MARCH: Multi-Agent Radiology Clinical Hierarchy for CT Report Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lin, Yi, Ding, Yihao, Wu, Yonghui, Peng, Yifan
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917416167014400
author Lin, Yi
Ding, Yihao
Wu, Yonghui
Peng, Yifan
author_facet Lin, Yi
Ding, Yihao
Wu, Yonghui
Peng, Yifan
contents Automated 3D radiology report generation often suffers from clinical hallucinations and a lack of the iterative verification found in human practice. While recent Vision-Language Models (VLMs) have advanced the field, they typically operate as monolithic "black-box" systems without the collaborative oversight characteristic of clinical workflows. To address these challenges, we propose MARCH (Multi-Agent Radiology Clinical Hierarchy), a multi-agent framework that emulates the professional hierarchy of radiology departments and assigns specialized roles to distinct agents. MARCH utilizes a Resident Agent for initial drafting with multi-scale CT feature extraction, multiple Fellow Agents for retrieval-augmented revision, and an Attending Agent that orchestrates an iterative, stance-based consensus discourse to resolve diagnostic discrepancies. On the RadGenome-ChestCT dataset, MARCH significantly outperforms state-of-the-art baselines in both clinical fidelity and linguistic accuracy. Our work demonstrates that modeling human-like organizational structures enhances the reliability of AI in high-stakes medical domains.
format Preprint
id arxiv_https___arxiv_org_abs_2604_16175
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle MARCH: Multi-Agent Radiology Clinical Hierarchy for CT Report Generation
Lin, Yi
Ding, Yihao
Wu, Yonghui
Peng, Yifan
Artificial Intelligence
Computer Vision and Pattern Recognition
Automated 3D radiology report generation often suffers from clinical hallucinations and a lack of the iterative verification found in human practice. While recent Vision-Language Models (VLMs) have advanced the field, they typically operate as monolithic "black-box" systems without the collaborative oversight characteristic of clinical workflows. To address these challenges, we propose MARCH (Multi-Agent Radiology Clinical Hierarchy), a multi-agent framework that emulates the professional hierarchy of radiology departments and assigns specialized roles to distinct agents. MARCH utilizes a Resident Agent for initial drafting with multi-scale CT feature extraction, multiple Fellow Agents for retrieval-augmented revision, and an Attending Agent that orchestrates an iterative, stance-based consensus discourse to resolve diagnostic discrepancies. On the RadGenome-ChestCT dataset, MARCH significantly outperforms state-of-the-art baselines in both clinical fidelity and linguistic accuracy. Our work demonstrates that modeling human-like organizational structures enhances the reliability of AI in high-stakes medical domains.
title MARCH: Multi-Agent Radiology Clinical Hierarchy for CT Report Generation
topic Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2604.16175