A groundbreaking study from the University of Dundee has shown that artificial intelligence (AI) can analyze routine diabetes eye screening images to detect early signs of kidney disease. This new approach, unveiled at the Diabetes UK Professional Conference 2025, has the potential to predict the onset of chronic kidney disease (CKD) years before symptoms emerge, allowing for earlier intervention and treatment.
How Diabetes Affects Kidney Health
Type 2 diabetes occurs when the insulin produced by the pancreas is insufficient or ineffective, leading to dangerously high blood sugar levels. Over time, this can cause severe damage to organs, increasing the risk of heart disease, stroke, vision loss, and kidney failure.
Diabetes-related kidney disease often progresses silently, remaining undetected until it becomes severe. One in five people with diabetes will require kidney disease treatment in their lifetime. Additionally, nearly one in three patients needing dialysis or a kidney transplant have diabetes, making it one of the leading causes of end-stage kidney disease.
AI’s Role in Early Detection
People with diabetes in the UK aged 12 and older are routinely invited for eye screenings. These screenings capture images of the retina to identify damage caused by diabetes. Researchers from the Universities of Dundee and Glasgow investigated whether AI could analyze these images to detect hidden patterns indicating kidney disease.
Led by Dr. Alexander Doney, the research team developed an AI tool trained on nearly one million eye screening photographs from approximately 100,000 individuals with type 2 diabetes in Scotland. The images were cross-referenced with kidney health data, allowing the AI to distinguish between individuals with and without kidney disease. The tool was then validated using data from an additional 30,000 individuals.
High Accuracy and Improved Predictions
The AI tool demonstrated an 86% accuracy rate in detecting existing kidney disease. For individuals without kidney disease, it successfully predicted who would develop it within five years with 78% accuracy. Notably, the AI outperformed traditional kidney function tests by identifying future kidney disease risks even when standard tests showed no warning signs.
Dr. Elizabeth Robertson, Director of Research at Diabetes UK, highlighted the significance of this discovery:
“Kidney damage often progresses silently until it becomes severe, and early detection is critical. This fascinating research has offered a new window into kidney health – through the eyes. By revealing invisible patterns in images taken during eye screenings, this AI tool could alert healthcare professionals to early signs of kidney damage. This would offer a vital opportunity to provide tailored support to slow or halt the progression of kidney disease, ultimately saving lives.”
Transforming Diabetic Eye Screenings
Dr. Alex Doney, the study’s lead researcher, emphasized the potential impact of AI in medical diagnostics:
“The retina at the back of the eye is the only place where the fragile network of blood vessels, critical to the health of all organs throughout the body, can be conveniently visualized and photographed. AI can be trained to detect very early features and patterns in these images that humans may miss. These indicators provide an earlier opportunity for doctors to act before permanent kidney damage occurs.”
This study was funded by AstraZeneca, Innovate UK, and other organizations supporting medical research.
Future Implications and Advancements
The ability to detect kidney disease using AI analysis of eye screenings could revolutionize diabetes care. If widely adopted, this approach could transform routine screenings into a multi-purpose diagnostic tool, helping doctors predict and prevent various diabetes-related complications beyond kidney disease.
By leveraging AI-driven healthcare innovations, medical professionals can improve early intervention strategies, enhance patient outcomes, and reduce the burden of diabetes-related complications worldwide.
For more updates on groundbreaking medical research, visit Daljoog News.