A Review of Longitudinal Radiology Report Generation: Dataset Composition, Methods, and Performance Evaluation

2025-10-14

AI Summary

Current chest X-ray report generation methods predominantly rely on single-image inputs, limiting their capacity to model disease progression and resulting in reports lacking longitudinal comparability—critical for clinical decision-making. This work pioneers the first systematic investigation of longitudinal radiology report generation, introducing a comprehensive framework spanning dataset construction, model architecture design, and domain-specific evaluation. Methodologically, we integrate vision-language models with longitudinal imaging modeling techniques to emulate radiologists’ diagnostic reasoning. Ablation studies rigorously assess the impact of longitudinal information, architectural choices, and evaluation protocols. Experiments demonstrate that longitudinal modeling substantially improves report accuracy and clinical consistency. Furthermore, we identify five recurrent limitations across existing approaches and propose structured directions for future research, establishing foundational benchmarks and methodological guidelines for longitudinal radiology NLP.

Technology Category

• Natural Language Processing: Language Grounding & Multi-modal NLP
• Machine Learning: Life-Long and Continual Learning
• Computer Vision: Multi-modal Vision

Application Category

• User Modeling, Personalization and Recommendation: Studies of user behavior, including longitudinal effects of personalized systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Chest Xray imaging is a widely used diagnostic tool in modern medicine, and its high utilization creates substantial workloads for radiologists. To alleviate this burden, vision language models are increasingly applied to automate Chest Xray radiology report generation (CXRRRG), aiming for clinically accurate descriptions while reducing manual effort. Conventional approaches, however, typically rely on single images, failing to capture the longitudinal context necessary for producing clinically faithful comparison statements. Recently, growing attention has been directed toward incorporating longitudinal data into CXR RRG, enabling models to leverage historical studies in ways that mirror radiologists diagnostic workflows. Nevertheless, existing surveys primarily address single image CXRRRG and offer limited guidance for longitudinal settings, leaving researchers without a systematic framework for model design. To address this gap, this survey provides the first comprehensive review of longitudinal radiology report generation (LRRG). Specifically, we examine dataset construction strategies, report generation architectures alongside longitudinally tailored designs, and evaluation protocols encompassing both longitudinal specific measures and widely used benchmarks. We further summarize LRRG methods performance, alongside analyses of different ablation studies, which collectively highlight the critical role of longitudinal information and architectural design choices in improving model performance. Finally, we summarize five major limitations of current research and outline promising directions for future development, aiming to lay a foundation for advancing this emerging field.

Problem

• Automating longitudinal chest X-ray report generation to reduce radiologist workload
• Incorporating historical imaging data for clinically accurate comparison statements
• Addressing limitations in existing surveys by providing systematic framework guidance

Innovation

• Incorporates longitudinal data for clinical context
• Leverages historical studies to mirror diagnostic workflows
• Uses tailored architectures for longitudinal report generation