Horizon Information Portal

MSOE Library Home

MSOE Library Home

Home

Modeling and analysis of human airway dynamics utilizing data obtained from a cine MRI : a report submitted to the faculty of the Milwaukee School of Engineering in partial fulfillment of the requirements for the degree of Master of Science in Engineering / by Wesley A. Zloza.

by Zloza, Wesley A., author., Kumpaty, Dr. Subha , Elluru, Dr. Ravi, thesis co-advisor , Kalra, Dr. Maninder, thesis co-advisor , Shimek, Gary , Milwaukee School of Engineering

[Milwaukee School of Engineering], [2017]

Airway (Medicine)

Respiratory organs -- Obstructions.

Computer simulation

Magnetic resonance imaging

142 leaves : illustrations, some of which are in color ; 29 cm

Introduction -- Project description -- Review of literature -- Background -- Model generation, refinement, and simulation -- Conclusions -- Completed tasks and milestones -- Appendix A: IRB certification -- B: Medical illustrations -- C: Additional CFD analysis.

The objective of this project was to determine the validity of using MRI (magnetic resonance imaging) data to identify abnormalities in human airways and to analyze the internal flow behavior, which is conducted by means of fluid flow analysis. CT (computed tomography) scans have long been used to develop three-dimensional (3D) models of internal anatomy because of their fine resolution and contrast between tissue densities. The negative aspect of these scans is that they expose patients to medical risks because of the radiation that is emitted when capturing the images. Because of these risks, MRIs are considered to be a safer imaging option; however, their image quality isn't as fine.

The original goals of this project were to employ static (standard) and dynamic (cine) MRI images from pediatric patients to develop accurate models of their airways, and to use the models to perform computational fluid dynamics (CFD) simulations. CFD simulations results would then be used to identify areas of large pressure differentials and air recirculation, which could indicate areas of concern. These results would then aid medical personnel in determining which medical procedures could be performed to minimize irregular or problematic air flow. Because of time constraints and challenges associated with the MRI images, the project goals were refined. Standard and cine MRI data from a single patient, and CT data from an additional patient, were employed in an attempt to establish a technique for developing accurate 3D fluid domain models from MRI data.

This report explains the work performed during the two-phase project. The first phase of the project focused on the generation of 3D models from cine MRI data, and the second phase focused on use of the models in CFD simulation. To generate models from standard MRI data in their original DICOM format, a complicated series of data manipulation techniques was necessary. Using 3D Slicer, along with thresholding and image segmentation, a 3D mesh model was extracted and created. The mesh model was refined and converted to a STL format with a number of tools, including, Blender, Meshmixed, and Remake. Fusion 360 was employed to convert this model into a solid model needed for simulation. The solid model was imported into ANSYS Fluent, where the mesh was optimized for simulation, and laminar and turbulent flow simulations were successfully conducted.

An attempt was made to employ cine MRI data to analyze the changing geometry of the airway. Difficulties were encountered in working with the cine MRI data, because they had to be analyzed manually. A recommendation for future research is the development of a software program that can automate the analysis of cine MRI data. CFD simulations of the MRI data revealed recirculation within the airway, but no other areas of concern were discerned. The project was successful in verifying that 3D models suitable for simulation can be extracted from MRI data.

https://msoe.tind.io/record/914

Horizon Information Portal 3.25_9885

Powered by SirsiDynix

© 2001-2013 SirsiDynix All rights reserved.

Horizon Information Portal