Hong Kong researchers develop algorithm to predict diabetic kidney disease risk

A team of researchers from the Chinese University of Hong Kong and California-based medical research institute Sanford Burnham Prebys has come up with a new computational model to predict whether a patient with Type 2 diabetes will develop kidney disease.

The said model, according to a press statement, measures markers of DNA methylation – a process that occurs when subtle changes accumulate in the DNA – from blood samples to check both present kidney function and how the kidneys will function in the future. 

It was developed using data from more than 1,200 patients with T2D in the Hong Kong Diabetes Register and has been tested even on a group of Native Americans with T2D.

WHY IT MATTERS

It is said that about half of cases of end-stage kidney disease and dialysis around Asia are due to diabetes. While there has been significant progress in treating diabetic kidney disease, it remains difficult to assess a patient’s risk of developing the condition using clinical factors alone, said Ronald Ma, professor of Diabetes at CUHK Faculty of Medicine. 

By combining AI and data, the research team was able to come up with a computational model that has the potential to prevent kidney disease among T2D patients and optimise their treatment. 

They are now working to refine the model while also expanding its application by incorporating other data that can enhance its ability to predict other diabetes-related outcomes. A patent application for the same model has also been filed. 

MARKET SNAPSHOT

There is already an existing AI-powered tool that also assesses a T2D patient’s risk of getting kidney disease. Developed by New York-based company Renalytix AI, KidneyIntelX generates a kidney disease risk score based on a patient’s blood sample and medical history.