Virtual Coiling for Intracranial Aneurysm Based on Geometric Path Planning using Coil Pre-Shape
Patel, Palak Kalpeshbhai
MetadataShow full item record
Embolic coiling is the mainstay endovascular intervention for intracranial aneurysms. To enable better plan treatment, various computational algorithms are developed to simulate coil deployment in patient-specific aneurysms. Some fast algorithms use simplified physics to model a coil as a series of connected segments randomly marching to fill the aneurysmal space, resulting in unrealistic coil configurations. To make the coil deployment more realistic, we implemented a new geometric path planning algorithm using a governing constraint based on a coil’s tendency to restore its preshape geometry. Each successive coil segment advances from the tip of the previous segment along the preshape to mimic actual coil behavior upon release from the catheter. Upon collision with the aneurysm wall or already deployed coils, the segment would rotate its direction just enough to avoid the collision. Deployment of the entire coil is followed by a smoothing operation, emulating bending energy minimization to remove sharp kinks. We simulated coil deployments in several patient-specific aneurysms and compared results qualitatively against clinical angiograms. Furthermore, we virtually deployed coils in idealized geometries and evaluated simulated results against in-vitro experiments as well as published simulation results from previous methods. Our method did not produce unrealistic kinks on the coils near the corners and aneurysmal surfaces as with the previous methods as well as captures the gross coil behavior and configurations exhibited in experiments. Therefore, incorporating the coil preshape greatly improves coil behavior and generating more realistic coil configuration than previous pseudo-physics based models.