Multi-Agent AI System for Radiology Report Structuring and Quality Assurance (arXiv 2608.18072)

Forum topic · 小凯 · 2026-08-20

Summary

Researchers developed and evaluated a locally deployed multi-agent AI system that combines radiology report structuring and quality assurance in a single workflow. In a retrospective study of 638 chest, abdomen, and pelvis CT reports dictated by 15 board-certified radiologists (2023–2024), the pipeline used regex rules and local large language models to structure reports at the sentence level into standardized anatomical sections. The system also flagged Findings–Impression mismatches, gender-anatomy conflicts, and undocumented communication of critical findings, marking 90 reports (14.1%), most commonly section mismatches (80 reports, 12.5%). Two radiologists independently reviewed a 45-report subset, agreeing that 31 reports (69%) were correctly restructured, 2 (4%) incorrectly, with no omitted clinically important information or fabricated content. Overall QA performance was rated excellent or good in 84% of evaluated reports. The study suggests locally deployed multi-agent systems can support report standardization and QA in radiology practice.

Overview

Field: NLP / Medical AI
Authors: Iryna Hartsock, Cesar Lam, Christopher Otteni, Aliya Qayyum, Robert Gatenby, Cyrillo Araujo, Ghulam Rasool
Published: 2026-08-18
arXiv: 2608.18072

Purpose

To develop and evaluate a locally deployed multi-agent AI system for radiology report structuring and quality assurance (QA).

Materials and Methods

Retrospective study of 638 radiology reports from chest, abdomen, and pelvis CT examinations, dictated by 15 board-certified radiologists in 2023–2024.
A multi-agent AI pipeline performs both report structuring and QA.
Reports are structured into standardized anatomical sections at the sentence level using regex rules and local large language models.
The QA agent detects:
Mismatches between the Findings and Impression sections, or within sections
Gender–anatomy conflicts
Undocumented communication of critical findings
Two board-certified radiologists independently evaluated a 45-report subset.

Results

The system structured the Findings sections of all reports (22,270 sentences) into the predefined anatomical format while preserving original report content.
90 reports (14.1%) were flagged, most commonly for section mismatches (80 reports, 12.5%).
Radiologist review of the subset: both reviewers agreed 31 reports (69%) were correctly restructured, 2 (4%) incorrectly; they disagreed on the remaining 12 (27%).
Both reviewers agreed that no clinically important information was omitted and no fabricated content (hallucinations) was introduced.
Overall QA performance was rated "excellent" or "good" in 84% of evaluated reports; the remainder were rated "fair".

Conclusion

A locally deployed multi-agent AI system can combine radiology report structuring and quality assurance into a single workflow, showing good performance in radiologist evaluation. Such systems may support report standardization and quality assurance in radiology practice.

---

*Auto-collected on 2026-08-20.*