Automated Structured Radiology Report Generation

2025-05-30

AI Summary

Current automated chest X-ray (CXR) report generation faces key bottlenecks: inconsistent free-text formatting, poor clinical coherence, and coarse-grained evaluation. To address these, we introduce Structured Radiology Report Generation (SRRG)—a novel task—and present the first clinically compliant, structured CXR report dataset rigorously aligned with radiological reporting standards. We propose SRR-BERT, a hierarchical multi-label classification model grounded in a disease taxonomy comprising 55 clinically relevant categories. Furthermore, we design F1-SRR-BERT, a fine-grained evaluation metric that bridges the assessment gap between free-text and structured reports. Validated by five board-certified radiologists, our dataset exhibits high clinical fidelity. SRR-BERT achieves statistically significant improvements over baselines in fine-grained disease identification (p < 0.01). Reader studies confirm its clinical interpretability and practical utility. This work establishes a standardized paradigm for modeling and evaluating medical imaging report generation.

Technology Category

• Machine Learning: Structured Learning
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Computer Vision: Medical and Biological Imaging

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Web Mining and Content Analysis: Bridging structured and unstructured data
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

Automated radiology report generation from chest X-ray (CXR) images has the potential to improve clinical efficiency and reduce radiologists' workload. However, most datasets, including the publicly available MIMIC-CXR and CheXpert Plus, consist entirely of free-form reports, which are inherently variable and unstructured. This variability poses challenges for both generation and evaluation: existing models struggle to produce consistent, clinically meaningful reports, and standard evaluation metrics fail to capture the nuances of radiological interpretation. To address this, we introduce Structured Radiology Report Generation (SRRG), a new task that reformulates free-text radiology reports into a standardized format, ensuring clarity, consistency, and structured clinical reporting. We create a novel dataset by restructuring reports using large language models (LLMs) following strict structured reporting desiderata. Additionally, we introduce SRR-BERT, a fine-grained disease classification model trained on 55 labels, enabling more precise and clinically informed evaluation of structured reports. To assess report quality, we propose F1-SRR-BERT, a metric that leverages SRR-BERT's hierarchical disease taxonomy to bridge the gap between free-text variability and structured clinical reporting. We validate our dataset through a reader study conducted by five board-certified radiologists and extensive benchmarking experiments.

Problem

• Standardizing free-text radiology reports into structured format
• Improving consistency and clinical relevance of automated reports
• Developing precise evaluation metrics for structured radiology reports

Innovation

• Restructured reports using large language models
• Introduced SRR-BERT for disease classification
• Proposed F1-SRR-BERT for report evaluation