Validation of natural language processing for surgical complication surveillance: detecting 11 postoperative complications from electronic health records
Emilie Even Dencker, Alexander Bonde, Anders Troelsen, Martin Sillesen · 2026
Read Validation of natural language processing for surgical complication surveillance: detecting 11 postoperative complications from electronic health records by Emilie Even Dencker, Alexander Bonde, Anders Troelsen, Martin Sillesen (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-13537478.txt.
Description
Emilie Even Dencker and colleagues tested whether language software can spot surgical complications in health records. Published in 2026, the look-back Danish study covered 17,486 cases across 18 hospitals over six years. Problems after surgery guide quality checks, yet cheap, exact tracking methods were lacking.
The models scored from just over 0.9 to near perfect on a standard accuracy curve and beat diagnosis-code tracking by a wide margin. Sensitivity ran from seven tenths to full marks except for heart attacks at one half, while true-hit rates varied widely. Pairing software with human review of just over sixteen percent of cases reached a perfect true-hit rate. The authors present the software as a support for, not a swap for, manual checks, offering wider tracking.
Catalog record
| CREATOR | Emilie Even Dencker, Alexander Bonde, Anders Troelsen, Martin Sillesen |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 39 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |