Thesis (Ph. D.)--State University of New York at Binghamton, Thomas J. Watson School of Engineering and Applied Science, Department of Systems Science and Industrial Engineering, 2009. / Includes bibliographical references.

Thesis (Ph. D.)--University of Missouri-Columbia, 2005. / The entire dissertation/thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file (which also appears in the research.pdf); a non-technical general description, or public abstract, appears in the public.pdf file. Vita. "July 2005." Includes bibliographical references.

Thesis (M.Phil.)--Hong Kong University of Science and Technology, 2006. / Includes bibliographical references (leaves 96-100). Also available in electronic version.

Thesis (Ed.D.)--Duquesne University, 2009. / Title from document title page. Abstract included in electronic submission form. Includes bibliographical references (p. 86-113) and index.

We present Information System Hazard Analysis (ISHA), a novel systemic hazard analysis technique focused on Clinical Information System (CIS)s. The method is a synthesis of ideas from United States Department of Defense Standard Practice System Safety (MIL-STD-882E), System Theoretic Accidents Models and Processes (STAMP) and Functional Resonance Analysis Method (FRAM). The method was constructed to fill gaps in extant methods for hazard analysis and the specific needs of CIS. The requirements for the method were sourced from existing literature and from our experience in analysis of CIS related accidents and near misses, as well as prospective analysis of these systems. The method provides a series of iterative steps which are followed to complete the analysis. These steps include modelling phases that are based on a combination of STAMP and FRAM concepts. The method also prescribes the use of triangulation of hazard identification techniques which identify the effects of component and process failures, as well as failures of the System Under Investigation (SUI) to satisfy its safety requirements. Further to this new method, we also contribute a novel hazard analysis model for CIS as well as a safety factor taxonomy. These two artifacts can be used to support execution of the ISHA method. We verified the method composition against the identified requirements by inspection. We validated the method’s feasibility through a number of case studies. Our experience with the method, informed by extant safety literature, indicates that the method should be generalizable to information systems outside of the clinical domain with modification of the team selection phase. / Graduate

Falls are the second leading cause of accidental or unintentional injury deaths worldwide. Many factors contribute to an increased risk of falling, such as age, disease state, and medication use. The purpose of the current investigation was to compare an electronic medical record (EMR) fall risk calculator, the theoretical Timed Up and Go (T-TUG), which utilizes gender, age, BMI, and prescription and OTC drug counts as variables, to other established performance- and paper-based assessments of fall risk. The National Social Life, Health, and Aging Project (NSHAP) Database was used to develop the T-TUG. Data was analyzed from participants in Wave 1 of the Irish Longitudinal Study on Ageing (TILDA) to validate the T-TUG. Performance-based assessments included mean grip force for both dominant and nondominant hands, Timed Up and Go (TUG), and a paper-based assessment titled the Steadiness Index. The latter is a series of 3 questions assessing steadiness when walking, standing, or getting up from a chair. Those participants of the TILDA cohort passing the inclusion criteria were divided into those who reported a fall in the previous year (N=1159) and those reporting no falls (N=4746). Two group comparisons were analyzed by Mann-Whitney U Test (p<0.05) and a Receiver Operator Characteristics (ROC) curve analysis was used to detect separation of fall and non-fall groups. For the Mann-Whitney U test the fall and no fall groups were statistically different for the T-TUG (p<0.001), TUG (p<0.001), dominant and nondominant grip forces (p<0.001), and the steadiness index (p< 0.001). In the fall group, the grip forces were weaker, T-TUG and TUG time longer, and the steadiness index scores lower. For the grip force assessments and steadiness index, lower scores are more likely to be associated with a higher fall risk. In the T-TUG and TUG, longer times are more likely to be associated with a higher fall risk. In the ROC curve analyses, the T-TUG (0.567, p<0.001) demonstrated similar outcomes compared to dominant (AUC=547, p<0.09) and non-dominant (AUC=0.550, p<0.01) grip forces, and the TUG (AUC=0.558, p<0.001). The steadiness index ROC analysis was slightly better than the T-TUG (AUC=0.579, p<0.001). Sensitivity (52-58%) and specificity (50-57%) ranges were equivalent for all performance-based assessments, whereas for the Steadiness Index, the sensitivity (40%) was lower than the specificity (75%). The EMR fall-risk calculator (T-TUG) is a valid triage tool to estimate fall risk in older community dwellers. The EMR calculator has the potential for real-time assessment of patients using current data compared to other performance- and paper-based assessments, which would allow the healthcare team to spend more time with higher fall risk patients.

Learning representations of medical concepts from the Electronic Health Records (EHRs) has been shown effective for predictive analytics in healthcare. The learned representations are expected to preserve the semantic meanings of different medical concepts, which can be treated as features and thus benefit a variety of applications. Medical ontologies have also been explored to be integrated with the EHR data to further enhance the accuracy of various prediction tasks in healthcare. Most of the existing works assume that medical concepts under the same ontological category should share similar representations, which however does not always hold. In particular, the categorizations in the categorical medical ontologies were established with various factors being considered. Medical concepts even under the same ontological category may not follow similar occurrence patterns in the EHR data, leading to contradicting objectives for the representation learning. In addition, these existing works merely utilize the categorical ontologies. Actually, it has been noticed that ontologies containing multiple types of relations are also available. However, studies rarely make use of the diverse types of medical ontologies. In this thesis research, we propose three novel representation learning models for integrating the EHR data and medical ontologies for predictive analytics. To improve the interpretability and alleviate the conflicting objective issue between the EHR data and medical ontologies, we propose techniques to learn medical concepts embeddings with multiple ontological representations. To reduce the reliance on labeled data, we treat the co-occurrence statistics of clinical events as additional training signals, which help us learn good representations even with few labeled data. To leverage the various domain knowledge, we also consider multiple medical ontologies (CCS, ATC and SNOMED-CT) and propose corresponding attention mechanisms so as to take the best advantage of the medical ontologies with better interpretability. Our proposed models can achieve the final medical concept representations which align better with the EHR data. We conduct extensive experiments, and our empirical results prove the effectiveness of the proposed methods. Keywords: Bio/Medicine, Healthcare-AI, Electronic Health Record, Representation Learning, Machine Learning Applications

A personal health record (PHR) allows a patient to exert control over his/her healthcare by enhancing communication with healthcare providers. According to research, patients find value in having access to information contained in their medical records. Often a glossary is required to aid in interpreting the information and understanding the content. However, giving patients the ability to speak with providers about their medical conditions empowers them to participate as informed healthcare consumers. The majority of patients (75%) at Medical Specialists expressed their intention to adopt the PHR if it is made available to them. Although the perceived usefulness of a PHR was a significant determining factor, comfort level with technology, health literacy, and socioeconomic status were indirectly related to intention to adopt as well. Perceived health status was not found to be a significant factor in this population for determining intention to adopt a PHR. The majority of patients in each category of gender, age, marital status, and race/ethnicity (except American Indian/Alaska Native) expressed interest in adopting a PHR, with most categories being above 70%. Findings indicate a broad acceptance of this new technology by the patients of Medical Specialists. Improvement of adoption and use rates may depend on availability of office staff for hands-on training as well as assistance with interpretation of medical information. Hopefully, over time technology barriers will disappear, and usefulness of the information will promote increased demand.

Structured electronic health record (EHR) data are commonly incomplete and can lack diagnostic detail. Clinical reports, on the other hand, are typically comprehensive and contain a wealth of detailed medical information. Pathologists invest considerable time and specialized training to create information-rich pathology reports, but the necessary manual review of these reports for clinical or research use is a high barrier to their routine utilization. The automated extraction of clinical targets directly from pathology reports would allow for the structured aggregation of relevant patient data that are not currently routinely captured in the EHR. In this dissertation, I apply recently developed transformer models to predict clinical targets from cancer pathology report text. In the first chapter, I present a pathology report corpus that I fully processed and made publicly available, and perform a proof-of-concept cancer type classification. In the second chapter, I discuss a set of cancer stage classification models that I fine-tune on the pathology report corpus and then externally validate on reports from Columbia University Irving Medical Center (CUIMC). In the last chapter, I explore additional applications for this methodology, developing a generalizable model to classify prostate cancer reports into primary Gleason score categories, applying a transformer model to classify reports into diagnosis categories for a Barrett’s esophagus patient cohort in a low-data environment, and performing a proof-of-concept prediction of adverse drug events from 1D drug representations.

No description available.