<p>Between October 2001 and May 2012 approximately 70% of U.S. military personnel killed in action and 75% wounded in action were the direct result of exposure to an explosion. As of 2008, it was estimated that close to 20% of all Operation Iraqi Freedom and Operation Enduring Freedom (OIF/OEF) veterans had sustained some form of traumatic brain injury (TBI). Further, blast exposure is also a civilian problem due to the increased usage of explosives in terrorist attacks. Blast injury research has historically focused on the pulmonary system and the other air-containing organs which have been shown through extensive experimentation to be susceptible to blast overpressure injury. A shift in injury pattern during recent conflicts is characterized by decreased incidence of pulmonary injuries with an increase in TBI thought to be associated with blast exposure. This increase in observation of blast TBI has resulted in a large research effort to understand mechanisms and thresholds. However, due to the relatively sudden shift, much of this research is being conducted without a proper understanding and consideration of blast mechanics and interspecies scaling effects.</p><p>This dissertation used experimental and computational finite element (FE) analysis to investigate some large questions surrounding blast TBI research. An experimental investigation was conducted to determine the effects of modern thoracic body armor usage on blast pressure exposure seen by the body. To improve FE modeling capabilities, brain tissue mechanics in common blast TBI animal model species were investigated experimentally and computationally to determine viscoelastic constitutive behavior and measure interspecies variation. Meta-analysis of blast pulmonary literature was conducted to update interspecies scaling and injury risk models. To derive interspecies scaling and injury risk models for blast neurotrauma endpoints a meta-analysis of existing experimental data was used.</p><p>This dissertation makes major contributions to the field of injury biomechanics and blast injury research. Research presented in this dissertation showed that modern thoracic body armor has the ability to lower the risk of pulmonary injury from blast exposure by attenuating and altering blast overpressure. The study shows that the use of soft body armor results in the pulmonary injury threshold being similar to that for neurotrauma. The use of hard body armor results in the threshold for pulmonary injury occurring at higher levels than that of neurotrauma. This finding is important, as it helps to explain the recent shift in injury types observed and highlights the importance of continued widespread usage of body armor not only for ballistic protection but for protection from blast as well.</p><p>This dissertation also shows the importance of interspecies scaling for investigation of blast neurotrauma. This work looks at existing in vivo animal model data to derive appropriate scaling across a wide range of brain size. Appropriate scaling for apnea occurrence and fatality for blast isolated to the head was found to be approximately equal to a characteristic length scaling of brain size, assuming similar brain geometry. By combining the interspecies scaling developed and existing tests data, injury risk models were derived for short duration blast exposures.</p><p>The contributions and conclusions of this dissertation serve to inform the injury biomechanics field and to improve future research efforts. The consideration by researchers of the recommendations presented in this dissertation for in vivo animal model testing will serve to maximize the value gained from experimentation and improve our understanding of blast injury mechanisms and thresholds. The injury risk models presented in this work help to improve our ability to prevent, diagnose, and treat blast neurotrauma.</p> / Dissertation

Several efforts have been made in the past half century to characterize the behavior of brain tissue under different modes of loading and deformation rates; however each developed model has been associated with limitations. This dissertation aims at addressing the non-linear and rate dependent behavior of brain tissue specially in high strain rates (above 100 s-1) that represents the loading conditions occurring in blast induced neurotrauma (BINT) and development of a universal constitutive model for brain tissue that describes the tissue mechanical behavior from medium to high loading rates.. In order to evaluate the nature of nonlinearity of brain tissue, bovine brain samples (n=30) were tested under shear stress-relaxation loading with medium strain rate of 10 s-1 at strain levels ranging from 2% to 40% and the isochronous stress strain curves at 0,1 s and 10 s after the peak force formed. This approach enabled verification of the applicability of the quasilinear viscoelastic (QLV) theory to brain tissue and derivation of its elastic function based on the physics of the material rather than relying solely on curve fitting. The results confirmed that the QLV theory is an acceptable approximation for engineering shear strain levels below 40% that is beyond the level of axonal injury and the shape of the instantaneous elastic response was determined to be a 5th order odd polynomial with instantaneous linear shear modulus of 3.48±0.18 kPa. To investigate the rate dependent behavior of brain tissue at high strain rates, a novel experimental setup was developed and bovine brain samples (n=25) were tested at strain rates of 90, 120, 500, 600 and 800 s-1 and the resulting deformation and shear force were recorded. The stress-strain relationships showed significant rate dependency at high rates and was characterized using a QLV model with a 739 s-1 decay rate and validated with finite element analysis. The results showed the brain instantaneous elastic response can be modeled with a 3rd order odd polynomial and the instantaneous linear shear modulus was 19.2±1.1 kPa. A universal constitutive model was developed by combining the models developed for medium and high rate deformations and based on the QLV theory, in which the relaxation function has 5 time constants for 5 orders of magnitude in time (from 1 ms to 10 s) and therefore, is capable of predicting the brain tissue behavior in a wide range of deformation rates. Although the universal model presented in this study was developed based on only shear tests and the material parameters could not be found uniquely, by comparing the results of this study with previously available data in the literature under tension unique material parameters were determined for a 5 parameter generalized Rivlin elastic function (C10=3.208±0.602 kPa, C01=4.191±1.074 kPa, C11=79.898±18.974 kPa, C20=-37.093±7.273 kPa, C02=-37.712±5.678 kPa). The universal constitutive model for brain tissue presented in this dissertation is capable of characterizing the brain tissue behavior under large deformation in a wide range of strain rates and can be used in computational modeling of Traumatic Brain Injury (TBI) to predict injuries that result from falls and sports to automotive accidents and BINT. / Mechanical Engineering