The Vanderbilt University Free Electron Laser (FEL) is a tunable source of pulsed infrared radiation with pulse characteristics unlike those of most laser systems. A primary objective of the research presented in this dissertation is to investigate the effects of chromophore and pulse characteristics in the ablation of soft tissues with the (FEL). The working hypothesis of the research project is that results of ablation of soft tissues with the FEL cannot be solely explained by the selective absorption of protein components in the tissue, and that the pulse characteristics of the laser play an important role. Three related studies are presented in this dissertation. First, the ablation depth and ablation threshold of rat dermis irradiated with the FEL at many wavelengths were measured and analyzed to reveal gross effects of the ablation process. Second, acoustic transients generated during the ablation of rat dermis and gelatin samples were measured and analyzed to reveal the effects of protein absorption and mechanical strength in the ablation process. Finally, numerical modeling of the ablation process was employed to investigate the effect of the temporal pulse structure of the laser and the effect of dynamic absorption of water on the ablation process. The results of the studies presented here led to the following conclusions. First, the ablation of soft tissues irradiated with the FEL is largely described by a steady-state ablation model, indicating that the mechanism of ablation is predominantly photothermal in nature. Second, the ablation of soft tissues with infrared FEL radiation is a surface-mediated process, similar to that of traditional ultraviolet laser tissue ablation. Third, the dynamic absorption of water plays a significant role in the process. Finally, protein absorption of the incident radiation results in the targeted destruction of the tissue structural matrix at wavelengths where the absorption of protein represents a significant fraction of the overall absorption cross-section.

<p>A traditional medical school curriculum consists of a large amount of information presented by a large number of faculty. Faced with a growing and evolving flood of information, medical educators require and seek assistance to manage this knowledge base. Between the fall of 2001 and summer of 2002, researchers and educators at the Vanderbilt University Department of Biomedical Informatics have constructed KnowledgeMap (KM), a web-based knowledge management tool to support medical instruction at the Vanderbilt University School of Medicine. The KM knowledge store contains all curricular documents in a searchable database. The KM interface makes available online this information to students, faculty members, and administrators. This thesis analyzes the use of KM during its pilot implementation in the first year medical school anatomy course during the fall of 2002.</p><p>Data was collected from first year medical students and the anatomy course faculty through server log files, a survey, and interviews. The data revealed that students have utilized the system near unanimously and that the majority of them have expressed satisfaction with the system. Computer proficiency and test anxiety were identified as factors affecting the adoption of the system. The study addressed issues relating to both students and faculty that arise from the use KM. They include student time constraints and learning styles, class attendance, intellectual property concerns, communication among faculty, and the properties of hypermedia as a medium for instruction at the Vanderbilt University School of Medicine.</p>

The purpose of this research is to study and quantify the effects of system and human factors on objective workload, subjective workload and physiological stress in residents and attending physicians working in an emergency department (ED) at a Level I trauma center. The study design is a time-motion task analysis that incorporates objective, subjective, and physiological measures of stress and workload. Several procedural methods and workload assessment techniques were developed, integrated and used to dissect the dynamic ED work environment. Descriptive statistics characterizing this environment are calculated and compared to previous studies. Methodologies developed for measuring workload continuously are implemented and discussed. The study demonstrates the applicability of human factors engineering to describe a medical work environment and identify potential shortcomings in system and provider-level care.

BIOMEDICAL ENGINEERING THE EFFECT OF PULSE STRUCTURE ON SOFT TISSUE LASER ABLATION AT MID-INFRARED WAVELENGTHS MARK ANDREW MACKANOS Dissertation under the direction of Professor E. Duco Jansen A series of experimental investigations have demonstrated that targeting a mid-infrared Mark-III Free-Electron Laser to wavelengths near 6.45 Ým results in tissue ablation with minimal collateral damage and substantial efficiency useful for human surgery. Thermodynamic reasoning suggests that the minimal collateral damage at this wavelength is due to the differential absorption of protein and water; which causes compromised tissue integrity by laser heating of the non-aqueous components prior to explosive vaporization. These properties are advantageous for surgery because they reduce the structural integrity of the tissue, thus reducing amount of energy needed for ablation. While the FEL, based on these findings, has been used successfully in eight human surgeries to date, it is unlikely that this laser will become broadly used clinically due to its expense and difficult implementation. Recent developments in conventional laser technology have provided access to this wavelength. While the average and peak power of these sources are still not equivalent to the FEL, recent data indicates that ablation studies are feasible. The research described here investigates the role of pulse structure with regards to soft tissue ablation to determine the feasibility of these sources as potential FEL replacements for clinical applications. Relevant parameters including the threshold radiant exposure and ablated crater depth were studied for comparison of the native FEL micropulse with a stretched FEL micropulse and a ZnGeP2 OPO. Brightfield imaging was used to analyze the effect of pulse structure on the dynamics of ablation, while histology on cornea and dermis was performed to study pulse effects on thermal damage. Mass spectrometry was also used to investigate whether non-linear effects are involved with the FEL micropulse in changing the chemical structure of proteins prior to ablation. The results of this analysis show that the micropulse structure of the FEL does not play a role in the efficient ablation of soft tissue with minimal collateral damage that has been shown previously, and alternative sources remain viable alternatives to the FEL.

This dissertation covers research regarding the use of laser range scanning (LRS) during neurosurgery. Impetus for this work stems from the desire to provide relevant intraoperative data regarding the position and motion of the brain relative to preoperative image tomograms. LRS provides an excellent technology for providing fast, accurate, and well-resolved surface data of the exposed brain. Methods described in this dissertation represent a novel visualization system capable of providing real-time cortical surface characterization. More specifically, techniques have been developed that register intraoperative LRS data to preoperative MR tomograms and quantify the motion of the brain using serial LRS acquisitions. The results generated from these techniques are presented as graphical renderings that provide correspondence between the exposed cortical surface and anatomical structures in the preoperative tomograms. In vivo validation shows that cortical surface registration and motion tracking can be achieved to millimetric accuracy. The impact of these results allows enhanced recognition of cortical structures while providing meaningful assessment of brain deformation during surgery. In summary, this research provides a comprehensive examination of LRS for use within the operating theater and constitutes a significant step toward the use of intraoperative cortical surface data in image-guided neuronavigation.

Pancreatic islet transplantation is a promising treatment for type 1 diabetes. However, current efforts to study islet transplantations are hampered by the lack of a non-invasive method of imaging islets and quantifying islet mass post transplantation. Transplanted pancreatic islets can be imaged and quantified non-invasively using in vivo bioluminescence imaging (BLI). Pancreatic islets transfected with the firefly reporter gene, luciferase, emit light that can be quantified using photon-counting measurements. Pancreatic islet number is linearly related to light emission both in vitro and in vivo. Application of bioluminescence imaging for this application can be greatly enhanced by relating light emission to the number of islets surviving post-transplantation. Determining this relationship requires detailed knowledge of the factors that influence photon-counting measurements. Bioluminescence was modeled using constant light emitting phosphorescent beads implanted at the two common sites of islet transplantation: the renal capsule and liver. This model was used to quantify light attenuation by tissues overlying the islet transplantations. The ratio of implanted light emission to in vitro light emission was found to be .2394+/- 0.0261 for renal implantation and 0.0645 +/- 0.0140 for hepatic implants. Mathematical modeling of light propagation using Monte Carlo simulation is in excellent agreement with these experimental results. Monte Carlo modeling yields an in vivo to in vitro luminescence ratio for renal and hepatic sources to be 0.2860 and 0.0495, respectively. Surgical artifacts were found to influence bioluminescence measurements. Surgical scar tissue leads to lower light emission the week immediately post-op, but this attenuation is negligible two weeks after surgery. The orientation of the subject also influences quantification of bioluminescence. Rotation of 50 degrees from flat can lead to a 73% decrease in light transmission for renal implants and 52% decrease for hepatic implants. The rate of luminescence decrease with increasing angle depends on the surface light is projected upon. Flatter surfaces lead to a slower decrease in luminescence while higher curvature leads to more rapid decrease in luminescence. Spot size of bioluminescence was found to increase with increasing tissue depth. The spot size of hepatic implants was found to be 17% larger than renal implants, as measured by full width at half maximum measurements. Constant light emission modeling of transplanted islet bioluminescence permits quantification of actual islet number from photon counting measurements and insight into factors which influence these measurements.

A new inter-modality surface registration algorithm that uses textured point clouds and mutual information is presented within the context of model-updated image guided procedures. The algorithm has been developed to capitalize on the information generated by a laser range scanner. The current iteration of the algorithm is optimized for cortical surface registration. Intra-modality validation for the algorithm is provided in both physical and imaging phantoms. The physical phantom is generated using a laser range scanner that reports texture coordinates. The imaging phantom is generated from gadolinium enhanced MR volumes of the brain. Simulated inter-modality registration experiments on a cortical surface are also presented. Results of the experiments show successful registration accuracies on the order of the resolution of the surfaces (i.e. submillimetric). The results demonstrate that the registration algorithm and laser range scanner have potential application in deformation tracking during surgery and model-updated image-guided procedures.

Cell encapsulation is a novel therapeutic approach for the treatment of Type I Diabetes Mellitus that circumvents both the immunosuppression and limited allograft donor source dilemma. Current methods for scoring the biocompatibility of the alginate-based capsules that sequester Islets of Langerhans include fabrication and implantation into the peritoneal cavity of mice, incubation for specified periods of time, retrieval via peritoneal lavage, and observation of the number of cells or cell layers surrounding the capsules. This method allows only one data point to be obtained per animal. In this experiment we propose to measure biocompatibility real time and in vivo This new method of monitoring immune response using bioluminescent technology and a Nuclear Factor-kappa Beta sensitive transgenic mouse model allows unlimited data points to be acquired per animal, reduces the number of animals required to obtain statistically significant immune response data over time, and in turn reduces error associated with animal variability. NF-kB is a transcription factor that plays a critical upstream role in the coordination of the inflammatory and would healing cascades by initiating the transcription of many cytokines, chemokines, adhesion molecules, and proinflammatory genes. Five types of capsules were monitored over 6 six weeks after transplantation into the dorsal-cervical fat pad, a capsule group, a bead group, a non-coated capsule group, a sham surgery group, and a control group. The bead, capsule, and non-coated capsule transplant groups allow the effects of capsule size and capsule wall composition on NF-kB activity to be monitored. This imaging modality was able to detect statistically significant differences in NF-kB activity between pre and post-operative data points per mouse. It was also able to discern with significance an unexpected increase in NF-kB activity due to capsule size instead of capsule wall composition over a six week time period.

Trait heterogeneity, which exists when a trait has been defined with insufficient specificity such that it is actually two or more distinct traits, has been implicated as a confounding factor in traditional statistical genetics of complex human disease. In the absence of detailed phenotypic data collected consistently in combination with genetic data, unsupervised computational methodologies offer the potential for discovering underlying trait heterogeneity. The performance of three such methodsBayesian Classification, Hypergraph-Based Clustering, and Fuzzy k-Modes Clusteringthat are appropriate for categorical data were compared. Also tested was the ability of these methods to additionally detect trait heterogeneity in the presence of locus heterogeneity and gene-gene interaction, which are two other complicating factors in discovering genetic models of complex human disease. Bayesian Classification performed well under the simplest of genetic models simulated, and it outperformed the other two methods, with the exception that the Fuzzy k-Modes Clustering performed best on the most complex genetic model. Permutation testing showed that Bayesian Classification controlled Type I error very well but produced less desirable Type II error rates. Methodological limitations and future directions are discussed.

The success of treatment of patients with cancer depends on establishing an accurate diagnosis. To this end, we have built a system called GEMS (Gene Expression Model Selector) for the automated development and evaluation of high-quality cancer diagnostic models and biomarker discovery from microarray gene expression data. In order to determine and equip the system with the best performing diagnostic methodologies in this domain, we first conducted a comprehensive evaluation of classification algorithms using 11 cancer microarray datasets. After the system was built, we performed a preliminary evaluation of the system with 5 new datasets. The performance of the models produced automatically by GEMS is comparable or better than the results obtained by human analysts. Additionally, we performed a cross-dataset evaluation of the system. This involved using a dataset to build a diagnostic model and to estimate its future performance, then applying this model and evaluating its performance on a different dataset. We found that models produced by GEMS indeed perform well in independent samples and, furthermore, the cross-validation performance estimates output by the system approximate well the error obtained by the independent validation. GEMS is freely available for download for non-commercial use from http://www.gems-system.org.