ADVERTISEMENTs
Researchers develop AI tool to detect kidney failure
The tool predicts which patients with chronic kidney disease are likely to progress to end-stage renal disease.
-
Bhavana P - Wed, 10, Sep 2025
- 16:38 EDT
- 12
Rema Padman / Carnegie Mellon University
Carnegie Mellon University researchers, led by Indian-American scholar Rema Padman, have developed advanced artificial intelligence models that can help doctors predict kidney failure earlier and more accurately.
By integrating electronic health records with insurance claims data, the models predict which patients with chronic kidney disease (CKD) are likely to progress to end-stage renal disease (ESRD). It provides physicians stronger tools to detect risks earlier, plan treatment more effectively, and potentially reduce disparities in kidney care.
Also read: Scientists find genetic ‘switches’ driving hidden disease pathways
Padman, the trustees professor of management science and healthcare informatics at Carnegie Mellon’s Heinz College, emphasized the broader significance of the study. “Our study presents a robust framework for predicting ESRD outcomes, improving clinical decision-making through integrated multisourced data and advanced analytics. Future research will expand data integration and extend this framework to other chronic diseases,” she explained.
CKD is a long-term condition where kidney function steadily declines, sometimes leading to ESRD, when patients require dialysis or a kidney transplant to survive. It affects up to 16 percent of the global population, with 5 percent to 10 percent eventually reaching ESRD.
The Carnegie Mellon study analyzed data from more than 10,000 CKD patients between 2009 and 2018. Researchers tested a range of machine learning and deep learning models, ultimately finding that combining clinical and claims data produced the most accurate predictions. A 24-month observation period emerged as the optimal balance for early detection and prediction accuracy.
Co-author Yubo Li, a Ph.D. student at Heinz, noted the innovation, “By minimizing the observation window needed for accurate predictions, our approach balances clinical relevance with patient-centered practicality; this integration enhances both predictive accuracy and clinical utility, enabling more informed decision-making to improve patient outcomes.”
While researchers acknowledge limitations such as reliance on single-institution data, Padman's team believes the framework represents a major step toward more equitable and proactive kidney care.
Related
ADVERTISEMENT
ADVERTISEMENT
ADVERTISEMENT
E Paper
ADVERTISEMENT
.jpg)
Video
Comments
Start the conversation
Become a member of New India Abroad to start commenting.
Sign Up Now
Already have an account? Login