Hybrid machine learning for real-time prediction of edema trajectory in large middle cerebral artery stroke
In treating malignant cerebral edema after a large middle cerebral artery stroke, clinicians need quantitative tools for real-time risk assessment. Existing predictive models typically estimate risk at one, early time point, failing to account for dynamic variables. To address this, we developed Hybrid Ensemble Learning Models for Edema Trajectory (HELMET) to predict midline shift severity, an established indicator of malignant edema, over 8-h and 24-h windows. The HELMET models were trained on retrospective data from 623 patients and validated on 63 patients from a different hospital system, achieving mean areas under the receiver operating characteristic curve of 96.6% and 92.5%, respectively. By integrating transformer-based large language models with supervised ensemble learning, HELMET demonstrates the value of combining clinician expertise with multimodal health records in assessing patient risk. Our approach provides a framework for accurate, real-time estimation of dynamic clinical targets using human-curated and algorithm-derived inputs.
Authors
- Ethan Phillips (ORCID: https://orcid.org/0000-0002-6662-6888)
- Stefanos Chatzidakis (ORCID: https://orcid.org/0000-0002-4652-397X)
- Stelios M. Smirnakis (ORCID: https://orcid.org/0000-0002-1929-2811)
- Odhran O’Donoghue (ORCID: https://orcid.org/0000-0001-6507-8990)
- Charlene Ong (ORCID: https://orcid.org/0000-0003-2377-3025)
- Agni Orfanoudaki (ORCID: https://orcid.org/0000-0003-3737-4826)
- Yumeng Zhang (ORCID: https://orcid.org/0009-0008-7711-2103)
- Jonathan J Song (ORCID: https://orcid.org/0000-0002-4349-2442)
- Benjamin Brush (ORCID: https://orcid.org/0009-0002-1918-2169)
- Leigh Ann Mallinger
- Yili Du (ORCID: https://orcid.org/0000-0001-8444-5477)
- Jack Pohlmann (ORCID: https://orcid.org/0009-0007-0896-4356)
- Ivy Kim
- Panos Tsimpos
Institutions
- Boston University (US)
- Boston Medical Center (US)
- United States Department of Veterans Affairs (US)
- Harvard University (US)
- North Carolina State University (US)
- University of Oxford (GB)
- New York University Langone Orthopedic Hospital (US)
- Massachusetts Institute of Technology (US)
Publication Details
- Journal
- npj Digital Medicine
- Published
- 2025-05-17
- DOI
- https://doi.org/10.1038/s41746-025-01687-y
- Citations
- 5
- Primary Topic
- Acute Ischemic Stroke Management
- Type
- article
- Field-Weighted Citation Impact
- 6.77
Funders
- University of Oxford
- National Institutes of Health
- National Institute of Neurological Disorders and Stroke