Chronic kidney disease (CKD) is a progressive condition that can lead to end-stage renal disease (ESRD), when kidney function drops to critically low levels. Globally, CKD affects between 8% and 16% of people, with up to 10% progressing to ESRD. This not only threatens survival, requiring dialysis or transplantation, but also imposes high healthcare costs. New research now shows that artificial intelligence (AI) models can improve the prediction of CKD progression to ESRD.

In the study conducted at Carnegie Mellon University (Pittsburgh, PA, USA), researchers applied machine learning, deep learning, and explainable AI to integrate clinical and claims data. Unlike models using isolated datasets, this approach combined multiple sources to create a more accurate framework for outcome prediction. The research analyzed data from more than 10,000 CKD patients collected between 2009 and 2018.

Various statistical and AI models were tested across five observation windows, with a 24-month window providing the best balance between early detection and prediction accuracy. The findings, published in the Journal of the American Medical Informatics Association, showed that integrated models outperformed single-source models, and the 2021 estimated glomerular filtration rate equation reduced racial bias, particularly in African American patients.

By improving predictive accuracy, these AI-driven models could enhance CKD management and guide timely interventions to prevent disease progression. The framework also supports patient-centered care by enabling clinicians to tailor treatment strategies. While limited by data from a single institution and potential biases in electronic health records, the researchers plan to expand data integration and adapt the framework for other chronic diseases.

“Our study presents a robust framework for predicting ESRD outcomes, improving clinical decision-making through integrated multi-sourced data and advanced analytics,” said Rema Padman, professor of management science and healthcare informatics at Carnegie Mellon’s Heinz College. “Future research will expand data integration and extend this framework to other chronic diseases.”

http://www.allmycbs.com/en/index.asp