Time-updated explainable machine learning predicts short-term mortality in peritoneal dialysis patients
Quan Wang, Yanqiong Ding, Qing Luo, Sheng Wan, Yanmin Zhang · 2026
Read Time-updated explainable machine learning predicts short-term mortality in peritoneal dialysis patients by Quan Wang, Yanqiong Ding, Qing Luo, Sheng Wan, Yanmin Zhang (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-13220578.txt.
Description
Quan Wang and colleagues build an explainable machine learning system for death risk in home dialysis patients in a 2026 study. People on peritoneal dialysis carry a high risk of dying, from cardiovascular causes, and their measurements change month to month. A warning system that updates with each result could flag the six-month danger window while action is still possible.
Working retrospectively with 1,484 patients from one dialysis centre, the team trained eight machine learning models on demographics, features, and laboratory values refreshed over time. The winners were validated internally and again on a later, separate group of patients. A gradient boosting model performed best, with accuracy of 0.888 internally and 0.850 in the temporal check. An explanation method, SHAP, opened the model up, naming the features that drove its warnings so clinicians can sanity-check them. The authors present the system as decision support for personalized treatment, not a black-box verdict.
Catalog record
| CREATOR | Quan Wang, Yanqiong Ding, Qing Luo, Sheng Wan, Yanmin Zhang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 35 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |