Domain Generalization and Adaptation in Intensive Care with Anchor Regression
RSS: Data Science and Artificial Intelligence · 2026
DOI: 10.1093/rssdat/udag001 ↗Kurz gefasst
Fachartikel in der Zeitschrift RSS: Data Science and Artificial Intelligence, 2026. Verfasst von Malte Londschien und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Themen: Intensivpflege
Abstract (Originalsprache)
The performance of predictive models in clinical settings often degrades when deployed in new hospitals due to distribution shifts. This paper presents a large-scale study of causality-inspired domain generalization on heterogeneous multi-center intensive care unit (ICU) data. We apply anchor regression and introduce anchor boosting, a novel, tree-based nonlinear extension, to a large dataset comprising 400,000 patients from nine distinct ICU databases. We find that anchor regularization yields improvements of out-of-distribution performance, particularly for the most dissimilar target domains. The methods appear robust to violations of theoretical assumptions, such as anchor exogeneity. Furthermore, we propose a novel conceptual framework to quantify the utility of large external data datasets. By evaluating performance as a function of available target-domain data, we identify three regimes: (i) a domain generalization regime, where only the external model should be used, (ii) a domain adaptation regime, where refitting the external model is optimal, and (iii) a data-rich regime, where external data provides no additional value.
Zitieren
APA
Londschien, M., Burger, M., Rätsch, G., & Bühlmann, P. (2026). Domain Generalization and Adaptation in Intensive Care with Anchor Regression. RSS: Data Science and Artificial Intelligence. https://doi.org/10.1093/rssdat/udag001
Vancouver
Londschien M, Burger M, Rätsch G, Bühlmann P. Domain Generalization and Adaptation in Intensive Care with Anchor Regression. RSS: Data Science and Artificial Intelligence. 2026. doi:10.1093/rssdat/udag001
BibTeX
@article{Londschien2026,
author = {Londschien, Malte and Burger, Manuel and Rätsch, Gunnar and Bühlmann, Peter},
title = {Domain Generalization and Adaptation in Intensive Care with Anchor Regression},
journal = {RSS: Data Science and Artificial Intelligence},
year = {2026},
doi = {10.1093/rssdat/udag001}
}
EUVIMED