Developing a more scientific way to determine the load threshold for capillary wall failure would be a big step forward in characterizing whether bruising is result from an abuse or an accident. In this thesis, the upper portion of the human arm was modeled and analyzed under dynamic loading conditions. Since the diameter of the arm is much larger than that of the capillary, a four-level hierarchical sub-modeling method was used to mathematically link the transient response of the global arm model to the response of a small volume in the muscle tissue containing one capillary. Soft tissue in the arm was modeled in two distinct ways. In one method each component of soft tissue was modeled used isotropic linear elastic properties to find the loading threshold that produces a hoop stress in the capillary wall equal to the capillary failure stress. In the other approach, nonlinear, hyper-elastic properties for skin, adipose, muscle tissue and capillary wall were employed to make the tissue behavior more realistic to that of a human arm. Material-appropriate constitutive functions were chosen for each layer. A mathematical technique implement in MATLAB was used to estimate and subtract rigid body motion from the total displacement to avoid excessive displacements of sub-models and focus more on the deformation-only displacement. It was found that modeling the skin, adipose, muscle and capillary as hyper-elastic resulted in significantly smaller deformations but larger loads that resulted in capillary failure.

<p>Traumatic brain injury (TBI) is characterized by a violent or sudden blow to the head that causes tearing or bruising of the brain tissue and its supporting blood vessels. Determination of the mechanical properties of gray and white matter is critical for the creation of computational models of healthy and TBI-damaged brain tissues. Current in vivo methods to characterize brain tissue, such as 3D amplified MRI (aMRI) and magnetic resonance elastography (MRE), are highly vulnerable to motion artifacts and have limited techniques to exert mechanical loads on the brain. Therefore, in vitro testing was employed to estimate the chemical composition of gray and white matter using Fourier Transform Infrared (FTIR) spectroscopy and the stress responses of the brain tissues to high compressive deformations via unconfined compression. Attenuated total reflectance (ATR) was run in conjunction with FTIR spectroscopy to eliminate the need for sample preparation. Unconfined compression of gray and white matter samples was performed to 70% of the total sample height at a constant strain rate of 0.35/s. Results showed significant increases in the absorbances of white matter (<em>p</em> < 0.05) in the characteristic lipid and carbohydrate regions of the FTIR spectra when compared to gray matter. Within the initial 10% toe-region of the stress-strain curve, white matter is observed to absorb significantly greater compressive loads (<em>p </em>< 0.05) than gray matter. These results indicate an incomplete characterization of brain tissue; therefore, additional in vitro and in vivo methods are still necessary, separately or in combination, to accurately characterize brain tissue mechanics in TBI and non-TBI patients.</p>

No description available.

No description available.

No description available.

This thesis developed a novel theory of fiber jamming and experimentally verified it. The theory relates the performance, which is the ratio between the stiff and soft states of a fiber jamming chamber, to three relative design parameters: the ratio of the wall thickness to the membrane inner diameter, the ratio of the fiber diameter to membrane inner diameter, and the number of fibers. These three parameters, when held constant across different chamber sizes, hold the performance constant. To test the theory, three different types of fiber jamming chambers were built in three different sizes. Each chamber was set up as a cantilever beam and deflected 10mm in both the un-jammed (soft) and jammed (stiff) states. When the three design parameters were held constant, the performance of the chamber was consistent within 10\%. In contrast, when the parameters were altered, there was a statistically significant $p < .0001$ and noticeable effect on chamber performance. These two results can be used in tandem to design miniaturized fiber jamming chambers. These results also have a direct application in soft robots designed for minimally invasive surgery.

The objective of this thesis is to develop design approaches and models for prosthetic ankle joints using kinematic models of the human ankle and compliant mechanisms technology. Compliant mechanisms offer several potential design advantages over traditional rigid-body designs including high reliability and low cost. These design advantages are ideal for use in prosthetics. Some prosthetic ankle/foot systems currently on the market have multiple degrees of freedom yet are expensive. Additionally, even though these systems have multiple degrees of freedom, none of them are designed after the actual movements of the biological ankle. In this thesis a two, single degree-of-freedom hinge joint model, which is a kinematic model based on the biological ankle during walking, is used to develop compliant prosthetic ankle joints. The use of the model together with compliant mechanisms may provide the ability to develop highly functional prosthetic ankle joints at a lower cost than current high-performance prosthetic systems. Finally, a design approach for ankles may facilitate future development for knees, hips or other biological joints.

This dissertation presents a method for the design of customizable products that interface with the human body. The method presented involves first, a consistent method of capturing and representing the human model so that the model can be used with CAx tools and solid modeling techniques. Second, it provides a design methodology based on feature structure planning and assembly modeling that provides a consistent structure to the design process so that it can be reused and parameterized. Third, a strategy for identifying parametric variables that are referenced to the human body is introduced. The core of this method is the definition of biomechanical products as an assembly model, where human data is defined as the base part. This research expands on traditional mating conditions in assembly model methods by identifying different ways products can interface with the human body. With the identification of these mating conditions, products can be designed to interact with the body in definable ways through the definition of parametric strategies. This dissertation also presents the necessary theoretical and numerical methods for implementation of these mating conditions in a CAD system.

Developmental dysplasia of the hip (DDH) is a common newborn condition where the femoral head is not located in its natural position in the acetabulum (hip socket). Several treatment methods are being implemented worldwide to treat this abnormal condition. One of the most effective methods of treatment is the use of Pavlik Harness, which directs the femoral head toward the natural position inside the acetabulum. This dissertation presents a developed method for identifying the least energy path that the femoral head would follow during reduction. This is achieved by utilizing a validated computational biomechanical model that allows the determination of the potential energy, and then implementing the principle of stationary potential energy. The potential energy stems from strain energy stored in the muscles and gravitational potential energy of four rigid-body components of lower limb bones. Five muscles are identified and modeled because of their effect on DDH reduction. Clinical observations indicate that reduction with the Pavlik Harness occurs passively in deep sleep under the combined effects of gravity and the constraints of the Pavlik Harness. A non-linear constitutive equation, describing the passive muscle response, is used in the potential energy computation. Different DDH abnormalities with various flexion, abduction, and hip rotation angles are considered, and least energy paths are identified. Several constraints, such as geometry and harness configuration, are considered to closely simulate real cases of DDH. Results confirm the clinical observations of two different pathways for closed reduction. The path of least energy closely approximated the modified Hoffman-Daimler method. Release of the pectineus muscle favored a more direct pathway over the posterior rim of the acetabulum. The direct path over the posterior rim of the acetabulum requires more energy. This model supports the observation that Grade IV dislocations may require manual reduction by the direct path. However, the indirect path requires less energy and may be an alternative to direct manual reduction of Grade IV infantile hip dislocations. Of great importance, as a result of this work, identifying the minimum energy path that the femoral head would travel would provide a non-surgical tool that effectively aids the surgeon in treating DDH.

The objectives of this study were to 1) calculate participant-specific segment inertial parameters using dual energy X-ray absorptiometry (DXA) data (referred to as full DXA-driven parameters) and compare the pitching arm kinetic predictions using full DXA-driven inverse dynamics vs scaled, DXA mass-driven (using DXA masses but scaled centers of mass and radii of gyration), and DXA scaled inverse dynamics(ID) (using the full DXA-driven inertial parameters averaged across all participants), 2) examine associations between full DXA-driven kinetics and body mass index (BMI) and 3) examine associations between full DXA-driven kinetics and segment mass index (SMI). Eighteen 10- to 11- year-olds pitched 10 fastballs. DXA scans were conducted and examined to obtain 3D inertial parameters of the upper arm, forearm, and hand. Full DXA-driven and scaled inertial parameters were compared using paired t-tests. Pitching arm kinetic predictions calculated with the four methods (i.e. scaled ID, DXA mass-driven ID, full DXA-driven ID, and DXA scaled ID) were compared using a repeated measures ANOVA with Tukey post-hoc tests. The major results were that 1) full DXA-driven participant specific inertial parameters differed from scaled inertial parameters 2) kinetic predictions significantly varied by method and 3) full DXA-driven ID predictions for shoulder compression force and shoulder internal rotation torque were significantly associated with BMI and/or SMI.