For many people living with diabetes, vision problems are among the most feared complications of the disease.
Diabetes can damage the eyes slowly over many years, often without causing obvious symptoms at first. By the time vision changes become noticeable, significant and sometimes irreversible damage may already have occurred.
Researchers around the world have been searching for ways to detect eye damage much earlier. A new study suggests that the answer may be found in a simple blood sample combined with artificial intelligence.
The study, published in PLOS Medicine, was led by Huangdong Li and colleagues from the Guangdong Provincial Clinical Research Center for Ocular Diseases in China. The researchers developed a new AI-assisted model that uses information from dozens of blood proteins to estimate a person’s risk of developing diabetic retinal neurodegeneration.
Diabetic retinal neurodegeneration, often called DRN, occurs when nerve cells inside the retina gradually become damaged. The retina is a thin layer of tissue at the back of the eye that detects light and sends visual information to the brain. Healthy retinal nerve cells are essential for clear vision.
Scientists increasingly believe that DRN is more than just an eye problem. Research suggests it may reflect broader nerve damage caused by diabetes throughout the body. Some studies indicate that retinal degeneration may be linked to cognitive decline, dementia, and nerve damage affecting the hands and feet.
Because DRN develops silently in its early stages, doctors currently face challenges in identifying which patients are most at risk. This is where the new research could make an important difference.
The investigators analyzed blood samples from 1,492 people with type 2 diabetes who had not yet developed retinal neurodegeneration. They then followed participants over time and performed detailed retinal imaging examinations on 1,218 individuals during a six-year period.
To strengthen their findings, the researchers also tested their results using data from another group of 502 people with diabetes from the United Kingdom Biobank.
The team searched for biological markers in the blood that might signal future retinal damage. They identified 71 proteins that appeared to be strongly associated with the development of DRN.
Many of these proteins are involved in important biological processes, including inflammation, cellular repair, and maintenance of healthy tissues. These findings suggest that subtle biological changes may occur long before retinal damage becomes visible during routine eye examinations.
Using machine learning technology, the scientists combined information from the 71 proteins to create a prediction model called Pro-DRN. Artificial intelligence excels at recognizing complex relationships that would be difficult for humans to detect using traditional statistical methods.
When tested, the model performed remarkably well. It improved prediction accuracy by 26% compared with the strongest existing prediction approach. This means doctors may eventually be able to identify high-risk patients much earlier than before.
Earlier identification could lead to more frequent monitoring and potentially earlier treatment. While there are currently limited options specifically designed to prevent retinal neurodegeneration, researchers hope future therapies could be directed toward those identified as highest risk.
The scientists have already made the model available online for clinical evaluation. They believe the approach may help shift diabetic eye care from simply detecting damage to actively predicting who is most likely to develop it.
The broader implications are also intriguing. Since the retina is part of the nervous system, understanding retinal degeneration may provide clues about what is happening elsewhere in the body. Future research may reveal whether blood protein patterns associated with DRN can also predict other neurological complications of diabetes.
If you care about diabetes, please read studies about bananas and diabetes, and honey could help control blood sugar.
For more health information, please see recent studies about Vitamin D that may reduce dangerous complications in diabetes and plant-based protein foods may help reverse type 2 diabetes.
Source: Guangdong Provincial Clinical Research Center for Ocular Diseases.
