FT-MDT: Extracting Decision Trees from Medical Texts via a Novel Low-rank Adaptation Method
This addresses the need for efficient clinical decision support systems by automating MDT extraction, though it is incremental as it builds on existing parameter-efficient fine-tuning approaches.
The paper tackles the problem of automatically extracting medical decision trees from clinical texts to reduce manual annotation, proposing a novel low-rank adaptation method that achieves state-of-the-art accuracy with reduced model complexity.
Knowledge of the medical decision process, which can be modeled as medical decision trees (MDTs), is critical to building clinical decision support systems. However, current MDT construction methods rely heavily on time-consuming and laborious manual annotation. To address this challenge, we propose PI-LoRA (Path-Integrated LoRA), a novel low-rank adaptation method for automatically extracting MDTs from clinical guidelines and textbooks. We integrate gradient path information to capture synergistic effects between different modules, enabling more effective and reliable rank allocation. This framework ensures that the most critical modules receive appropriate rank allocations while less important ones are pruned, resulting in a more efficient and accurate model for extracting medical decision trees from clinical texts. Extensive experiments on medical guideline datasets demonstrate that our PI-LoRA method significantly outperforms existing parameter-efficient fine-tuning approaches for the Text2MDT task, achieving better accuracy with substantially reduced model complexity. The proposed method achieves state-of-the-art results while maintaining a lightweight architecture, making it particularly suitable for clinical decision support systems where computational resources may be limited.