Gold University of Minnesota M. Skip to main content. University of Minnesota.
Driven to Discover.
DMD

Computational Modeling & Regulatory Science 1:
Patient-specific Simulation for Treatment Planning and Clinical Decision-Making

Tuesday, April 10, 10:30-12:00
Ski-U-Mah, McNamara Alumni Center

Organizers: Marc Horner, Technical Lead, Healthcare, ANSYS, Inc.
Dawn Bardot, Director, Healthcare Innovation, Medtronic, Inc.

"An In Silico Investigation of a Lobe-Specific Targeted Pulmonary Drug Delivery Method"
Yu Feng, Assistant Professor, School of Chemical Engineering, Oklahoma State University

"Digital Approach to Customized Biomedical Solutions: Customized Smart Braces and Prosthetics"
Fluvio Lobo Fenoglietto, Research and Development Engineer, Prototype Development and 3D Print Lab (PD3D), Institute for Simulation and Training, University of Central Florida

"High Resolution Simulation of Diastolic Left Ventricular Hemodynamics Guided by Four-dimensional Flow Magnetic Resonance Imaging Data"
Trung Bao Le, Research Scientist, Medical College of Wisconsin

"Towards Democratization of Clinical Diagnostics Tool Development for Hemodynamic Simulation"
Dimitris Mitsouras, Associate Professor, Faculty of Medicine, University of Ottawa


Session Abstract:

Advances in medical imaging and numerical simulation methods are enabling the development of patient-specific anatomic and physiologic computer models at a rapid pace. These models are playing an ever-increasing role in personalized diagnostics and treatment planning. Attendees will learn about progress in the development of engineering simulation solutions that utilize medical image data and other patient-specific data to generate patient-specific diagnostic information, supply clinical decision-support, and optimize the outcome of an intervention for a specific patient.


Session Organizer Bios:

Marc Horner, Technical Lead, Healthcare, ANSYS, Inc.
Dr. Marc Horner is the lead healthcare specialist at ANSYS Inc. Marc joined ANSYS after earning his PhD in Chemical Engineering from Northwestern University in 2001. Marc began by providing support and professional services for biomedical clients, primarily in the areas of cardiovascular devices, drug delivery, packaging, microfluidics and orthopedics. During this time, Marc developed numerous modeling approaches that can be used to establish the efficacy and safety of medical devices. Marc now helps coordinate business and technology development for the health care sector in North America.

Dawn Bardot, Director, Healthcare Innovation, Medtronic, Inc.
Dr. Bardot has more than 15 years of experience in computational model validation and uncertainty quantification. She is passionate about the application of modeling and simulation to improve health care and lower the cost of bringing products to market. She is an active member of the American Society of Mechanical Engineers (ASME) and the Committee on Verification and Validation in Computational Methods for Medical Devices. Dr. Bardot has a BS and MS in mechanical engineering from Kansas State University, a PhD in mechanical engineering from the University of Washington, was an Innovation Fellow at the University of Minnesota's Earl E. Bakken Medical Devices Center, and spent two summers as a Faculty Fellow at NASA Marshal Space Flight Center.


Speaker Bios:

Yu Feng, Assistant Professor, School of Chemical Engineering, Oklahoma State University
Dr. Yu Feng is an assistant professor in the School of Chemical Engineering at Oklahoma State University, and a center investigator in the Oklahoma Center for Respiratory and Infectious Diseases (OCRID). Yu Feng was a Research Assistant Professor and Lab Manager of the Computational Multi-Physics Laboratory (CM-PL) at North Carolina State University. He has also held an affiliation with the DoD Biotechnology HPC Software Applications Institute (BHSAI) as a Research Scientist II. He completed his B.S. in Engineering Mechanics in 2007 from the School of Aeronautics and Astronautics, Zhejiang University, Hangzhou, China. He then joined the Department of Mechanical and Aerospace Engineering at North Carolina State University and obtained his M.S. and Ph.D. degrees in 2010 and 2013, respectively. He founded the Computational Biofluidics and Biomechanics Laboratory (CBBL) at Oklahoma State University, focusing on developing and applying advanced CFPD models towards multiple applications associated with pulmonary healthcare. He has over 10 years experience modeling lung aerosol dynamics on ANSYS CFX and Fluent platforms, with 25 publications in top-ranked fluid dynamics and aerosol science journals. The overall goal of his research group is to understand and consider more underlying physics and chemistry and provide non-invasive, cost-effective and accurate numerical tools for multiple biomedical applications, e.g., health risk assessment, pulmonary drug evaluation and improvement, and non-invasive disease diagnosis. We make contributions to the medical world and human life by providing well-posed solutions to patient-specific pulmonary health problems using multi-scale modeling techniques. Outside of work, Yu Feng enjoys running (17 half marathons and 6 marathons so far), hiking, singing performance, photography, basketball, playing accordion, and cooking.

Fluvio Lobo Fenoglietto, Research and Development Engineer, Prototype Development and 3D Print Lab (PD3D), Institute for Simulation and Training, University of Central Florida
Fluvio is a Research and Development Engineer for the Prototype Development and 3D Print Lab (PD3D) at the Institute for Simulation and Training (IST) within the University of Central Florida (UCF). Fluvio leads the design and development of augmented medical devices and systems capable of assessing user performance. These systems include integrated sensors, microcontrollers, software and electronic development. Trained as a Biomedical Engineer, Fluvio holds an MS degree in the field of Tissue Mechanics. Fluvio was instrumental in the development of the Tissue Characterization Lab at the University of Minnesota, which focused on the characterization of biological tissues’ mechanical behavior.

Trung Bao Le, Research Scientist, Medical College of Wisconsin
Trung Bao Le is a Research Scientist at the Medical College of Wisconsin. His research focuses on fundamental phenomena in fluid-structure interaction problems at a variety of scales in biomedical engineering. His techniques involve the development for scalable numerical algorithms that can run from desktop computer to supercomputers. His current effort focuses on translating simulation technology into clinical practice.

Dimitris Mitsouras, Associate Professor, Faculty of Medicine, University of Ottawa
Dr. Mitsouras is Associate Professor in the Faculty of Medicine of the University of Ottawa. Dr. Mitsouras co-founded and later directed the internationally-recognized Applied Imaging Science Lab at Harvard Medical School and Brigham and Women’s Hospital whose research focuses on developing novel biomedical imaging technologies that empower physicians to better detect, quantify, and visualize disease toward improving the quality of life of their patients via computational and algorithmic techniques. He has formed and led collaborative multi-disciplinary teams that have developed diverse methods now incorporated in clinical magnetic resonance and computed tomography systems. Dr. Mitsouras’ lab is currently working to (a) establish optimal computational physiologic models to non-invasively assess the hemodynamic significance of coronary artery disease and predict the risk of plaque rupture which can lead to myocardial infarction and death, and (b) accelerate personalized precision medicine via 3D virtual and printed models to enable optimal therapy planning and delivery for cardiac, vascular, oncologic, and orthopedic surgery patients.


Presentation Abstracts:

"An In Silico Investigation of a Lobe-Specific Targeted Pulmonary Drug Delivery Method"
Nowadays, “personalized medicine” is starting to replace current “one size fits all” treatment approaches. The goal is to have the right drug with the right dose for the right patient at the right time and location. Indeed, conventional pulmonary drug delivery devices still have poor efficiencies (<25%) for delivering drugs to lung tumor sites. Major portions of the aggressive medicine deposit on healthy tissue, which causes severe side effects and induces extra health care expenses. Therefore, a new targeted pulmonary drug delivery method is proposed and evaluated using the Computational Fluid-Particle Dynamics (CFPD) method to achieve lobe-specific delivery. By controlling the release position and velocity of the drug particles at the mouth inlet, drug deposition efficiency (DE) in a designated lobe can be increased up to 90%. Intersubject variability has also been investigated using this noninvasive in silico tool. Results indicate that the glottis constriction ratio is a key factor that influences the effectiveness of the proposed targeted drug delivery method. Although lobe-specific pulmonary drug delivery can be realized, the actuation flow rate must be lower than 2 L/min, and the glottis constriction ratio has a significant impact on the effectiveness of the targeting method. Also, a design idea using e-cigarette as the prototype is proposed as the next-generation inhaler to accommodate operational flexibility restrictions.

"Digital Approach to Customized Biomedical Solutions: Customized Smart Braces and Prosthetics"
Patient’s external anatomy, surface features and color can be mapped using handheld 3D optical scanners. The resulting scans can be used for the design of custom orthotics or prosthetics for limbs or residual limbs. Designs can be further optimized through the implementation of engineering simulation methods and topology optimization software. 3D printers are then used for rapid fabrication of the resulting tailor-made orthotics or prosthetics. The entire process requires a combination of software and expertise from the fields of medicine, engineering and graphic arts and design. The goal of this presentation is to guide the audience along the prototype development and 3D printing (PD3D) process for the design of customized smart braces and prosthetics.

"High Resolution Simulation of Diastolic Left Ventricular Hemodynamics Guided by Four-dimensional Flow Magnetic Resonance Imaging Data"
We investigate the diastolic hemodynamics in a patient-specific left ventricle (LV) of a healthy subject using four-dimensional flow magnetic resonance imaging (4D-Flow MRI) measurement data and numerical simulation. From four-dimensional cardiac magnetic resonance (CMR) imaging data, the kinematics of the endocardium is reconstructed. The endocardial kinematics and the time varying velocity distribution from 4D-Flow MRI at the mitral orifice are prescribed as boundary conditions for the numerical simulation. Both 4D-Flow MRI data and numerical results show the classical formation of the mitral vortex ring (MVR) during E-wave filling. The in−vivo data reveals that a large three-dimensional vortex structure forms near in the mid-level region of LV during diastasis (mid-level vortex). This mid-level vortex is formed simultaneously with the MVR and has not been reported in the literature. Quantitative comparison shows that the computed kinetic energy (KE) evolves in a similar manner to one derived from 4D-Flow MRI data during early E-wave filling.

We acknowledge the support of computational time from the Minnesota Supercomputing Institute (University of Minnesota) and the Institute for Advanced Computational Science (Stony Brook University). We also acknowledge the support for the computational time from the Argonne Leadership Computing Facility, which is a DOE Office of Science User Facility supported under Contract DE-AC02-06CH11357

"Towards Democratization of Clinical Diagnostics Tool Development for Hemodynamic Simulation"
Advances in medical imaging and computational modeling technologies are enabling the development of patient-specific models of various biological transport processes. Recently, results from such models gained approval for clinical diagnostic use in e.g., coronary artery disease, with risk evaluation and clinical treatment planning/optimization approaches now also emerging. Barriers to adoption of physics-based diagnosis and prognosis tools include a lack of expertise in applying image segmentation and numerical simulation techniques, time required to provide a result, and determining patient-specific simulation parameters from the available clinical data. This presentation will review our progress addressing these barriers through the development of a simulation platform that automates many of the steps associated with modeling blood flow in stenotic coronary arteries for guiding appropriate percutaneous coronary intervention (PCI).

The presence of a coronary stenosis, or narrowing, may or may not affect flow rates and pressure. While a coronary angiogram can be used to identify the presence of a stenosis, the resulting impact on the hemodynamics is currently measured invasively using a pressure wire while the patient is in a pharmacologically-induced stress state. The pressure loss measured across each stenosis normalized to mean aortic pressure, called the fractional flow reserve (FFR) is the reference standard to determine which patients will benefit from PCI (FFR<80%).

Computational fluid dynamics is now approved by the US Food and Drug Administration to perform patient-specific FFR prediction from non-invasive coronary angiography imaging. Imaging yields the vascular geometry and computational fluid dynamics yields pressure for the given hemodynamics conditions in that geometry. Developing patient specific hemodynamic input conditions is the major challenge. A key feature of the hemodynamic simulations described herein is the incorporation of the distal vasculature using lumped parameter outlet boundary conditions. The input parameters of the outlet condition are estimated from the medical image data and known features of coronary physiology. The combination of the segmented image data and image-based parameters results in a fully personalized three-dimensional analysis of blood flow in the patients’ coronary vasculature. This talk will review progress towards an automated computational fluid dynamics platform that significantly decreases the learning curve associated with deploying clinical engineering simulation tools such as noninvasive FFR. Specifically, the framework has reduced the set-up, solution, and post-processing steps of an FFR analysis to four simple steps. This framework is continuing to be developed to provide a non-invasive assessment of a patient’s disease state.


Related Sessions:

Computational Modeling & Regulatory Science 2
Computational Modeling & Regulatory Science 3

Register Now


Stay connected with the DMD Conference through LinkedIn Twitter Facebook Icon G+ YouTube