1 Copyright © 20xx by ASME
ASME 2010 First Global Conference on NanoEngineering for Medicine and Biology
NEMB2010
Feb. 7-10, 2010, Houston, TX, USA
NEMB2010-13098
MRI BASED FE MODEL OF HEAD TRAUMA: IMAGING AND MESH GENERATION
Marcus Slavenas
University of Illinois
Urbana-Champaign, IL, USA
Martin Ostoja-Starzewski
University of Illinois
Urbana-Champaign, IL, USA
ABSTRACT
In order to perform simulations of the human head under
impact loading, a finite element (FE) mesh was developed from
a structural magnetic resonance image set. A T1 and a T2
weighted image were acquired from a single subject and co-
registered. FSL was used to segment the image according to
different tissue types: scalp, skull, cerebral spinal fluid (CSF),
grey matter, white matter, or non-object. A custom C++
program concatenates the image files, removes all non-object
voxels, and creates a FE mesh based upon the FSL tissue
identification and the image geometry. We present a highly
refined FE mesh of the human head with scalp thickness, skull
thickness, and white and grey matter distribution that are
consistent with anatomical norms. The mesh is exported to
ABAQUS and used to simulate a frontal impact by Y. Chen
and M. Ostaja-Starzewski [1].
INTRODUCTION
Traumatic Brain Injury (TBI) is a serious medical problem
from auto accidents, sports, and combat. A better understanding
of the mechanics of an impact to the head leads to better design
of protective headgear and improved diagnostic abilities of
trauma physicians. Modern imaging technologies like CT
(computed tomography) and structural MRI (magnetic
resonance imaging) have provided clinicians the ability to look
into the living brain and rapidly diagnose damage or
abnormalities. Functional imaging technologies like SPECT
(single photon emission computed tomography) and fMRI
(functional MRI) improved understanding of brain function.
But, despite the great advances in brain science, the mechanical
models of traumatic brain injury (TBI) remain immature. We
introduce a highly refined finite element (FE) mesh based upon
structural MRI data.
RESEARCH APPROACH
Imaging
Each subject was scanned using the 3T Siemens Alegra
Magnetom Scanner at the Biomedical Imaging Center at the
University of Illinois, Urbana-Champaign. Two structural
imaging sequences, MPRAGE and T2-VFL, provide high-
resolution images with two different contrast weightings, T1
and T2, respectively. T1 signal is produced by the relaxation of
nuclei from an energized state back to equilibrium, also called
spin-lattice relaxation because nuclei lose energy to the object
lattice. T2 signal is produced by the dephasing of adjacent
nuclear spins, also called spin-spin relaxation because nuclei
exchange energy with each other. The contrast is selected by
changing the timing of the applied pulse sequence. The T1 and
T2 images have resolutions of 1.33×1.33×1.30mm and
0.50×0.50×1.00mm (Figure 1).
a) b)
c) d)
Figure 1) a) A T1 weighted image, sagittal view b) A T2 image of the
same subject c,d) Close-ups of the same region from the two images
show quite different output