinic London and professor of ophthalmology for the Institute of Ophthalmology at University College London, specialising in medical and surgical retinal imaging and research, welcomes the introduction of AI.

“I believe that AI will allow for earlier diagnosis and is going to be essential for telemedicine and for those practices without the benefit of an in-house retina specialist,” he said. “It is going to be very beneficial, perhaps, for optometrists making a referral.”

Optos has been at the forefront of the AI revolution, in line with its commitment to advancing retinal imaging and helping eye care professionals deliver the best possible patient care.

Evidence shows AI has the potential to improve eye care solutions and assist eye care professionals through improved visualisation, measurement, and monitoring.

As an early adopter of AI, the company developed OptosAI, which is a tool used for screening diabetic retinopathy (DR).

The new tool combines optomap® ultra-widefield (UWF^TM) imaging, which captures a 200° single-capture digital image of the retina with an AI solution for DR. OptosAI is a tool for AI-based DR management and diabetic macular edema (DME) screening and is intended to be used for the rapid AI-based DR screening with UWF images.

Diabetes can affect the eye in many ways, and every year more than 1700 people have their sight affected by diabetes in the UK – that’s more than 30 people every week.1 People typically have no symptoms in the early stages of DME, which is why routine examinations combined with innovative technology will prove vital for capturing early indicators of disease.

The device works by evaluating an optomap image for signs of diabetic retinopathy and, based on the lesions that are detected, a report is generated to advise whether the subject has referrable retinopathy. It has been proven to have 96% sensitivity and 93%² specificity. There will be a case for referral for above mild and moderate and for severe and urgent referral recommendations.

The speed and accuracy of OptosAI is particularly advantageous when dealing with an elderly patient cohort that may be living with DR for long periods of time. It's important that the right intervention is made in a timely manner so the advances in treatment can have the most impact on both the patient and the healthcare economics.

So, what do practitioners think of OptosAI?

Professor Tunde Peto, professor of clinical ophthalmology at Queen’s University Belfast, said: “OptosAI for DR has the potential to save sight and save healthcare practitioner’s time, whilst allowing clinicians to deliver the best clinical care.”

Despite impressive advancements, further investment and development is required to create a unified approach across industry, regulatory bodies, and end users. This will be done through open and fluent conversation.

Derek Swan, Optos chief technology officer, said: "There is huge opportunity for the adoption of AI across both optometry and ophthalmology. Industry and clinical progress to date has driven significant growth in available health data, enabling delivery of AI services which can help practitioners make informed decisions in a very tightly constrained healthcare sector.

“AI has the potential to be an additional and assistive part of a practitioner’s tool set, to complement the patient exam and the data output to improve practice efficiency, referral accuracy, better detection and diagnosis of disease.”

See how multimodal ultra-widefield technology combined with AI can help you improve the way you manage your patients. For more information call +44 (0)1383 843350/0808 100 4546 (Freephone UK Only) or email [email protected].

1. Diabetes UK: https://www.diabetes.org.uk/guide-to-diabetes/managing-your-diabetes/diabetic-eye-screening
2. Performance of a Diabetic Retinopathy Deep Learning Model for Ultra-widefield Imaging. IOVS. 2022.
3. Evaluation of a new model of care for people with complications of diabetic retinopathy: The EMERALD Study. Ophthalmology. 2020
4. Comparison of ETDRS Standard 7-field Imaging versus Ultrawide Field Imaging for Determining Diabetic Retinopathy Severity. JAMA Ophthalmology. 2018.
5. Nonmydriatic Ultrawide Field Retinal Imaging Compared with Dilated Standard 7-Field 35mm Photography and Retinal Specialist Examination for Evaluation of Diabetic Retinopathy. American Journal of Ophthalmology. 2012.