High AUROC can mask decision failure in sepsis transcriptomic classifiers: Preprocessing stability outweighs post hoc calibration across cohorts
Hongwei Zheng, Wenbiao Chen · 2026
Read High AUROC can mask decision failure in sepsis transcriptomic classifiers: Preprocessing stability outweighs post hoc calibration across cohorts by Hongwei Zheng, Wenbiao Chen (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-13541173.txt.
Description
Hongwei Zheng and Wenbiao Chen warn in a 2026 paper that a top score can hide failure in sepsis prediction tools. Such tools read gene activity in blood to flag sepsis, and a common score says how well they rank sick above well. But a tool can rank well and still fail at the fixed cutoff doctors must use to act.
The team tested four data sets and several ways of scaling the data before training. Inside one data set all ways looked strong, yet on new data the standard ways collapsed to chance at the fixed cutoff. A method that ranks values within each sample held steady across cohorts. Extra tuning after the fact did not rescue the failing ways. The authors urge checks that match real clinical use.
Catalog record
| CREATOR | Hongwei Zheng, Wenbiao Chen |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 51 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |