Over the past decade, the tissue engineering paradigm has gained attention as a potential means to restore native tissue functionality. Although attractive, the wide variety of scaffold materials, cell sources, and mechanical conditioning regimes coupled with the paucity of structurally-based, finite deformation framework constitutive models found in the literature hinders the elucidation of extracellular matrix (ECM) formation and remodeling in engineered tissues. Therefore, the overall objective of this work is to develop structurally guided generalized finite deformation based constitutive models than can be used to gain an understanding of tissue formation and remodeling in tissue engineering applications. Further, it is the intent of this work to apply such an approach to investigate tissue formation and remodeling in tissue engineered pulmonary arteries. In the first part of this work, a novel technique for acquiring and quantifying high resolution three dimensional structural data was used on bone-marrow stem cell-seeded polymeric scaffold composites, and it was shown that the continuous anisotropic scaffold phase transitioned to a highly discontinuous isotropic scaffold phase after twelve weeks in vivo. Next, structural constitutive models were developed based on the scaffold continuity. For continuous scaffold composites, scaffold-ECM interactions were included in the model as extensional and shearing terms, while it was shown that such effects were negligible in the discontinuous scaffold composites. A parameter estimation and model validation procedure was described using a tunable tissue-analog system of polyacrylamide (PAM) gel. It was found that the scaffold-ECM interaction due to fiber extension was highly non-linear, showing a reinforcing effect larger than from rule of mixtures predictions. Experimental validation with PAM gel supported the models. Finally, both models were used to investigate tissue formation and remodeling in in vivo engineered pulmonary arteries. At early timepoints (7 days), little change in ECM mechanical properties was observed. In later timepoints (42 to 140 days), the collagen effective modulus and collagen recruitment parameters changed substantially, suggesting collagen maturation via increased cross-linking and crimp organization. Ultimately, a methodical approach to understanding tissue formation and remodeling via structural constitutive models was presented and successfully applied to a clinically-relevant tissue engineering system.

Stem cells are highly valued for their capacity to aid in the functional recovery of damaged or diseased tissue. They are defined by their remarkable ability to maintain their undifferentiated state through countless cycles of cell division and to differentiate into variable types of specialized cells. Since ethical controversy has hindered funding for embryonic stem cell research and induced pluripotent stem cells are in the initial stages of investigations, much research has been conducted using adult stem cells. The use of adult stem cells in clinical applications is gradually becoming a reality; however, the major limitation is the difficulty to isolate, purify and expand them in culture. Matrix metalloproteinases (MMPs) have been regarded as a group of zinc-endopeptidases that influence tissue remodeling by degrading constituents of the extracellular matrix to actively promote cell proliferation, migration, apoptosis and differentiation. They have been suggested to play important roles in the regeneration of amputated newt limbs by contributing to a population of undifferentiated stem cells, called a blastema, which is likely formed by cell dedifferentiation. The research presented here builds on previous work investigating the therapeutic use of MMP1. Investigations have demonstrated the ability of MMP1 to aid in the recovery of skeletal muscle tissue by degrading fibrous scar tissue to facilitate cell migration and differentiation. This work examines the potential of MMP1 in skeletal muscle healing to stimulate stem cell behavior by the expression of certain muscle stem cell markers and its impact on cell differentiation. In addition, stem cells derived from skeletal muscle tissue were investigated to thoroughly elucidate the effect of blocking MMP signaling. MMP inhibition using GM6001 was observed to negatively impact muscle stem cell migration, stem cell associated markers and their differentiation capacity thus indicating the key role of MMPs in muscle stem cell behavior.

Extracorporeal blood purification is a promising therapeutic modality for sepsis, a potentially fatal, dysfunctional immune disorder caused by infection. During sepsis, dysregulation of the innate immune system leads to excessive release of inflammatory mediators known as cytokines into the bloodstream. Removal of cytokines from the circulating blood may attenuate hyper-inflammatory signaling and promote immunologic homeostasis. We are developing an extracorporeal blood purification device to remove cytokines from the blood using biocompatible, porous, polymeric beads. Hemoadsorption therapy using our device has demonstrated improved survival in a murine sepsis model, and may serve as a novel adjuvant therapy to improve patient outcomes in the setting of severe sepsis and septic shock. We developed a mathematical model to characterize cytokine adsorption dynamics within the device, and used confocal laser scanning microscopy (CLSM) to quantify cytokine transport within single sorbent beads. Finite element modeling was utilized to estimate model parameters based on best fits to CLSM data, and the fitted model was used to simulate cytokine adsorption behavior under clinically relevant conditions. We investigated intraparticle cytokine transport under competitive and non-competitive adsorption conditions, and demonstrated that effects due to coadsorption of serum solutes are likely negligible under physiologic cytokine concentrations. CLSM results indicate that less than 20% of available sorbent surface area is utilized for cytokine adsorption. Tumor necrosis factor (TNF) is a pleiotropic, pro-inflammatory cytokine, and serves as a primary initiator of systemic inflammation during sepsis. Removal of TNF within the device is slow, putatively due to hindered diffusion of the large TNF molecule (51kD) within the sorbent pores. We induced deoligomerization of trimeric TNF into its monomeric subunits, and demonstrated significantly accelerated capture of monomerized TNF within the device, compared to native TNF. We investigated small molecules capable of facilitating TNF deoligomerization, and proposed techniques to immobilize such molecules on the sorbent surface. Functionalized sorbent beads capable of locally dissociating TNF at the bead surface may significantly accelerate capture of TNF from the circulating blood. This concept could be expanded to enhance capture of oligomeric biomolecules using size exclusion filtration materials for a variety of disease states.

Background: Bleeding and thromboembolism continue to be the greatest cause of morbidity and mortality during extracorporeal membrane oxygenation (ECMO). The activated clotting time (ACT) remains a standard for heparin management in both ECMO and cardiopulmonary bypass (CPB), despite its weak correlation to the lower plasma heparin levels of ECMO. Furthermore, little is known about the hemostatic alterations related to ECMO versus CPB in children. We hypothesize that the hemostatic profile of ECMO patients differs significantly from those undergoing CPB, and furthermore that age-specific differences exist between neonatal and pediatric ECMO patients. Methods: A prospective observational study evaluating antithrombin III (ATIII), fibrinolysis, thrombin generation, platelet activation and platelet response to agonist during CPB and ECMO at baseline, 4 and 24 hrs. Differences between age groups (neonatal vs. pediatric) in the ECMO group were also compared at baseline, 4, 12, 24, 72, and 144 hrs respectively. Results: The ECMO group was younger, weighed less, and had a lower body surface index than the CPB group. There were significant differences in baseline ATIII levels, platelet functionality, D-Dimer, and F1+2 concentrations between groups. Dichotomization of the ECMO group revealed that D-Dimer and F1+2 generation increased with time in neonatal patients. Conclusions: ECMO generally had greater impact on the hemostatic system than CPB. Circulating platelets were less responsive and markers of fibrinolytic and coagulation activation were greater in the ECMO group at all 3 time points compared between the two therapies. Fibrinolytic and coagulation activation increased with time significantly in the neonatal ECMO population.

Functional electrical stimulation is an important therapeutic tool for improving the quality of life of patients following spinal cord injury. Investigators have developed neural interfaces of varying invasiveness and implant location to stimulate neurons and evoke motor responses. Here we present an alternative interface with the ventral roots (VR) or dorsal root ganglia (DRG). We designed preliminary electrophysiology experiments to evaluate the performance of these interfaces, wherein we stimulated lumbar VR or DRG through a penetrating single-wire microelectrode while recording fixed endpoint force and bipolar electromyograms of hindlimb muscles. Data from rat experiments provided evidence for selectivity for target muscles, graded force recruitment, and nontrivial force magnitudes of up to 1 N. Electrophysiology experiments in cats produced similar results to those in rats. In addition, we developed a computational model to estimate the size and quantity of fibers recruited as a function of stimulus amplitude. This model confirmed electrophysiology results showing differences in the thresholds to detect activity in response to VR versus DRG stimulation. The model also provided insights into the mechanisms by which DRG stimulation is more likely to recruit smaller fibers than larger fibers. Finally, we discuss further work to develop and evaluate these potential interfaces.

Pressure ulcer incidence rates have remained constant [1] even though wheelchair seat cushion technologies have advanced. Shear stress is recognized as a risk factor for pressure ulcer development [2] and is a focus of many shear reduction technologies incorporated into cushions; however, shear reduction has not been quantified in the literature. This study evaluated 21 commercial wheelchair seat cushions using a methodology developed to quantify interface shear stress and calculate overall and local horizontal stiffness values. For statistical analyses, the cushions were grouped by Healthcare Common Procedure Coding System (HCPCS) codes. The general use cushion category (E2601) resulted in significantly greater interface shear stresses (p<.001) than all other categories and the adjustable skin protection cushion category (K0734) resulted in significantly less interface shear stress (p<.001) than all other categories. Additionally, this study provided evidence that the current horizontal stiffness test methodology (ISO 16840-2) [3] provides sufficient information to characterize wheelchair seat cushions, but does not directly quantify interface shear stress. Results from the evaluation of commercial wheelchair seat cushions provided evidence of materials and technologies that may reduce the risk of pressure ulcers. Based on these results, three prototype cushions were conceptualized and prototyped into a closed-loop control system. The closed-loop control system monitored interface stress amplitude to actively modulate cushion properties. None of the prototypes effectively reduced interface shear stress using the methodology developed for cushion testing. Subcutaneous buttock soft tissues were investigated using a finite element model. Researchers have previously used finite element models [4-13]; however, this study improved upon image collection methodology and validation techniques. MR images of one subject were collected in three seated postures and were used to create 3-D models of the buttock. A non-linear 3-D finite element model was developed with anatomical geometries using hyperelastic and viscoelastic constitutive models. Interface pressure, interface shear stress, and soft tissue displacements were used to validate the model. A parametric analysis resulted in a partially validated model that provided subcutaneous stresses and strains for the upright seated posture. The validated model will be used in future studies to evaluate the SCI population and to evaluate commercial and prototype wheelchair seat cushions.

Thesis (M. S.)--Public Policy, Georgia Institute of Technology, 2008. / Kingsley, Gordon, Committee Member ; Walsh, John, Committee Chair ; Castillo, Marco, Committee Member.

<p> A brain-computer interface (BCI) allows human to communicate with a computer by thoughts. Recent advances in brain decoding have shown the capability of BCIs in monitoring physiological and cognitive state of the brain, including drowsiness. Since drowsy driving has been an urgent issue in vehicle safety that causes numerous deaths and injuries, BCIs based on non-invasive electroencephalogram (EEG) are developed to monitor drivers&rsquo; drowsiness continuously and instantaneously. Nonetheless, on the pathway of transitioning laboratory-oriented BCI into real-world applications, there are major challenges that limit the usability and convenience for drowsiness detection (DD). To completely understand the association between human EEG and drowsiness, this study employed a large-scale dataset collected from simulated driving experiments with a lane-keeping task and EEG recordings. A DD-BCI that acquires EEG from only non-hair-bearing (NHB) areas was proposed to maximize the comfort and convenience. The performance of the NHB DD-BCI was validated and compared with that using whole-scalp EEG, showing no significant difference in the accuracy of alert/drowsy classification. In addition, a subject-transfer framework that leverages large-scale existing data from other subjects was proposed to reduce the calibration time of a DD-BCI. Alert baseline data were involved to enhance the efficiency of subject-to-subject model transfer. The subject-transfer approach significantly reduced the calibration time of the DD-BCI, exhibiting the potential in facilitating plug-and-play brain decoding for real-world BCI applications. Overall, this thesis presents the contributions to developing a DD-BCI for real-world use with maximal usability and convenience. The methodologies and findings could further catalyze the exploration of real-world BCIs in more applications. </p><p>

Short interfering RNA (siRNA) is a class of nucleotide drugs with a profound potential to improve patient health through its ability to silence the expression of specific genes at the post-transcriptional level. However, the clinical application of siRNA therapeutics remains hindered by a lack of efficient delivery systems that deposit siRNA into the cytoplasm of cells, a step necessary for siRNA’s silencing effect. Much research has focused on the development of siRNA delivery agents to overcome this challenge. There are no standard pre-clinical models for testing of siRNA delivery agents, and investigators have chosen to evaluate efficacy in a variety of systems including in vitro tissue culture and animal models. These systems have vastly different cellular microenvironments which may modulate cellular behavior and affect the response of cells to siRNA, thus altering the apparent efficacy of siRNA delivery agents. The substrate on which cells adhere is one aspect of the microenvironment that has been previously shown to alter cellular behavior. In this work, we tested the hypothesis that changing the properties of cellular adhesion substrates can change the apparent efficacy of a siRNA delivery agent. Specifically, we used a commonly employed in vitro cationic lipid siRNA delivery vector and evaluated siRNA silencing efficacy in U251 cells seeded on alginate hydrogel surfaces. These surfaces were synthesized to have systematic variation in integrin ligand arginine-glycine-aspartate (RGD) density and elastic modulus. We found that an eightfold increase in RGD content of the alginate grown substrate increased siRNA knockdown efficacy from 25 ± 12% to 52 ± 10%, with constant concentrations of siRNA and delivery agent. We found no difference in siRNA mediated knockdown efficacy over the elastic modulus range tested (53-133 kPa). These results indicate that the cell-adhesion substrate interaction can modulate siRNA protein silencing efficacy, a finding important for evaluation of siRNA therapeutics in the in vitro setting.

<p> In this study, structure-property relationships in the ironclad beetle (Zopherus nodulosus haldemani) exoskeleton are quantified to develop novel bio-inspired impact resistance technologies. The hierarchical structure of this exoskeleton was observed at various length scales for both the ironclad beetle pronotum and elytron. The exocuticle and endocuticle layers provide the bulk of the structural integrity and consist of chitin-fiber planes arranged in a Bouligand structure. The pronotum consists of a layered structure, while elytron consists of an extra layer with &ldquo;tunnel-like&rdquo; voids running along the anteroposterior axis along with smaller interconnecting &ldquo;tunnel-like&rdquo; voids in the lateral plane. Energy dispersive X-ray diffraction revealed the existence of minerals such as calcium carbonate, iron oxide, zinc oxide, and manganese oxide. We assert that the strength of this exoskeleton could be attributed to its overall thickness, the epicuticle layer thickness, the existence of various minerals embedded in the exoskeleton, and its structural hierarchy. The thickness of the exoskeleton correlates to a higher number of chitin-fiber planes to increase fracture toughness, while the increased thickness of the epicuticle prevents hydration of the chitin-fiber planes. In previous studies, the existence of minerals in the exoskeleton has been shown to create a tougher material compared to non-mineralized exoskeletons.</p><p>