Improving automatic cerebral 3D-2D CTA-DSA registration

Journal Article (2025)
Author(s)

Charles Downs (Erasmus MC)

P. Matthijs van der Sluijs (Erasmus MC)

Sandra A.P. Cornelissen (Erasmus MC)

Frank te Nijenhuis (Erasmus MC, TU Delft - Electrical Engineering, Mathematics and Computer Science)

Wim H.van Zwam (Maastricht University Medical Center)

Vivek Gopalakrishnan (Massachusetts Institute of Technology)

Xucong Zhang (TU Delft - Pattern Recognition and Bioinformatics)

Ruisheng Su (Eindhoven University of Technology, Erasmus MC, TU Delft - Biomechanical Engineering)

Theo van Walsum (TU Delft - Biomechanical Engineering, Erasmus MC)

Faculty
Electrical Engineering, Mathematics and Computer Science
DOI related publication
https://doi.org/10.1007/s11548-025-03412-2
More Info
expand_more
Publication Year
2025
Language
English
Faculty
Electrical Engineering, Mathematics and Computer Science
Journal title
International Journal of Computer Assisted Radiology and Surgery
Issue number
7
Volume number
20
Article number
102392
Pages (from-to)
1451-1460
Downloads counter
278
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

Purpose : Stroke remains a leading cause of morbidity and mortality worldwide, despite advances in treatment modalities. Endovascular thrombectomy (EVT), a revolutionary intervention for ischemic stroke, is limited by its reliance on 2D fluoroscopic imaging, which lacks depth and comprehensive vascular detail. We propose a novel AI-driven pipeline for 3D CTA to 2D DSA cross-modality registration, termed DeepIterReg. Methods : The proposed pipeline integrates neural network-based initialization with iterative optimization to align pre-intervention and peri-intervention data. Our approach addresses the challenges of cross-modality alignment, particularly in scenarios involving limited shared vascular structures, by leveraging synthetic data, vein-centric anchoring, and differentiable rendering techniques. Results : We assess the efficacy of DeepIterReg through quantitative analysis of capture ranges and registration accuracy. Results show that our method can accurately register 70% of a test set of 20 patients and can improve capture ranges when performing an initial pose estimation using a convolutional neural network. Conclusions : DeepIterReg demonstrates promising performance for 3D-to-2D stroke intervention image registration, potentially aiding clinicians by improving spatial understanding during EVT and reducing dependence on manual adjustments.

Files

S11548-025-03412-2.pdf
(pdf | 1.43 Mb)
License info not available