CEVAR: Centerline Embedding Extraction for Endovascular Aneurysm Repair

2026-06-14

AI Summary

Endovascular aneurysm repair (EVAR) can fail due to loss of seal at the stent-graft anchoring zones, potentially leading to aneurysm rupture; however, current CT-based assessment relies on manual centerline editing, which is time-consuming and requires expert intervention. This work proposes the first Transformer-based framework that integrates image-to-graph modeling with embedded geometric prediction to enable fully automatic aorto-iliac centerline tracing guided by the EVAR4C protocol, simultaneously estimating stent-graft position, vessel diameter, and seal length. The method significantly outperforms commercial semi-automatic workflows on both the full test set and a contrast-free subset, markedly improving evaluation efficiency and consistency.

Abstract

Long-term mortality rates after endovascular aneurysm repair (EVAR) remain elevated due to post-EVAR rupture caused by loss of seal in stent graft sealing zones. Structured CT review using centerline measurements improves detection, but current workflows require manual centerline editing and expert operators. We propose a transformer framework for automated, protocol-driven sealing zone assessment that combines 3D centerline tracking with embedding-based geometric prediction. Two state-of-the-art image-to-graph models are evaluated for aorto-iliac centerline extraction from follow-up CT and for measurement of stent position, vessel diameters, and seal lengths according to EVAR4C protocol. Across the full test set and a challenging no-contrast subset, the proposed fully automatic method outperforms the commercial semi-automatic workflow.

Problem

• Endovascular Aneurysm Repair
• Centerline Extraction
• Sealing Zone Assessment
• CT Imaging
• Aorto-iliac

Innovation

• transformer framework
• centerline embedding
• automated EVAR assessment
• 3D centerline tracking
• EVAR4C protocol