Simulating Vein Compressibility for Ultrasound-Guided Femoral Access: Speed, Accuracy, and Phantom Validation
Organizations: Department of Mechanical Engineering, University of Minnesota, Minneapolis, Minnesota 55455, United States of America · Lightside Surgical, Bellevue, Washington 98005, United States of America · Healtcare Corporation of America, HCA HealthONE Trauma Quality Improvement Program Univ., Ocala, Florida · Visible Heart Laboratory, Department of Anesthesiology, University of Minnesota Medical School, Minneapolis, Minnesota 55455, United States of America
Abstract
In ultrasound-guided femoral access, probe pressure deforms the tissue and changes the vascular anatomy a robot observes. A simulator must reproduce these mechanics fast enough for robotic interaction, yet the computational effort needed for a clinically meaningful signal like vein collapse before artery is unclear. We test the feasibility of a physics-based simulation in NVIDIA's Isaac Sim for Healthcare, with Stable Neo-Hookean tissue solved by Vertex Block Descent on one GPU. A simulation of a low-fidelity cylindrical benchtop phantom with known geometry and mechanical properties, pressed by a robot, predicted its artery within 5 percentage points at every depth. It matched the vein within 3 points through 11 mm of a 25.4 mm press and followed the collapse behavior of the low-fidelity physical phantom. Across element size, time step, and solver iterations, wall clock time per frame scaled linearly with solver work, but accuracy did not improve with more work. The chosen setting reduced solver time per frame by a factor of 976, from 29,700 ms to 30.4 ms, while remaining within 1.03 points of the finest reference setting. The physics solver alone took 30.4 ms per frame, while the full stack, including ultrasound and graphics rendering, took 56.4 ms per frame. Finally, when seeded with realistic literature-based mechanical properties, the digital leg model compressed the vein far more than the artery at every setting, reproducing the expected clinical signal of vein collapse before artery, which is used to distinguish them. In total, these results suggest that the NVIDIA Isaac Sim platform may be suitable for augmenting reinforcement learning training or online simulation for autonomous femoral access with ultrasound.
Figures & tables
| Structure | Geometry (source) | (kPa) | |
|---|---|---|---|
| Skin and fat (one body) | shell 16.0 mm; limb diameter 184.9 mm (set) | 3 | 0.45 |
| Muscle | core, radius 76.5 mm | 20 | 0.45 |
| CFA wall | lumen 7.0 mm, wall 1.0 mm (set); intima 19.5 mm deep [ 33 ] | 600 | 0.45 |
| CFV wall | lumen 10.0 mm, wall 0.5 mm (set); intima 17.8 mm deep [ 33 ] ; outer far wall 29.2 mm vertical (28.2 mm perpendicular) | 20 | 0.45 |
| Femoral nerve | ellipse mm [ 34 ] ; 13.0 mm lateral to CFA (about 12.7 mm [ 35 ] ; measured mm [ 36 ] ) | 40 | 0.45 |
| Femoral sheath | constant section in the mesh (funnel in CAD only); –35 mm (set; usually 30–40 mm long [ 37 ] ); wall thickness set | 20 | 0.45 |
| Parameter | Leg model (CAD) | Benchtop |
| Length along the limb | 81 mm | 81 mm |
| Diameter | 184.9 mm | 184.1 mm |
| Skin and fat thickness | 16.0 mm | 15.9 mm |
| Artery lumen diameter | 6.96 mm | 9.00 mm |
| Vein lumen diameter | 9.95 mm | 10.00 mm |
| Artery wall thickness | 1.00 mm | 1.00 mm |
| Parameter | Slab (parameter study) | Leg model |
|---|---|---|
| Domain | block cut from the limb 20 mm below the inguinal ligament; side and deep faces held fixed | whole thigh segment, 81 mm long, held by a small fixed patch where the thigh rests on the table |
| Lateral width | 72.0 mm | 184.9 mm |
| Axial length | 16.5 mm | 81.0 mm |
| Depth | 82.5 mm | 184.9 mm |
| Nodes | 31,740 | 23,688 |
| Tetrahedra | 166,386 | 129,783 |
Physics-Based Simulation
Physics-based simulation models physical systems with equations and numerical solvers, increasingly augmented by learned models for speed and control.
See all 239 papers