← brief for 2026-09-04 · Slowing decline

A metabolomics and machine learning-based predictive model for AKI-to-CKD progression risk

Key takeaways

This team built a computer model that uses body chemistry readings from blood or urine to guess who will go from a sudden kidney injury to long-term kidney disease. The model was trained on past patient records, so it shows a pattern, not proof of cause, and it has not been shown to change what happens to anyone. A tool like this would sort people into higher and lower risk after a kidney injury, so follow-up could be tighter. The record has no abstract, so how many patients it was built and tested on is not available here.

Who did this work

Zhiqian Xiong (Guiyang Medical University) · Ran Yan (Guizhou Provincial People's Hospital) · Yanzhe Peng (Guiyang Medical University) · Yu Yang (Guizhou Provincial People's Hospital) · Yuqi Yang (Guizhou Provincial People's Hospital) · Yan Zha (Guiyang Medical University)

Source

Paper · OpenAlex, CKD works · 2026-09-03
https://doi.org/10.1038/s41598-026-69680-3