Integrating clinical reasoning into large language model-based diagnosis through etiology-aware attention steering

2025-07-31

AI Summary

Large language models (LLMs) exhibit insufficient diagnostic reliability and weak interpretability in complex clinical scenarios. Method: This paper proposes a causality-aware attention-guidance framework that leverages clinical guidelines as structured reasoning priors, incorporates causality-aware attention heads to identify critical diagnostic evidence, and introduces a reasoning-guided, parameter-efficient fine-tuning mechanism with a dedicated loss function to explicitly align model attention distributions with medical reasoning logic. Results: On a consistency diagnosis cohort, the framework improves average diagnostic accuracy by 15.65% and reasoning focus score by 31.6%. External validation demonstrates significant gains in diagnostic reliability and decision interpretability on real-world complex cases. To our knowledge, this is the first work to deeply integrate causality-driven attention modulation with structured clinical reasoning, establishing a novel paradigm for trustworthy medical AI.

Technology Category

• Reasoning under Uncertainty: Causality
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Machine Learning: Causal Learning

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Objective: Large Language Models (LLMs) demonstrate significant capabilities in medical text understanding and generation. However, their diagnostic reliability in complex clinical scenarios remains limited. This study aims to enhance LLMs' diagnostic accuracy and clinical reasoning ability. Method: We propose an Etiology-Aware Attention Steering Framework to integrate structured clinical reasoning into LLM-based diagnosis. Specifically, we first construct Clinical Reasoning Scaffolding (CRS) based on authoritative clinical guidelines for three representative acute abdominal emergencies: acute appendicitis, acute pancreatitis, and acute cholecystitis. Next, we develop the Etiology-Aware Head Identification algorithm to pinpoint attention heads crucial for the model's etiology reasoning. To ensure reliable clinical reasoning alignment, we introduce the Reasoning-Guided Parameter-Efficient Fine-tuning that embeds etiological reasoning cues into input representations and steers the selected Etiology-Aware Heads toward critical information through a Reasoning-Guided Loss function. Result: On the Consistent Diagnosis Cohort, our framework improves average diagnostic accuracy by 15.65% and boosts the average Reasoning Focus Score by 31.6% over baselines. External validation on the Discrepant Diagnosis Cohort further confirms its effectiveness in enhancing diagnostic accuracy. Further assessments via Reasoning Attention Frequency indicate that our models exhibit enhanced reliability when faced with real-world complex scenarios. Conclusion: This study presents a practical and effective approach to enhance clinical reasoning in LLM-based diagnosis. By aligning model attention with structured CRS, the proposed framework offers a promising paradigm for building more interpretable and reliable AI diagnostic systems in complex clinical settings.

Problem

• Enhance LLMs' diagnostic accuracy in complex clinical scenarios
• Integrate structured clinical reasoning into LLM-based diagnosis
• Improve reliability of AI diagnostic systems via attention steering

Innovation

• Etiology-Aware Attention Steering Framework
• Clinical Reasoning Scaffolding construction
• Reasoning-Guided Parameter-Efficient Fine-tuning