Explainable AI for infection prevention and control: modeling CPE acquisition and patient outcomes in an Irish hospital with transformers
This study presents a robust and explainable AI framework for analyzing complex EMR data to identify key risk factors and predict CPE-related outcomes. Our findings underscore the superior performance of the Transformer models and highlight the importance of diverse clinical and network features. The transparent interpretability offered by our XAI approach provides actionable insights for infection prevention and control, paving the way for more targeted interventions and ultimately enhancing patient safety within acute healthcare settings.
Authors
- Nick McDonald (ORCID: https://orcid.org/0000-0002-1417-1148)
- Marie Ward (ORCID: https://orcid.org/0000-0002-6638-8461)
- Úna Geary (ORCID: https://orcid.org/0000-0001-7052-5863)
- Martin Crane (ORCID: https://orcid.org/0000-0001-7598-3126)
- Marija Bezbradica (ORCID: https://orcid.org/0000-0001-9366-5113)
- Minh-Khoi Pham (ORCID: https://orcid.org/0000-0003-3211-9076)
- Brian O’Connell (ORCID: https://orcid.org/0000-0003-4529-7664)
- Declan Byrne (ORCID: https://orcid.org/0000-0002-1836-4043)
- Colm J Bergin (ORCID: https://orcid.org/0000-0002-6651-1132)
- Tai Tan (ORCID: https://orcid.org/0000-0001-6657-0872)
- Rob Brennan (ORCID: https://orcid.org/0000-0001-8236-362X)
- Donncha Creagh
Institutions
- University College Dublin (IE)
- Trinity College Dublin (IE)
- Health Service Executive (IE)
- Fraunhofer Research Institution for Additive Manufacturing Technologies IAPT (DE)
- St. James's Hospital (IE)
- Dublin City University (IE)
Publication Details
- Journal
- BMC Medical Informatics and Decision Making
- Published
- 2025-10-22
- DOI
- https://doi.org/10.1186/s12911-025-03214-1
- Primary Topic
- Machine Learning in Healthcare
- Type
- article
- Field-Weighted Citation Impact
- 0.00