Leveraging Evidence-Guided LLMs to Enhance Trustworthy Depression Diagnosis

2025/11/22 by Yuan, Yining, Tamo, J. Ben, Nnamdi, Micky C. +2
Computer Science · Medicine · Psychology · #Artificial Intelligence (cs.AI) #Artificial Intelligence in Healthcare and Education #FOS: Computer and information sciences #Information Retrieval (cs.IR) #Machine Learning in Healthcare #Mental Health via Writing

paper · doi:10.48550/arxiv.2511.17947

openalex publication_date 2025/11/22 · openalex created_date 2025/11/27 · openalex updated_date 2026/07/28

Abstract

Large language models (LLMs) show promise in automating clinical diagnosis, yet their non-transparent decision-making and limited alignment with diagnostic standards hinder trust and clinical adoption. We address this challenge by proposing a two-stage diagnostic framework that enhances transparency, trustworthiness, and reliability. First, we introduce Evidence-Guided Diagnostic Reasoning (EGDR), which guides LLMs to generate structured diagnostic hypotheses by interleaving evidence extraction with logical reasoning grounded in DSM-5 criteria. Second, we propose a Diagnosis Confidence Scoring (DCS) module that evaluates the factual accuracy and logical consistency of generated diagnoses through two interpretable metrics: the Knowledge Attribution Score (KAS) and the Logic Consistency Score (LCS). Evaluated on the D4 dataset with pseudo-labels, EGDR outperforms direct in-context prompting and Chain-of-Thought (CoT) across five LLMs. For instance, on OpenBioLLM, EGDR improves accuracy from 0.31 (Direct) to 0.76 and increases DCS from 0.50 to 0.67. On MedLlama, DCS rises from 0.58 (CoT) to 0.77. Overall, EGDR yields up to +45% accuracy and +36% DCS gains over baseline methods, offering a clinically grounded, interpretable foundation for trustworthy AI-assisted diagnosis.

Citations

• Causal Machine Learning for Surgical Interventions
• Claim Verification in the Age of Large Language Models: A Survey
• ClaimVer: Explainable Claim-Level Verification and Evidence Attribution of Text Through Knowledge Graphs
• Few shot chain-of-thought driven reasoning to prompt LLMs for open ended medical question answering
• Large Language Models are Clinical Reasoners: Reasoning-Aware Diagnosis Framework with Prompt-Generated Rationales
• Explainable Claim Verification via Knowledge-Grounded Reasoning with Large Language Models
• MDDial: A Multi-turn Differential Diagnosis Dialogue Dataset with Reliability Evaluation
• Diagnostic Reasoning Prompts Reveal the Potential for Large Language Model Interpretability in Medicine
• Large Language Models Perform Diagnostic Reasoning
• Retrieval-Augmented and Knowledge-Grounded Language Models for Faithful Clinical Medicine

Related

• MDD-Thinker: Towards Large Reasoning Models for Major Depressive Disorder Diagnosis
• Improving clinical reliability of LLM reasoning for depression assessment via structured generation and GRPO
• MDD-LLM: Towards Accuracy Large Language Models for Major Depressive Disorder Diagnosis
• Standardization of Psychiatric Diagnoses -- Role of Fine-tuned LLM Consortium and OpenAI-gpt-oss Reasoning LLM Enabled Decision Support System
• Self-Evolving Human-Centered Framework for Explainable Depression Symptom Annotation