Bridging Data Gaps of Rare Conditions in ICU: A Multi-Disease Adaptation Approach for Clinical Prediction

2025-07-08

AI Summary

Clinical prediction for rare diseases in intensive care units (ICUs) suffers from poor performance due to data scarcity and high intra-disease heterogeneity. Method: We propose KnowRare, a domain-adaptive deep learning framework that jointly leverages self-supervised pretraining and a conditional knowledge graph (CKG) to learn disease-agnostic representations and enable targeted knowledge transfer across clinically similar conditions. The model integrates graph neural networks with a multi-task domain adaptation mechanism to jointly mitigate data sparsity and explicitly model heterogeneity. Results: Evaluated on two real-world ICU datasets across five critical clinical prediction tasks, KnowRare consistently outperforms state-of-the-art deep learning models and conventional scoring systems (e.g., APACHE IV), demonstrating superior generalizability and intrinsic interpretability through CKG-guided reasoning.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Knowledge Representation and Reasoning: Knowledge Acquisition
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Artificial Intelligence has revolutionised critical care for common conditions. Yet, rare conditions in the intensive care unit (ICU), including recognised rare diseases and low-prevalence conditions in the ICU, remain underserved due to data scarcity and intra-condition heterogeneity. To bridge such gaps, we developed KnowRare, a domain adaptation-based deep learning framework for predicting clinical outcomes for rare conditions in the ICU. KnowRare mitigates data scarcity by initially learning condition-agnostic representations from diverse electronic health records through self-supervised pre-training. It addresses intra-condition heterogeneity by selectively adapting knowledge from clinically similar conditions with a developed condition knowledge graph. Evaluated on two ICU datasets across five clinical prediction tasks (90-day mortality, 30-day readmission, ICU mortality, remaining length of stay, and phenotyping), KnowRare consistently outperformed existing state-of-the-art models. Additionally, KnowRare demonstrated superior predictive performance compared to established ICU scoring systems, including APACHE IV and IV-a. Case studies further demonstrated KnowRare's flexibility in adapting its parameters to accommodate dataset-specific and task-specific characteristics, its generalisation to common conditions under limited data scenarios, and its rationality in selecting source conditions. These findings highlight KnowRare's potential as a robust and practical solution for supporting clinical decision-making and improving care for rare conditions in the ICU.

Problem

• Addressing data scarcity for rare ICU conditions
• Mitigating intra-condition heterogeneity in clinical predictions
• Improving prediction accuracy for rare diseases in ICU

Innovation

• Self-supervised pre-training for condition-agnostic representations
• Condition knowledge graph for selective knowledge adaptation
• Multi-task deep learning framework for rare ICU conditions