Applying a transformer architecture to intraoperative temporal dynamics improves the prediction of postoperative delirium
Niklas Giesa, Maria Sekutowicz, Kerstin Rubarth, Claudia Doris Spies, Felix Balzer, Stefan Haufe, Sebastian Daniel Boie · 2024
Read Applying a transformer architecture to intraoperative temporal dynamics improves the prediction of postoperative delirium by Niklas Giesa, Maria Sekutowicz, Kerstin Rubarth, Claudia Doris Spies, Felix Balzer, Stefan Haufe, Sebastian Daniel Boie (2024) 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-11603037.txt.
Description
Niklas Giesa and co-authors show that a transformer model reading surgery data over time predicts post surgery delirium better than models using summary values, in work published in 2024. Delirium after surgery raises the risk of dementia and death. Earlier predictors mostly averaged each measure over the whole operation.
Models that read thirty minutes of live parameter streams performed best, with the transformer ahead of memory based rivals. Invasive systolic blood pressure and opioid doses carried the most weight, matching simple statistical checks. The authors conclude that the shape of signals during surgery, not just their averages, holds key warning signs.
Catalog record
| CREATOR | Niklas Giesa, Maria Sekutowicz, Kerstin Rubarth, Claudia Doris Spies, Felix Balzer, Stefan Haufe, Sebastian Daniel Boie |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 72 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |