DEPARTMENT OF MECHANICAL ENGINEERING
WILLIAM MAXWELL REED SEMINAR SERIES
Analysis and Design of Functionally Optimized
Image-based Computational Analysis Aiming to Aid Clinical Decision Making for Hemodynamic Abnormality in
Cardiovascular Arteries
Huidan (Whitney) Yu
Department of Mechanical Engineering, Indiana University-Purdue University Indianapolis, IN
Surgery Division of Vascular Surgery, School of Medicine, Indiana University, IN
ABSTRACT: Patient-specific flow simulation in human subject has emerged as a powerful research tool to noninvasively reveal in vivo
flow pattern, pressure and wall shear stress (WSS) distribution. The attractive advantages include (1) the low cost of facility, personnel, and
supplies; (2) the fully human subject protection; (3) the short and flexible time cycle, and (4) the direct human subject results. Radio-graphical
evaluation and animal model experimentation cannot compete with these advantages and achieve similar results with the same investment.
We have developed a unique and efficient research tool for patient-specific computational hemodynamics and flow-vessel interaction using
lattice Boltzmann method (LBM), which seamlessly integrates anatomical-structure extraction from imaging data and flow modeling in one
computation mesh structure thus geometry formation and mesh generation are not required. The LBM first solves a level set equation for
image segmentation and then extract the boundary information. The obtained patient-specific vessel geometry, volumetric ratio of solid versus
fluid, and the orientation of the boundary are seamlessly fed to the next step for solving the Navier-Stokes equation. From CT/MARI images
to in vivo flow and stress quantification, there are no data transformation and software involved thus the computation can be efficiently
accelerated by GPU technology. It has been estimated that a typical cardiac simulation of blood flow in a human aortic or carotid artery can
be completed within 30 minutes. Application studies on renal and carotid artery systems will be presented respectively demonstrating the
reliability and applicability of the computational tool. Meanwhile, medical favorite results on flow pattern, pressure, and WSS will be shown.
Although validation and demonstration have been focused on computational hemodynamics in cardiovascular system, the computation
methodology is applicable to airflow, urinary flow, and peristalsis in other human systems as well.
Bio:
Dr. Huidan (Whitney) Yu is currently an Assistant Professor in Mechanical Engineering Department of IUPUI. Prior to
this position, she successively completed two PhD programs in Physics at Peking University, China, in 2001 and in Aerospace
Engineering at Texas A&M University, USA, in 2004, followed by two postdoctoral research positions at Los Alamos National
laboratory and Johns Hopkins University. Dr. Yu’s research field is in computational fluid dynamics in general with expertise on
kinetic theory based lattice Boltzmann method to model and simulate complex flows involving turbulence. Dr. Yu has extensive
research experience in dealing with a variety of fluid and thermal problems applications in energy, biomechanics, atmosphere,
etc. She has published more than 60 articles in peer-reviewed journals. Her research interest is to synergistically combine the
novel mesoscale computational paradigm with the emerging Graphic Processing Unit (GPU) technology to develop efficient and
unique approaches for computing multi-scale multi-physics flow-structure interaction problems with or without turbulence. The
current focuses are on computational methodologies for patient-specific hemodynamics in which fluid-structure interactions,
non-Newtonian effects, and pulsatility with the presence of turbulence are considered and for bubble dynamics in reactive microflow. Her research projects are supported by NSF, IU Health, Boston Scientific, and OVCR of IUPUI.
Date: November 20, 2015
Place: CB 106
Time: 3:00 to 4:00p
Contact: Dr. Alexandre Martin 257-4462
Meet the speaker and have refreshments
Attendance open to all interested persons
DEPARTMENT OF MECHANICAL ENGINEERING
UNIVERSITY OF KENTUCKY
LEXINGTON, KENTUCKY