Diabetic Blindness: New biomarkers may help detect early eye abnormalities
Diabetic retinopathy is the most common diabetic eye disease in the United States, and it is a leading cause of blindness.
NEW DELHI — According to a recent study done at the Indiana University School of Optometry, new biomarkers identified in the eyes might unlock the secret to helping control diabetic retinopathy and possibly even diabetes. The findings were published in the peer-reviewed journal ‘PLOS One.’
Diabetes can damage the eyes early before the alterations are evident with a routine clinical examination.
However, recent retinal research has established that these changes may be assessed earlier than previously assumed using sophisticated optical methods and computer analysis.
The capacity to detect biomarkers for this potentially blinding illness might lead to the early detection of those at risk for diabetes or vision loss and enhance clinicians’ ability to manage these patients.
Diabetic retinal damage can be detected early
Early detection of diabetic retinal damage is possible with painless methods. It may help identify undiagnosed patients early enough to mitigate the consequences of uncontrolled diabetes, according to Ann E. Elsner, a Distinguished Professor at the Indiana University School of Optometry and the study’s co-author.
Diabetic retinopathy is the most prevalent diabetic eye condition and a significant cause of blindness in people in the United States. Alterations cause it in the blood vessels in the retina.
The number of Americans with diabetic retinopathy is predicted to nearly double between 2010 and 2050, from 7.7 million to 14.6 million.
Artificial intelligence to detect diabetic retinopathy
The new study is part of a growing work focusing on using artificial intelligence to detect diabetic retinopathy in retinal scans. On the other hand, some of these algorithms detect changes based on characteristics that appear considerably later than the alterations shown in this study.
The IU-led strategy allows for early detection because of the retinal image processing methods disclosed in the paper.
In Elsner’s opinion, many algorithms utilize any visual information that differs between diabetic patients and controls to determine which individuals may have diabetes. However, he says, these can be vague.
He explains that their technique may be used in conjunction with other AI technologies to offer early information tailored to certain retinal layers or types of tissues.
The research on retinal image processing
Elsner and her co-author, Joel A. Papay, a PhD student in the IU School of Optometry’s Vision Science Program, worked on retinal image processing at their lab at the Borish Center for Ophthalmic Research.
They analyzed data from diabetic participants as well as healthy control subjects. A diabetic retinopathy screening of individuals of the underprivileged population at the University of California, Berkeley, and Alameda Health provided further data.
The computer analysis was done using retinal imaging data that is regularly gathered in well-equipped clinics. Yet, most of the information employed in this work is typically overlooked for patient diagnosis and care.
