Machine Learning–Based Prediction of Severe Outcomes in Acute Propranolol Poisoning: An Interpretable Model With Internal Validation
Acute propranolol poisoning can cause severe cardiovascular and neurological toxicity, but early risk stratification remains challenging. This is a code base for a machine learning study on a retrospective single-centre cohort of 720 patients. We developed and internally validated machine-learning models for a composite severe in-hospital outcome (death, intensive care unit admission, intubation, vasopressor or inotrope support, coma, seizure or respiratory arrest). At the time of this description, the study is at the peer-review step (not published on the time of this entry).
Authors
- Kamran Rezaei (ORCID: https://orcid.org/0000-0002-0451-3175)
- Akram Aboutorabi (ORCID: https://orcid.org/0000-0003-0958-4017)
- Babak Mostafazadeh (ORCID: https://orcid.org/0000-0003-4872-9610)
- Shahin Shadnia (ORCID: https://orcid.org/0000-0002-9401-0781)
- Mitra Rahimi (ORCID: https://orcid.org/0000-0002-8785-8207)
- Parnian Eslahi (ORCID: https://orcid.org/0000-0002-6075-749X)
- Sayed Masoud Hosseini (ORCID: https://orcid.org/0000-0002-2046-1272)
- Peyman Erfan Talab Evini (ORCID: https://orcid.org/0000-0002-0995-4973)
- Negin Rajabi
Institutions
- Shahid Beheshti University (IR)
- Shahid Beheshti University of Medical Sciences (IR)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-03
- DOI
- https://doi.org/10.5281/zenodo.23913784
- Primary Topic
- Poisoning and overdose treatments
- Type
- article
- Field-Weighted Citation Impact
- 0.00