RadTimeline: Timeline Summarization for Longitudinal Radiological Lung Findings

2026-03-24

AI Summary

This work addresses the challenges of tracking lesions and the time-consuming nature of manual summarization in longitudinal pulmonary radiology reports by formulating abstractive summarization as a structured timeline construction task. The proposed method employs a three-stage large language model pipeline: first extracting imaging findings, then generating semantic group names, and finally organizing these into a cross-temporal comparative timeline structure. By innovatively introducing group naming as an intermediate step, the approach significantly improves the accuracy of finding categorization. The authors also construct RadTimeline, the first dataset tailored for pulmonary imaging timelines. Experimental results demonstrate that the method achieves human-level performance in grouping while maintaining high recall, thereby validating the effectiveness and practicality of the proposed framework.

Technology Category

• Natural Language Processing: Summarization
• Data Mining & Knowledge Management: Data Visualization & Summarization
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Tracking findings in longitudinal radiology reports is crucial for accurately identifying disease progression, and the time-consuming process would benefit from automatic summarization. This work introduces a structured summarization task, where we frame longitudinal report summarization as a timeline generation task, with dated findings organized in columns and temporally related findings grouped in rows. This structured summarization format enables straightforward comparison of findings across time and facilitates fact-checking against the associated reports. The timeline is generated using a 3-step LLM process of extracting findings, generating group names, and using the names to group the findings. To evaluate such systems, we create RadTimeline, a timeline dataset focused on tracking lung-related radiologic findings in chest-related imaging reports. Experiments on RadTimeline show tradeoffs of different-sized LLMs and prompting strategies. Our results highlight that group name generation as an intermediate step is critical for effective finding grouping. The best configuration has some irrelevant findings but very good recall, and grouping performance is comparable to human annotators.

Problem

• timeline summarization
• longitudinal radiology
• lung findings
• radiological report
• disease progression

Innovation

• timeline summarization
• longitudinal radiology
• structured summarization
• large language models
• finding grouping