Regeneration of large bone defects presents a critical challenge to orthopaedic clinicians as the current treatment strategies are severely limited. Tissue engineering has therefore emerged as a promising alternative to bone grafting techniques. This approach features the delivery of bioactive agents such as stem cells, genes, or proteins using biomaterial delivery systems which together stimulate endogenous repair mechanisms to regenerate the tissue. Because bone is a highly mechanosensitive tissue which responds and adapts dynamically to its mechanical environment, application of mechanical stimuli may enhance endogenous tissue repair. While mechanical loading has been shown to stimulate bone fracture healing, the ability of loading to enhance large bone defect regeneration has not been evaluated. The goal of this thesis was to evaluate the ability of sustained osteogenic growth factor delivery and functional biomechanical loading to stimulate vascularized repair of large bone defects in a rat segmental defect model. First, we evaluated the hypothesis that the relationship between protein dose and regenerative efficacy depends on delivery system. We determined the dose-response relationship between dose of recombinant human bone morphogenetic protein-2 (rhBMP-2) and bone regeneration in a hybrid alginate-based protein delivery system and compared with the current clinically-used collagen sponge. The hybrid delivery system improved bone formation and reduced the effective dose due to its sustained delivery properties in vivo. Next, we tested the hypothesis that transfer of compressive ambulatory loads during segmental defect repair enhances bone formation and subsequent limb regeneration. We found that delayed application of axial loads enhanced bone regeneration by altering bone formation, tissue differentiation and remodeling, and local strain distribution. Finally, we evaluated the hypothesis that in vivo mechanical loading can enhance neovascular growth to influence bone formation. We found that early mechanical loading disrupted neovascular growth, resulting in impaired bone healing, while delayed loading induced vascular remodeling and enhanced bone formation. Together, this thesis presents the effects of dose and delivery system on BMP-mediated bone regeneration and demonstrates for the first time the effects of in vivo mechanical loading on vascularized regeneration of large bone defects.

Modifications of biomaterial surface properties are employed to increase osteoblast differentiation and bone formation. Microtextured metallic surfaces promote osteoblast differentiation and high surface energy- achieved by controlling surface hydrocarbon contamination- increases osteoblast differentiation and peri-implant bone formation. Recombinant human bone morphogenic protein 2 (BMP2) is approved to induce bone formation in a number of applications. It is used clinically in combination with biomaterials to improve peri-implant bone formation and osseointegration. The amount of BMP2 that is required is large and inflammatory (swelling/seroma) and bone-related (ectopic bone/bone resorption) complications have been reported after BMP2 treatment. The aim of this study was to examine potential deleterious effects of BMP2 on the inflammatory environment and apoptosis of osteoblasts. Surface roughness and energy decreased pro-inflammatory interleukins and increased anti-inflammatory interleukins. In contrast, BMP2 abolished the surface effect, increasing pro-inflammatory interleukin (IL) 6, IL8, and IL17 in a surface roughness-dependent fashion and decreasing anti-inflammatory IL10 on rough surfaces. 5Z-7-Oxozeaenol and Dorsomorphin, but not H-8, blocked the effect of BMP2 on IL1A expression. There was an increase in expression of IL6 when treated with BMP2 for the control and H-8 groups, but both 5Z-7-Oxozeaenol and Dorsomorphin blocked the effect. Both 5Z-7-Oxozeaenol and H-8 blocked the effect of BMP2 on IL10 expression. BMP2 treatment had little effect on apoptosis in human mesenchymal stem cells (MSCs). Exogenous BMP2 had no effect on TUNEL. Caspase-3 activity was increased only at 200ng/ml BMP2. BAX/BCL2 decreased in MSCs treated with 50 and 100ng/ml BMP2. In contrast, BMP2 increased caspase-3 activity and TUNEL at all doses in normal human osteoblasts (NHOst). BAX/BCL2 increased in NHOst treated with BMP2 in a dose-dependent manner. Cells treated with 200 ng/ml BMP2 had an 8-fold increase in BAX/BCL2 expression in comparison with untreated cells. Similarly, BMP2 increased DNA fragmentation in NHOst cells. The BMP2-induced increase in DNA fragmentation was eliminated by 5-Z7-Oxozeaenol and Dorsomorphin. The results suggest that while surface features modulate an initial controlled inflammatory response, the addition of BMP2 induces a pro-inflammatory response. The effect of BMP2 on apoptosis depends on cell maturation state, inducing apoptosis in committed osteoblasts. BMP2 together with microtextured orthopaedic and dental implants may increase inflammation and possibly delay bone formation. Dose, location, and delivery strategies are important considerations in BMP2 as a therapeutic and must be optimized to minimize complications.

Thesis (D.Min.)--Dallas Theological Seminary, 2008. / Includes abstract. Description based on print version record. Includes bibliographical references (leaves 146-149).

Thesis--University of Oklahoma. / Includes bibliographical references (leaves 51-55).

Mobile devices are playing an increasingly prevalent role in K-12 education, as school systems are adopting the technology to enhance student learning. Consequently, teachers have to learn how to incorporate the devices into their classrooms, with the help of professional development activities. This case study examined the professional development of four teachers who participated in their school district’s iPod touch initiative for English Language Learners (ELLs), as well as the perspective of the instructional media specialist charged with assisting the teachers. The study aimed to understand what district-led and independent training activities K-12 teachers engage in and the influence the activities have on how mobile devices are used by the teachers. The educational activities of the teachers and their impact on the teachers’ technology integration were documented through a series of interviews, a training observation, and teacher questionnaires. The research identified four types of professional development support provided by the district and five types of self-guided or incidental learning activities that teachers engaged in that directly impacted the way they used the device. The district-led trainings are 1) group trainings 2) in-class demonstrations 3) one-on-one training and 4) ongoing support. The self-guided and incidental learning activities found are 1) research for resources and ideas 2) brainstorming 3) experimenting with apps and activities 4) collaboration with others and 5) students serving as trainers. Increased personal instruction, cultivating formal learning through mentoring and an online forum, and developing online training resources for a mobile format are proposed for professional development that would aid in the integration of mobile devices in a K-12 environment. / text

Performance of modern pipelined processor depends on steady flow of useful instructions for processing. Branch instruction disrupts sequential flow of instructions by presenting multiple paths through which a program can proceed. By predicting branch outcome early, branch predictors allow processor to continue fetching instructions from the predicted path. History-based dynamic branch predictors have shown to reach high prediction accuracy, yet certain branch types continue to mispredict. These are multitarget indirect branches and data-dependent direct and indirect branches. These are hard-to-predict branches since their outcome do not always exhibit repeatable patterns. This thesis describes branch prediction schemes for improving prediction accuracy of hard-to-predict branches using data value correlation. In these schemes, instead of relying on branch history information, compiler identifies program instructions whose output value strongly correlates with branch outcome. These correlated instructions are tracked at run-time, and their output is used for making branch predictions. Specifically, this thesis proposes following two branch prediction schemes: (i) Value-based BTB indexing (VBBI) is a low cost, compiler-guided scheme for predicting multi-target indirect branches. For every indirect branch, compiler identifies an instruction whose output strongly correlates with targets taken by the indirect branch. At run-time, multiple branch targets are stored and subsequently accessed from BTB using index formed by hashing indirect branch PC with output of the correlated instruction. (ii) Store-Load-Branch (SLB) predictor is a compiler-assisted branch prediction scheme for data-dependent branches. Typically, data-dependent branches are associated with program data structures such as arrays, linked list etc., and follow store-load-branch execution sequence. A set of memory locations is written at an earlier point in a program. Later, these locations are read, and used for evaluating branch condition. Branch outcome depends on values stored in data structure, which, normally do not have repeatable patterns. In SLB scheme, compiler identifies all program points where data structure associated with a data-dependent branch is modified. These marked store instructions are tracked at run-time, and stored values are used for computing branch flags ahead of time. Later, when branch instruction is fetched, pre-computed flags are read, and used for making predictions. This thesis introduces new branch prediction schemes, describes hardware structures and compiler analysis for implementing these schemes, evaluates their performance impact, and estimates their area, power and timing overhead. / text

Protein-protein interactions are central to most biological processes. Although in the field of drug discovery there is a great interest in targeting protein-protein interactions, the discovery and development of small-molecules, which effect these interactions has been challenging. The purpose of this project is to determine if in situ click chemistry is a practical approach towards testing whether Bcl-XL is capable of assembling it's own inhibitory compounds. Abbott laboratories developed compound ABT-737, which binds with high affinity (Ki < 1 nM) to the binding sites of Bcl-XL.³⁶ Based on ABT-737, two acetylene anchor molecules AM3 and AM4 have been synthesized. These anchor molecules are distinguished by the reactivity of the their carbon-carbon triple bond. Compound AM3 contains an electron withdrawing carbonyl in the alpha-position to the acetylene resulting in an activating effect towards the [1,3]-dipolar cycloaddition compared to compound AM4. To determine the reactivity of the activated system, ¹ H-NMR kinetic studies were performed to compare the relative rates of these two systems by reacting model alkynes 1,2,3, and 4 with azide AZ7. It was shown that the activated systems, 1 and 3, produce triazoles in an accelerated rate compared to the unactivated systems 2 and 3. To test for the self-assembly of inhibitory triazoles, the acetylenes AM3 and AM4 were incubated with Bcl-XL and 14 azide building blocks (AZ1-AZ12) for 12 hours at 37 degrees C. Subjecting these mixtures to LC/MS-SIM led to the discovery of two hit compounds, 35 and 36, of which 35 has been chemically synthesized confirming the hit. Future work includes the synthesis of all hit compounds. Since hit triazoles can be syn or anti, both need to be synthesized for each hit to investigate which regioisomer Bcl-XL generates. Tests to confirm if hit compounds are actually modulating Bcl-XL activity will be done using conventional bio-assays. This will validate that Bcl-XL is capable of assembling its own inhibitor via the in situ click chemistry approach to drug discovery.

This work presents an evaluation of a reform teaching practice, known as peer-led guided inquiry, that combines guided inquiry and cooperative learning for college chemistry teaching. Integral to implementing the reform in a large class (greater than 100 students) was the role of peer leaders, undergraduate students who have successfully completed the target course. These peer leaders facilitated cooperative learning groups during weekly guided inquiry activities in general chemistry. The evaluation, using data collected over a 3-year period, had two main foci: effective teaching and promotion of equity in the classroom. Both of these aims were evaluated using hierarchical linear models. The reform was found to be effective, with a progressive increase in the test scores of those students in the reform classes versus the students in the traditional classes. Furthermore, students in the reform outperformed their counterparts on an externally-constructed national exam. Both findings also held true when controlling for student SAT scores. Effectiveness is not sufficient cause for recommendation amid concerns that distinct groups of students may be disadvantaged by a reform. The evaluation therefore had special concern for students who were at significant risk of low performance in a college chemistry course, such as those with poor high school preparation. No evidence was found that the reform made the situation worse for these students; in fact, the reform was determined to be effective regardless of preparation as measured by SAT scores. In addition, formal thought ability was found to be an important factor in chemistry performance, independent of SAT scores, with low formal thought ability placing students at-risk. The evaluation data indicated that the reform may have allowed students who entered the course with low formal thought ability to overcome this disadvantage, though this effect could be attributed to chance. Finally, to understand further the students in this setting beyond cognitive factors, an inventory of student study approaches was administered. Three specific approach profiles were prevalent: surface, surface achieving and achieving. Two less prevalent approach profiles, deep and deep achieving, were related to better understanding of chemistry as measured by the national exam.

In this dissertation different structural materials (aluminum and steel) with different geometrical shapes (plates, pipes and bars) are studied for damage detection with guided waves. Specific guided wave modes (also known as Lamb wave modes for plate type structures) are generated in a laminated aluminum plate for damage detection and quantification using a broad band piezoelectric transducer structured with a rigid electrode. Appropriate excitation frequencies and modes for inspection are selected from theoretical and experimental dispersion curves. Sensitivity of anti-symmetric and symmetric modes for delamination detection and quantification is investigated. Longitudinal guided waves are excited and recorded after transmission through reinforcing steel bars for monitoring its corrosion level. Instead of investigating the amplitude of the transmitted guided waves, or in other words, monitoring its attenuation, the differential time-of-flight (TOF) is recorded. A reliable guided wave mode is identified for the detection and quantification of corrosion in reinforcing steel bars. Hole type damage, and bonding/de-bonding or lamination/delamination in pipes are studied with Noncontact Electro-Magnetic Acoustic Transducers and PZT transducers. An adaptive method using phase of the recorded signals for detection and quantification of damages in pipes is established using multiple feature extraction techniques (Time-Frequency representations) and differential time-of-flight cross-correlation technique.

This research covers modeling of Electro Magnetic Acoustic Transducers (EMATs) and their application in excitation and detection of longitudinal guided Lamb wave modes for evaluation of flaws in cylindrical pipes. The combination of the configuration of transducers and the frequency of the input current is essential for successful excitation of desired guided wave modes and for proper interpretation of the results. In this study EMATs were successfully constructed and longitudinal modes L(0,1) and L(0,2) were excited in the pipe. From the recorded signals the level of simulated damage in pipes could be assessed. It is also possible to theoretically predict the location of the pipe flaws. Theoretical predictions are matched with experimental results. Dents and holes in pipes are detected by appropriate signal processing of received L(0,1) and L(0,2) modes.