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

Forum topic · 小凯 · 2026-08-20

Summary

A paper posted on zhichai.net presents a locally deployed multi-agent AI system that combines radiology report structuring with quality assurance (QA) in a single workflow. The retrospective study included 638 chest, abdomen, and pelvis CT reports dictated by 15 board-certified radiologists in 2023–2024. The pipeline structures reports at sentence level into standardized anatomical sections using regex rules and local large language models, while detecting mismatches between Findings and Impression sections, gender-anatomy conflicts, and undocumented communication of critical findings. The system structured 22,270 sentences across all reports while preserving original content, and flagged 90 reports (14.1%), most commonly section mismatches (80, 12.5%). In independent evaluation of a 45-report subset, two radiologists agreed 31 reports (69%) were correctly restructured, 2 (4%) incorrectly, and no clinically important information was omitted or hallucinated. Overall QA performance was rated excellent or good in 84% of evaluated reports. Paper link: https://arxiv.org/abs/2608.18072.

Paper 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

Abstract Translation (from the forum post)

Purpose: To develop and evaluate a locally deployed multi-agent AI system for radiology report structuring and quality assurance.

Materials and Methods: This retrospective study included 638 radiology reports from CT examinations of the chest, abdomen, and pelvis dictated by 15 board-certified radiologists in 2023 and 2024. A multi-agent AI pipeline was developed to perform report structuring and quality assurance (QA). The system structured reports into standardized anatomical sections at the sentence level using regex rules and local large language models. It also detected mismatches between the Findings and Impression sections (or within sections), gender-anatomy conflicts, and undocumented communication of critical findings. Two board-certified radiologists independently evaluated a 45-report subset.

Results: The multi-agent system structured the Findings sections of all reports (22,270 sentences) into predefined anatomical formats while preserving original report content. The system flagged 90 reports (14.1%), most commonly section mismatches (80 reports, 12.5%). In radiologist evaluation, both reviewers agreed that 31 reports (69%) were correctly restructured and 2 (4%) incorrectly restructured, with disagreement on the remaining 12 (27%). Both reviewers agreed that no clinically important information was omitted and no fabricated content was introduced. Overall QA performance was rated "excellent" or "good" for 84% of evaluated reports, with the remainder rated "fair."

Conclusion: The locally deployed multi-agent AI system integrates radiology report structuring and quality assurance into a single workflow, showing favorable performance in radiologist evaluation. Such systems could support report standardization and QA in radiology practice.

Key Takeaways

Fully local deployment — no cloud dependency, relevant for patient-data privacy in clinical settings
Sentence-level structuring into standardized anatomical sections via regex + local LLMs
Automated QA checks: Findings/Impression mismatches, gender-anatomy conflicts, undocumented critical-findings communication
84% of evaluated reports rated excellent/good QA; no hallucination or omitted clinical information observed

---

*Auto-collected on 2026-08-20*