AI Tool Predicts Chronic Kidney Disease Risk in Diabetes

Many risk models were built on Western cohorts and may not translate well to Asian populations, limiting timely prevention. Clinicians need accurate, locally validated tools to guide early intervention and follow-up. To help address this challenge, researchers have developed an artificial intelligence agent and risk model tailored for precision diabetes management in Hong Kong.

The Hong Kong Polytechnic University (PolyU) has introduced the AI Agent for Precision Diabetes Management—PIPE-AI—together with a companion disease risk prediction model designed for Asian populations. The system targets patients with type 2 diabetes by estimating the likelihood that complications will worsen over time. It aims to support earlier clinical decisions and coordinated care planning across settings.

Model development used 17 years of electronic health records from the Hospital Authority Data Collaboration Laboratory, covering more than 560,000 patients with diabetes. The model estimates 10‑year risk for complications such as chronic kidney disease and achieved an accuracy of 87.1% in development. The findings were published in npj Digital Medicine in 2026.

PIPE-AI functions as a clinical interface that translates complex medical information into language patients can understand and that facilitates clinician–patient communication. Operationally, it is positioned to support preliminary screening and risk stratification in family medicine and primary care. It can inform specialist referrals for high‑risk or complex cases, extend 24‑hour consultation support through district health centers, and assist patients with self‑management, including diet, exercise, medication adherence, and health indicator tracking.

Clinical safeguards include a nurse oversight mechanism that automatically alerts a registered nurse when abnormal risk levels or important health warnings are detected. A community clinical study is underway with the Department of Family Medicine and Primary Healthcare of the Hospital Authority’s New Territories West Cluster and the Yuen Long District Health Center to recruit individuals with prediabetes and type 2 diabetes. Participants will receive individualized risk assessments and health management recommendations.

“Diabetes management is not only about treating a single disease, but is also closely related to the long-term allocation of health care resources and public health strategy," said David Shum, Dean of the Faculty of Health and Social Sciences, Yeung Tsang Wing Yee and Tsang Wing Hing Professor in Neuropsychology, Chair Professor of Neuropsychology at PolyU.

“By translating advanced AI technology into a tool for clinical application, the PolyU research team has extended risk prediction capabilities, which were previously largely confined to hospitals, to primary health care and community services. This will help allocate health care resources more precisely and drive a shift in the health care model from ‘passive treatment’ to ‘proactive prevention,’ which in the long run is expected to alleviate the health care burden posed by chronic diseases in Hong Kong.”

Related Links
PolyU