Machine learning-based prediction of a high-risk kidney function trajectory class after acute kidney injury
Chien-Liang Liu, You-Lin Tain, Chih-Chien Lin, Yueh-Lung Peng, Chien-Ning Hsu · 2026
Read Machine learning-based prediction of a high-risk kidney function trajectory class after acute kidney injury by Chien-Liang Liu, You-Lin Tain, Chih-Chien Lin, Yueh-Lung Peng, Chien-Ning Hsu (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13218091.txt.
Description
Chien-Liang Liu and co-authors report a 2026 study that trains machine learning models to flag patients whose kidney function is headed for steep decline after acute kidney injury. Many survivors of a sudden kidney injury recover unevenly, and some slide toward dialysis without clear warning signs. A tool that spots the dangerous pattern early could trigger protective care in time.
The cohort covered 88,632 adults with repeated kidney tests after discharge, drawn from more than 665,000 readings taken between 2010 and 2017. Modeling separated patients into three trajectories, and the group whose filtration rate stayed low faced the strongest link to later kidney replacement therapy. A gradient boosting model using only ten variables identified membership in that risky group with an accuracy score of 0.974. The slope of kidney function in the first three months mattered most. The authors note the data came from one health system, so outside testing is needed.
Catalog record
| CREATOR | Chien-Liang Liu, You-Lin Tain, Chih-Chien Lin, Yueh-Lung Peng, Chien-Ning Hsu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 31 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |