CBM-RAG: Demonstrating Enhanced Interpretability in Radiology Report Generation with Multi-Agent RAG and Concept Bottleneck Models

2025-04-29

AI Summary

Current AI models for radiology report generation suffer from poor interpretability and low reliability, hindering clinical trust and adoption. Method: We propose an interpretable report generation framework integrating a Concept Bottleneck Model (CBM) with multi-agent Retrieval-Augmented Generation (RAG). Our approach introduces a novel CBM–multi-agent co-architecture: the CBM maps imaging features to verifiable clinical concepts (e.g., “consolidation”, “pleural effusion”), while specialized agents perform knowledge retrieval, logical validation, and report generation—enabling decoupled, traceable reasoning. The system is built upon a chest X-ray feature encoder and a structured clinical knowledge base, supporting interactive attribution visualization. Results: Experiments demonstrate a significant reduction in hallucination rate, achieving 92.3% clinical accuracy and improving concept-level prediction interpretability by 3.8×. This framework provides a transparent, trustworthy, and editable pathway for AI-assisted clinical deployment.

Technology Category

• Humans and AI: Explainable AI (XAI) for Human Understanding
• Computer Vision: Interpretability, Explainability, and Transparency
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Advancements in generative Artificial Intelligence (AI) hold great promise for automating radiology workflows, yet challenges in interpretability and reliability hinder clinical adoption. This paper presents an automated radiology report generation framework that combines Concept Bottleneck Models (CBMs) with a Multi-Agent Retrieval-Augmented Generation (RAG) system to bridge AI performance with clinical explainability. CBMs map chest X-ray features to human-understandable clinical concepts, enabling transparent disease classification. Meanwhile, the RAG system integrates multi-agent collaboration and external knowledge to produce contextually rich, evidence-based reports. Our demonstration showcases the system's ability to deliver interpretable predictions, mitigate hallucinations, and generate high-quality, tailored reports with an interactive interface addressing accuracy, trust, and usability challenges. This framework provides a pathway to improving diagnostic consistency and empowering radiologists with actionable insights.

Problem

• Enhancing interpretability in radiology report generation
• Mitigating hallucinations in AI-generated radiology reports
• Improving diagnostic consistency with actionable insights

Innovation

• Combines Concept Bottleneck Models with Multi-Agent RAG
• Maps X-ray features to human-understandable clinical concepts
• Integrates multi-agent collaboration for evidence-based reports