The evaluation of the literature is an increasingly integral part of biomedical research. Clinicians, researchers, librarians, and others routinely use the literature to answer questions for clinical care and research. The size of the literature prevents the manual review of all documents, and automated methods are necessary for identifying high quality articles as a major filtering step. This work aimed to improve the performance and usability of existing tools with machine learning methods. First, evaluation methods for journals, articles, and websites were studied to determine if their performance varied widely for different topics. Second, the feasibility of predicting article citation count was examined by training Support Vector Machine (SVM) models on content and bibliometric features. Third, SVM models were used to automatically classify instrumental and non-instrumental citations.

The goal of this work is to develop and demonstrate a novel model of superparamagnetic iron oxide nanoparticle (SPION) magnetic properties based on physical first principles and experimental mathematical relationships. SPIONs exhibit magnetic properties that differ from the bulk properties of iron oxide due to scale affects unique to nanoparticles. The developed model is able to predict the magnetic properties of any type of SPION at a given temperature and applied field strength based solely on the particle size. By predicting SPION magnetization and induced magnetic field, the model is a useful engineering tool for nanomaterials design. Using the model, it is possible to predict the magnetic behavior of even complex SPION based nanomaterials, facilitating materials design rather than pure discovery using costly high throughput methods. Using this model, we have investigated the magnetic properties of a clustered system of SPIONs to potentially be used as a magnetic detection device and image contrast agent. Using the model, it is possible to predict the ideal particle size for these particular nanomaterials by optimizing key magnetic parameters with regards to a specific application.

Algorithms for discovery of a Markov boundary from data constitute one of the most important recent developments in machine learning, primarily because they offer a principled solution to the variable/feature selection problem and give insight about local causal structure. Even though there is always a single Markov boundary of the response variable in faithful distributions, distributions with violations of the intersection property may have multiple Markov boundaries. Such distributions are abundant in practical data-analytic applications, and there are several reasons why it is important to induce all Markov boundaries from such data. However, there are currently no practical algorithms that can provably accomplish this task. To this end, I propose a novel generative algorithm (termed TIE*) that can discover all Markov boundaries from data. The generative algorithm can be instantiated to discover Markov boundaries independent of data distribution. I prove correctness of the generative algorithm and provide several admissible instantiations. The new algorithm is then applied to identify the set of maximally predictive and non-redundant molecular signatures. TIE* identifies exactly the set of true signatures in simulated distributions and yields signatures with significantly better predictivity and reproducibility than prior algorithms in human microarray gene expression datasets. The results of this thesis also shed light on the causes of molecular signature multiplicity phenomenon.

The correlation between changes in mechanical properties and the onset of disease has led to an increased interest in assessing the elastic modulus of soft tissues as a biomarker for disease progression. In addition, soft tissue mechanical properties are desired for biomechanical modeling for surgical procedure planning and intraoperative guidance. Unfortunately, soft tissue modulus evaluation has proven inherently difficult due to tissue consistency and shape, and the approaches are highly variant. The work presented in this thesis focuses on the development, application, and interrogation of a novel soft tissue mechanical property evaluation technique, termed the Model-Gel-Tissue (MGT) assay, which utilizes a combination of a gel embedding process, direct mechanical testing, and computational modeling to analyze the elastic properties of a soft tissue sample. The goal was to develop a repeatable and adaptable evaluation technique that also allowed for irregularly shaped specimens and standardization of the implementation. This was accomplished by a rapid-embedding of the tissue in a gel with surfaces of known and uniform shape. The mechanical testing output is then utilized in a finite element model of the system developed from computed tomography (CT) scans of the specimen, in order to evaluate the mechanical properties of the embedded tissue. Preliminary testing of the MGT assay was implemented using fibrotic murine livers to assess the capability of the technique relative to traditional indentation testing. The assay was then used to investigate the correlation between microstructural collagen content and macroscopic tissue modulus in a murine model of breast cancer. Subsequently, the assay was used to investigate the propensity of modulus as an indicator of treatment resistance in a second murine model of breast cancer. Finally, extensive sensitivity tests were performed to qualify the fidelity of the system. The results of this work show that modulus assessment via the MGT assay correlates to traditional testing, as well as to tissue collagen content, and the concatenation of the work indicates that the MGT assay serves as a reliable and adaptable soft tissue modulus evaluation system.

The purpose of this research was to evaluate the collaborative design of an online community for cancer survivorship in middle Tennessee. The four primary aims of this qualitative study were to define the local cancer survivorship community, identify its strengths and opportunities to improve, build an online community to address these opportunities, and evaluate the collaborative design and development of this online community. A total of 43 cancer survivors, family members, health-care professionals, and community professionals participated in key informant interviews, sense of community surveys, and the collaborative design of the online community over a one-year period. The results of this study include a formal definition of the local cancer survivorship community and illustrate how support for cancer survivors extends throughout the local community. Six opportunities were identified to improve the sense of community in the local cancer survivorship community, and an online community was successfully developed to address these opportunities. The evaluation of the collaborative process resulted in a seven element framework for the discovery and development of community partnerships for informatics design. These results demonstrate the potential for an informatics-based approach to bring local communities together to improve supportive care for cancer survivors. Implications of the findings call for a new initiative for cancer survivorship that uses emerging web-based technologies to improve collaborative cancer care and quality of life in local communities.

This study demonstrated the relationship between brain connectivity and childrens reading abilities. For the behavioral part, the participants received proper reading interventions based on their responsiveness, and the standardized behavioral tests were administered throughout the process. For the imaging part, both T1-weighted images and diffusion weighted images were acquired. Nine cortical regions in each brain hemisphere were identified as regions of interest (ROI). The probabilistic streamlines connecting each pairing of the nine regions were calculated and used to estimate brain connectivity. The estimates were then used to correlate with childrens reading measures. Eight significant correlations were found, four of which were connections between the insular cortex and angular gyrus. The results are suggestive of a key role of connection between insular cortex and angular gyrus in mediating reading behavior. In spite of the limited sample size, the redundancy in the spread of group clusters is indicative of a relation between brain connectivity and childrens responsiveness to intervention.

The use of 18FDG-PET imaging to predict treatment response to IGF-1R/IR targeted therapy in mouse models of human lung cancer is presented in this thesis. In vitro cell studies were first conducted to establish sensitivity to treatment with OSI-906 and changes in glucose metabolism in responding cells. In vivo xenograft studies demonstrated that reduced 18FDG-PET correlated with PI3K pathway inhibition and was able to predict tumor response to OSI-906 prior to changes in tumor volume could be ascertained. The in vivo imaging results were validated with molecular correlates. Based upon these results 18FDG-PET imaging appears to serve as a rapid non-invasive marker of IGF-1R/IR inhibition and should be explored clinically as a predictive clinical biomarker in patients undergoing IGF-1R/IR-directed cancer therapy.

Pediatric asthma exacerbations account for >1.8 million ED visits annually. Guidelines can decrease variability in asthma treatment and improve clinical outcomes; however, guideline adherence is inadequate. We evaluated a computerized asthma detection system based upon the NHLBI guidelines and evidence-based practice to improve care in a pediatric ED in a two phase study. Phase I looked at using an automatic disease detection system to identify eligible patients and then printing the paper-based guideline. Phase II implemented a fully computerized asthma management system. Although the time to disposition decision was not statistically significant, we believe this management system to be a sustainable computerized management system to help standardize asthma care. The computerized asthma management system represents a work-flow oriented, sustainable approach in a challenging environment.

Molecular medicine encompasses the application of molecular biology techniques and knowledge to the prevention, diagnosis and treatment of diseases and disorders. Statistical and computational models can predict clinical outcomes, such as prognosis or response to treatment, based on the results of molecular assays. For advances in molecular medicine to translate into clinical results, clinicians and translational researchers need to have up-to-date access to high-quality predictive models. The large number of such models reported in the literature is growing at a pace that overwhelms the human ability to manually assimilate this information. Therefore the important problem of retrieving and organizing the vast amount of published information within this domain needs to be addressed. The inherent complexity of this domain and the fast pace of scientific discovery make this problem particularly challenging. <p> This dissertation describes a framework for retrieval and organization of clinical bioinformatics predictive models. A semantic analysis of this domain was performed. The semantic analysis informed the design of the framework. Specifically, it allowed the development of a specialized annotation scheme of published articles that can be used for meaningful organization and for indexing and efficient retrieval. This annotation scheme was codified using an annotation form and accompanying guidelines document that were used by multiple human experts to annotate over 1000 articles. These datasets were then used to train and test support vector machine (SVM) machine learning classifiers. The classifiers were designed to provide a scalable mechanism to replicate human experts ability (1) to retrieve relevant MEDLINE articles and (2) to annotate these articles using the specialized annotation scheme. The machine learning classifiers showed very good predictive ability that was also shown to generalize to different disease domains and to datasets annotated by independent experts. The experiments highlighted the need for providing unambiguous operational definitions of the complex concepts used for semantic annotations. The impact of the semantic definitions on the quality of manual annotations and on the performance of the machine learning classifiers was discussed.

Both angiogenesis and inflammation are inescapable in vivo responses to any type of biomaterials implanted for regeneration. Continuous progress has been made in biomaterial design to facilitate tissue interactions with an implant by reducing inflammation and/or by inducing angiogenesis. However, it becomes increasingly clear that the physiological processes of angiogenesis and inflammation are interconnected through various molecular mechanisms. The role of implant-induced inflammation in the formation of new blood vessels into tissue surrounding the implant remains unclear. Therefore, we used a polyethylene glycol (PEG) cross-linked tyrosine derived polycarbonate hydrogel system as a model of implantable biomaterials. As opposed to the degradation rate, modulus and protein adsorption decreased as the cross-linking degree increased, due to hydrophilic repellent properties of PEG, indicating the unique and tunable hydrogel properties. The hydrogels were hybridized with pro- or anti-angiogenic (or inflammatory) peptides using collagen or fibrin gel and used for in vitro and in vivo biological studies. The results show a clear interconnectivity between angiogenic and inflammatory activities, indicating an inflammatory mechanism regulating follow-up angiogenic processes in hydrogels. This study suggests a new concept of biomaterial design that utilizes flexible inflammatory parameters to control angiogenesis for the eventual success of biomaterial implants.