Posted in | News | Medical Robotics

Artificial Intelligence Used to Support Instant Diagnosis of Diabetic Eye Disease

Diabetic retinopathy is the main cause of vision loss in adults and its impact is growing around the world, with 191 million people estimated to be affected by 2030.

A fundus image of a retina, with damaged areas highlighted by the image-processing algorithm. (Image credit: RMIT University)

There are no early-stage symptoms and the disease may already be progressive by the time people begin losing their sight. Early diagnosis and treatment can make a significant difference in how much vision a patient retains.

Now a team of Australian-Brazilian scientists led by RMIT University have created an image-processing algorithm that can automatically spot one of the important signs of the disease, fluid on the retina, with an accuracy rate of 98%.

Chief investigator Professor Dinesh Kant Kumar, RMIT, said the technique was instantaneous and economical.

We know that only half of those with diabetes have regular eye exams and one-third have never been checked. But the gold standard methods of diagnosing diabetic retinopathy are invasive or expensive, and often unavailable in remote or developing parts of the world. Our AI-driven approach delivers results that are just as accurate as clinical scans but relies on retinal images that can be generated with ordinary optometry equipment. Making it quicker and cheaper to detect this incurable disease could be life-changing for the millions of people who are currently undiagnosed and risk losing their sight.

Dinesh Kant Kumar, Professor and Chief investigator, RMIT

Fluorescein angiography and optical coherence tomography scans are presently the most accurate clinical approaches for diagnosing diabetic retinopathy.

An alternative and cheaper technique is examining images of the retina that can be taken with comparatively inexpensive equipment called fundus cameras, but the process is time-consuming, manual, and less reliable.

To automate the examination of fundus images, scientists in the Biosignals Laboratory in the School of Engineering at RMIT, along with collaborators in Brazil, applied deep learning and artificial intelligence methods.

The algorithm they created can accurately and reliably locate the presence of fluid from damaged blood vessels, or exudate, inside the retina.

The scientists hope their technique could ultimately be used for extensive screening of at-risk populations.

Undiagnosed diabetes is a massive health problem here and around the globe. For every single person in Australia who knows they have diabetes, another is living with diabetes but isn’t diagnosed. In developing countries, the ratio is one diagnosed to four undiagnosed. This results in millions of people developing preventable and treatable complications from diabetes-related diseases. With further development, our technology has the potential to reduce that burden.

Dinesh Kant Kumar, Professor and Chief investigator, RMIT

The scientists are in discussions with producers of fundus cameras about potential partnerships to advance the technology.

The study—with lead author Parham Khojasteh and collaborators from Federal University of Sao Carlos, Federal Institute of Sao Paolo, University of Campinas and Sao Paolo State University—is published in Computers in Biology and Medicine (January 2019, Volume 104, DOI:10.1016/j.compbiomed.2018.10.031).

Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback
Your comment type

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.