Related Experiment Video
Updated: Jun 5, 2026
07:46
Automatic Surgery in Transcatheter Aortic Valve Replacement Using Augmented Reality
Published on: August 9, 2024
SDFStent: Real-time interactive virtual stenting via SDF deformation fields
Arxiv
|June 4, 2026
Summary
SDFStent virtually models stent placement in congenital heart disease (CHD) faster and more accurately than existing methods. This open-source tool generates patient-specific models for improved computational fluid dynamics (CFD) simulations.
Area of Science:
- Biomedical Engineering
- Medical Imaging
- Computational Science
Background:
- Transcatheter stenting is a common intervention for congenital heart disease (CHD).
- Accurate computational fluid dynamics (CFD) simulations require precise post-operative vascular geometries, which are challenging to generate due to stent-induced shape changes.
- Existing methods for modeling stent-induced geometry changes are often time-consuming, require manual effort, or inadequately represent the complex shapes.
Purpose of the Study:
- To introduce SDFStent, a novel signed distance function (SDF) based method for real-time virtual stenting.
- To enable the generation of patient-specific vascular geometries that accurately reflect stent implantation for CFD simulations.
- To improve the efficiency and accuracy of pre-operative planning and post-operative outcome prediction in CHD interventions.
Main Methods:
- Developed SDFStent, a signed distance function (SDF) based mesh deformation technique for virtual stenting.
- Modeled the stent as a piecewise-capsule SDF surface combined with a smooth-minimum operator.
- Displaced mesh vertices along the SDF gradient using a compactly supported fall-off function and alpha blending for real-time mesh deformation, ensuring mesh integrity and junction geometry preservation.
Main Results:
- SDFStent achieved a mean stented diameter of 5.92 ± 0.08 mm (prescribed 6.0 mm) in 1.5 seconds, over 100 times faster than the best comparator.
- Generated watertight and self-intersection-free meshes for all tested cases.
- CFD simulations using SDFStent-generated models showed post-operative pressure drops within 4 mmHg (mean error 2 mmHg) of clinical measurements for tetralogy of Fallot (ToF) and coarctation of the aorta (CoA) patients.
Conclusions:
- SDFStent provides a rapid and accurate method for generating simulation-ready post-stent vascular models from pre-operative data.
- The tool's real-time performance and accuracy enhance the feasibility of virtual stenting for pre-operative planning and outcome prediction in CHD.
- The open-source availability and scriptable architecture facilitate large-scale synthetic cohort generation for data-driven surrogate modeling in cardiovascular research.
