Comparison of Multimodal Deep Learning Approaches for Predicting Clinical Deterioration in Ward Patients: Observational Cohort Study
A multimodal model combining structured data with embeddings using SapBERT had the highest area under the precision-recall curve, but performance was similar between models with and without CUIs. Although the addition of CUIs from notes to structured data did not meaningfully improve model performance for predicting clinical deterioration, models using CUIs could provide clinicians with relevant information and additional clinical context for supporting decision-making.
Authors
- Anoop Mayampurath (ORCID: https://orcid.org/0000-0002-3010-6960)
- Dana P. Edelson
- Dmitriy Dligach (ORCID: https://orcid.org/0000-0002-2585-2707)
- Majid Afshar (ORCID: https://orcid.org/0000-0002-6368-4652)
- Kyle A. Carey (ORCID: https://orcid.org/0000-0002-5038-6074)
- Jennie Martin
- Matthew M. Churpek (ORCID: https://orcid.org/0000-0002-4030-5250)
- Charles A Kotula
Institutions
- University of Wisconsin–Madison (US)
- Loyola University Chicago (US)
- University of Chicago (US)
- Madison Group (United States) (US)
Publication Details
- Journal
- Journal of Medical Internet Research
- Published
- 2025-06-02
- DOI
- https://doi.org/10.2196/75340
- Citations
- 6
- Primary Topic
- Machine Learning in Healthcare
- Type
- article
- Field-Weighted Citation Impact
- 14.89