GRACE: GRaph-based Addiction Care prEdiction

2025-10-23

AI Summary

To address the poor minority-class prediction performance in addiction treatment level (care tier) assignment—caused by severe class imbalance in clinical data—this paper proposes an unbiased meta-graph-based graph neural network (GNN) framework. The method formulates care tier prediction as a structured learning task, explicitly modeling associations between patient clinical features and heterogeneous healthcare resources via meta-graphs, while integrating deep feature engineering to mitigate label bias. Experiments on real-world electronic health record data demonstrate that the proposed approach improves minority-class F1-score by 11–35% over state-of-the-art baselines, substantially enhancing clinical decision support. Key contributions include: (i) the first application of an unbiased meta-graph GNN to tackle class imbalance in addiction care tier prediction; and (ii) simultaneous optimization of interpretability—through explicit patient-resource relational modeling—and generalization performance.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• 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
• Web Mining and Content Analysis: Large pretrained models with web data
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Determining the appropriate locus of care for addiction patients is one of the most critical clinical decisions that affects patient treatment outcomes and effective use of resources. With a lack of sufficient specialized treatment resources, such as inpatient beds or staff, there is an unmet need to develop an automated framework for the same. Current decision-making approaches suffer from severe class imbalances in addiction datasets. To address this limitation, we propose a novel graph neural network (GRACE) framework that formalizes locus of care prediction as a structured learning problem. Further, we perform extensive feature engineering and propose a new approach of obtaining an unbiased meta-graph to train a GNN to overcome the class imbalance problem. Experimental results in real-world data show an improvement of 11-35% in terms of the F1 score of the minority class over competitive baselines. The codes and note embeddings are available at https://anonymous.4open.science/r/GRACE-F8E1/.

Problem

• Predicting optimal addiction care placement using graph networks
• Addressing severe class imbalance in clinical decision-making data
• Automating locus of care determination to improve resource allocation

Innovation

• Graph neural network for structured care prediction
• Feature engineering to address class imbalance
• Unbiased meta-graph training for minority class improvement