Presentation Title

Simvascular: Optimizing and Comparing 3D models

Start Date

November 2016

End Date

November 2016

Location

HUB 302-#39

Type of Presentation

Poster

Abstract

Cardiovascular disease is the leading cause of death worldwide. Typically, cardiovascular disease is caused or enhanced by disruptions of blood flow in the arteries. Patient-specific blood flow modeling was pioneered in the late 1990’s to investigate and predict the progression of cardiovascular disease, and in recent years, has proven to be a powerful tool in clinical research [1].

In the 2000’s, researchers needed to purchase very expensive and robust specialized commercial tools for arterial model construction and blood flow simulation. Recently, a team of researchers at Stanford and UC Berkeley has promoted research in this area by creating an open source software pipeline called SimVascular. SimVascular provides a framework to reconstruct an arterial model from MRI or CT data, and then perform blood flow simulations on the reconstructed model. Despite being the leading computational tool for arterial modeling and blood flow simulation, it still takes an experienced user multiple hours to create an arterial geometry in SimVascular.

This paper will discuss our efforts to: (1) Lead the development of open source one-click conversion of image data into accurate 3D arterial models, (2) compare the open source models to previously created commercial models, and (3) make available a large number of open source arterial models that are ready for blood flow simulation and disease investigation. Through these efforts the group successfully created 15 open source models, a point based metric for comparing open source and commercial models, and an automatic model smoothing function decreasing the model creation time by 20 percent. Index Terms – Polydata, Parasolid, Segmentation, Hausdorff distance

This document is currently not available here.

Share

COinS
 
Nov 12th, 4:00 PM Nov 12th, 5:00 PM

Simvascular: Optimizing and Comparing 3D models

HUB 302-#39

Cardiovascular disease is the leading cause of death worldwide. Typically, cardiovascular disease is caused or enhanced by disruptions of blood flow in the arteries. Patient-specific blood flow modeling was pioneered in the late 1990’s to investigate and predict the progression of cardiovascular disease, and in recent years, has proven to be a powerful tool in clinical research [1].

In the 2000’s, researchers needed to purchase very expensive and robust specialized commercial tools for arterial model construction and blood flow simulation. Recently, a team of researchers at Stanford and UC Berkeley has promoted research in this area by creating an open source software pipeline called SimVascular. SimVascular provides a framework to reconstruct an arterial model from MRI or CT data, and then perform blood flow simulations on the reconstructed model. Despite being the leading computational tool for arterial modeling and blood flow simulation, it still takes an experienced user multiple hours to create an arterial geometry in SimVascular.

This paper will discuss our efforts to: (1) Lead the development of open source one-click conversion of image data into accurate 3D arterial models, (2) compare the open source models to previously created commercial models, and (3) make available a large number of open source arterial models that are ready for blood flow simulation and disease investigation. Through these efforts the group successfully created 15 open source models, a point based metric for comparing open source and commercial models, and an automatic model smoothing function decreasing the model creation time by 20 percent. Index Terms – Polydata, Parasolid, Segmentation, Hausdorff distance