Graph Inference Towards ICD Coding
This work addresses a domain-specific problem in healthcare informatics for medical coding automation, with incremental improvements over existing methods.
The paper tackles automated ICD coding from clinical narratives by addressing label space size and class imbalance, introducing LabGraph which reformulates it as a graph generation task and achieves consistent improvements in micro-F1, micro-AUC, and P@K on benchmark datasets.
Automated ICD coding involves assigning standardized diagnostic codes to clinical narratives. The vast label space and extreme class imbalance continue to challenge precise prediction. To address these issues, LabGraph is introduced -- a unified framework that reformulates ICD coding as a graph generation task. By combining adversarial domain adaptation, graph-based reinforcement learning, and perturbation regularization, LabGraph effectively enhances model robustness and generalization. In addition, a label graph discriminator dynamically evaluates each generated code, providing adaptive reward feedback during training. Experiments on benchmark datasets demonstrate that LabGraph consistently outperforms previous approaches on micro-F1, micro-AUC, and P@K.