Evaluating the generalisability of LiSep LSTM for early prediction of septic shock across US and european cohorts
Hong K Tan, Logan Froese, Daniel Wilhelms, Josef Fagerström, Michelle S Chew · 2026
Read Evaluating the generalisability of LiSep LSTM for early prediction of septic shock across US and european cohorts by Hong K Tan, Logan Froese, Daniel Wilhelms, Josef Fagerström, Michelle S Chew (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-13562518.txt.
Description
Hong Tan and co-authors carry a sepsis warning model trained on American hospital data across the Atlantic to test it on European patients, in a study from 2026. Sepsis kills quickly, and early warning buys the time that saves lives. A neural model called LiSep had forecast septic shock well at home, but models often fade when hospital routines and patient mixes change.
Tested on an Amsterdam intensive care database, the model kept strong accuracy and flagged shock a median of about two days ahead. It slightly beat its home performance there. The team also pooled both cohorts and ranked which input features carried the predictions. Checking a model abroad, feature by feature, is the kind of stress test clinical software needs before adoption.
Catalog record
| CREATOR | Hong K Tan, Logan Froese, Daniel Wilhelms, Josef Fagerström, Michelle S Chew |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 65 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |