Radiologist Copilot: An Agentic Assistant with Orchestrated Tools for Radiology Reporting with Quality Control

2025-12-02

AI Summary

Radiology report generation and quality control involve cumbersome, error-prone workflows; existing methods typically address only single-stage report generation without end-to-end quality assurance. This paper proposes the first agent-based AI framework tailored to the full radiology workflow, integrating region localization, image-grounded reasoning planning, dynamic clinical template matching, and multi-dimensional quality assessment—unified within a feedback-driven iterative optimization loop. Built upon a large language model backbone, the framework synergistically leverages medical vision-language models and feedback-aware fine-tuning to enable autonomous, multi-stage decision-making—from multimodal image understanding to compliant, clinically accurate report generation. Evaluated on multiple public benchmarks, our method significantly outperforms state-of-the-art approaches in report accuracy, clinical completeness, and guideline adherence, thereby enhancing both diagnostic reporting quality and clinical workflow efficiency.

Technology Category

• Computer Vision: Multi-modal Vision
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

Radiology reporting is an essential yet time-consuming and error-prone task for radiologists in clinical examinations, especially for volumetric medical images. Rigorous quality control is also critical but tedious, ensuring that the final report meets clinical standards. Existing automated approaches, including radiology report generation methods and medical vision-language models, focus mainly on the report generation phase and neglect the crucial quality control procedure, limiting their capability to provide comprehensive support to radiologists. We propose Radiologist Copilot, an agentic AI assistant equipped with orchestrated tools designed for automated radiology reporting with quality control. Leveraging large language models as the reasoning backbone, the agentic system autonomously selects tools, plans, and executes actions, emulating the behavior of radiologists throughout the holistic radiology reporting process. The orchestrated tools include region localization, think with image paradigm directed region analysis planning, strategic template selection for report generation, quality assessment and feedback-driven adaptive refinement for quality control. Therefore, Radiologist Copilot facilitates accurate, complete, and efficient radiology reporting, assisting radiologists and improving clinical efficiency. Experimental results demonstrate that Radiologist Copilot significantly surpasses other state-of-the-art methods in radiology reporting. The source code will be released upon acceptance.

Problem

• Automates radiology reporting with integrated quality control
• Addresses limitations of existing methods lacking quality assessment
• Assists radiologists in accurate, efficient clinical reporting processes

Innovation

• Agentic AI assistant with orchestrated tools for radiology reporting
• Uses large language models as reasoning backbone for autonomous tool selection
• Integrates quality assessment and feedback-driven refinement for quality control