Continuous predictive mortality risk monitoring after allogeneic hematopoietic stem cell transplantation
Nick Rucks, Sergej Korlakov, Sebastian Alexander Scharf, Anna Rommerskirchen, Rainer Haas, Stefan Conrad · 2026
Read Continuous predictive mortality risk monitoring after allogeneic hematopoietic stem cell transplantation by Nick Rucks, Sergej Korlakov, Sebastian Alexander Scharf, Anna Rommerskirchen, Rainer Haas, Stefan Conrad (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/PMC13280382.txt.
Description
Nick Rucks and colleagues present a 2026 proof of concept framework for live death risk monitoring after stem cell transplant. Allogeneic hematopoietic stem cell transplantation, which transplants donor blood stem cells, treats high risk blood cancers. Early mortality stays significant, with about six to eleven percent dead in the first one hundred days. Static pre transplant scores cannot track risk as care unfolds. Their framework updates risk in real time from routine clinical data.
Deep learning models forecast seven day mortality risk from twenty two lab values of the past fourteen days, plus five demographic features. An explainability method shows which inputs drive each forecast over time. The approach was evaluated on eight hundred ninety one patients in Dusseldorf from 2004 to 2019. An independent cohort from the MIMIC-IV database gave a second test. Live, explained risk could help teams act before decline sets in.
Catalog record
| CREATOR | Nick Rucks, Sergej Korlakov, Sebastian Alexander Scharf, Anna Rommerskirchen, Rainer Haas, Stefan Conrad |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 101058 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |