ALTER: Modeling Longitudinal Changes via Regional Differencing for 3D CT Report Generation

2026-08-06

AI Summary

This work addresses the challenge that existing 3D CT report generation methods struggle to accurately capture fine-grained longitudinal changes in patients, particularly due to the absence of anatomical region-level difference modeling. To overcome this limitation, we propose a change-aware report generation framework that integrates historical imaging and reports through Global Prior Integration (GPI). Our approach introduces a Region Proxy Differencing (RPD) mechanism to retrieve corresponding historical regions for each anatomical area and compute localized temporal differences. Furthermore, an Interval Change Fusion (ICF) strategy aggregates these multi-region temporal variations into a unified soft prompt to guide report generation. This method represents the first attempt to model longitudinal changes at the anatomical region level, significantly improving the accuracy and completeness of generated reports on the RadGenome-ChestCT and CTRG-Chest-548K datasets.

Abstract

Computed tomography (CT) is widely used for clinical diagnosis and longitudinal follow-up, yet automatically generating accurate and complete radiology reports from three-dimensional (3D) CT remains challenging. Existing methods improve fine-grained correspondence between images and text by modeling anatomical regions, but remain centered on the current examination. Consequently, patient-specific longitudinal changes within individual regions remain insufficiently modeled. Meanwhile, interval changes are often distributed across multiple anatomical regions, complicating a coherent assessment of the overall longitudinal state. We propose Anatomically Localized Temporal Evidence Representation (ALTER) to address these limitations. Global Prior Integration (GPI) incorporates the prior CT and report to establish historical context for the current examination. Regional Proxy Differencing (RPD) enables each current anatomical region to retrieve a historical proxy from a single shared encoding of the prior volume and to derive localized interval evidence. Interval Change Fusion (ICF) further combines current abnormality states with region-distributed differences, converting their joint representation into change-aware soft prompts that guide report generation. ALTER achieves state-of-the-art results on most evaluation metrics across the RadGenome-ChestCT validation and CTRG-Chest-548K test sets. Code and data preprocessing details are available at https://github.com/peytonkarlie/ALTER/tree/main.

Problem

• longitudinal changes
• 3D CT report generation
• anatomical regions
• interval changes
• radiology report

Innovation

• longitudinal modeling
• regional differencing
• 3D CT report generation
• temporal evidence representation
• change-aware prompting