Discovering Hierarchy-Grounded Domains with Adaptive Granularity for Clinical Domain Generalization

TL;DR

This paper introduces UdonCare, a method leveraging medical ontologies to dynamically discover hierarchy-grounded patient domains, improving clinical prediction models' generalization across diverse patient groups.

Contribution

The paper proposes a novel hierarchy-pruning approach that utilizes medical ontologies to identify latent patient domains without domain labels, enhancing model robustness in healthcare.

Findings

UdonCare outperforms eight baselines on clinical prediction tasks.

Medical ontology-guided domain discovery improves model generalization.

Method effectively handles substantial domain gaps in healthcare data.

Abstract

Domain generalization has become a critical challenge in predictive healthcare, where different patient groups often exhibit shifting data distributions that degrade model performance. Still, regular domain generalization approaches often struggle in clinical settings due to (1) the absence of domain labels and (2) the lack of clinical insight integration. To address these challenges in healthcare, we aim to explore how medical ontologies can be used to discover dynamic yet hierarchy-grounded patient domains, a partitioning strategy that remains under-explored in prior work. Hence, we introduce UdonCare, a hierarchy-pruning method that iteratively divides patients into latent domains and retrieve domain-invariant (label) information from patient data. On two public datasets, UdonCare shows superiority over eight baselines across four representative clinical prediction tasks with substantial domain gaps, highlighting the potential of medical knowledge for enhancing model generalization.