Answers are grounded in this paper and cite their sources.
The Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) 2009 equation is the most commonly used equation for estimating glomerular filtration rate (GFR) in clinical practice. However, recent studies have questioned the accuracy of this equation in estimating GFR in people with diabetes. We conducted a comprehensive review of existing literature on the role of artificial intelligence and machine learning in estimating GFR and the progression of chronic kidney disease (CKD), specifically in people with diabetes. Artificial intelligence, including machine learning and image-based deep learning algorithms have shown promise in improving the accuracy of GFR estimation. Artificial Neural Networks is the commonly used machine learning algorithm in GFR estimation studies. Other artificial intelligence methods include random forests, support vector machines, and ensemble learning models. Many of the studies included in this review reported that artificial intelligence-based GFR estimation equations exhibit lower bias, as well as higher precision and accuracy. However, these findings are not consistent across all the studies. In addition, currently available studies are limited to smaller sample sizes and majority of the studies are from selected countries or populations. Therefore, before implementing these methods in clinical practice, it is essential to validate them on larger sample sizes and diverse patient populations.
No takes yet. Share an insight, caveat, or question.
Koye et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: