Scholarship & Research
Permanent URI for this communityhttps://scholarworks.montana.edu/handle/1/1
Browse
3 results
Search Results
Item Increasing recommended testing compliance for persons with type II diabetes in primary care(Montana State University - Bozeman, College of Nursing, 2024) Fleming, Brandi Lynn; Chairperson, Graduate Committee: Elizabeth A. Johnson; This is a manuscript style paper that includes co-authored chapters.Background: Type II diabetes affects one in 14 Montanans (Centers for Disease Control and Prevention (CDC), 2023). The CDC estimates annual direct and indirect costs of diabetes in Montana exceed $800 million (2023). Constraints persist when incorporating National Quality Forum measures and Healthy People 2030 objective guidance to address known challenges in managing Type II diabetes in a community setting due to minimal resources and lack of workflow appraisal. The rurality and radical weather patterns in Montana pose challenges for sustaining healthy diets and regular exercise. Purpose: The quality improvement project aims at generating consistent clinical decision support system (CDSS) electronic health record platform (EHR) reminders, streamlining workflow processes, and delaying Type II diabetes' concomitant conditions. Methods: A Plan-Do-Study-Act (PDSA) cycle employing Amazing Charts EHR to consistent clinical decision support system reminders, workflow process modification, and shared decision-making interventions. Purposive sampling included persons with Type II diabetes, 18-75 years, presenting for an annual visit type encounter. Interventions: Rule query preference entry and workflow process modification were monitored to a short-term goal benchmark of 90% for completion of recommended testing for persons with Type II diabetes. Data collection evaluated generation of CDSS reminders and annual completion of comprehensive foot examinations, urine microalbumin to creatinine ratio testing, and dilated eye examinations. Results: A total of six patients participated in the project, n = 5 met criteria for Type II diabetes diagnosis, n = 1 miscoded. The EHR generated CDSS reminders, and staff completed annual comprehensive foot examinations 83.33% of eligible encounters. Urine microalbumin testing was completed 66.63% of eligible encounters with n = 1 (16.33%) deferred testing until their annual visit. Strengths emerging from Strengths, Weakness, Opportunities, and Threats (SWOT) analysis included simple streamlined guidelines that promote teamwork. Conclusion: Consistent CDSS reminder facilitates recommendation completion, benefiting patients and providers. Although short term goals were not achieved at the 90% benchmark, the project is deemed clinically significant representing the homogeneity of Montanans. Future recommendations include participation in Merit-based Incentive Payment System (MIPS), extension of interventions for utilization of other chronic diseases, and integration of Current Procedural Terminology (CPT) codes for reimbursement for services.Item Quanitfying snow depth distributions and spatial variability in complex mountain terrain(Montana State University - Bozeman, College of Letters & Science, 2021) Miller, Zachary Stephen; Chairperson, Graduate Committee: Eric A. SprolesThe spatial variability of snow depth is a major source of uncertainty in avalanche and hydrologic forecasting. Identification of spatial and temporal patterns in snow depth is further complicated by the interactions of complex mountain topography and localized micro-meteorology. Recent studies have dramatically improved our understanding of snow depth spatial variability by utilizing increasingly accessible remote sensing technologies such as satellite imagery, terrestrial laser scanning, airborne laser scanning and uninhabited aerial systems (UAS) to map spatially continuous snow depths over a variety of spatiotemporal scales. However, much of this work focuses on relatively low-relief topographies or limited temporal frequencies. Our research presents a thorough evaluation of the evolution of snow depth spatial variability at the slope scale in steep complex mountain terrain (45.834 N, -110.935 E) using analysis from UAS imagery. We apply 13 spatially complete UAS-derived snow depth datasets collected throughout the course of the 2019/2020 winter to analyze spatial and temporal patterns of snow depth and snow depth change variability. Our results show greater spatial variability in steep complex mountain terrain than an adjacent mountain meadow both in the seasonal context and during individual meteorological periods. We analyze 2 cm horizontal resolution snow depth models by (i) comparing spatial patterns with coincident meteorological data, (ii) analysis of the temporal elevation specific patterns of snow depth, and (iii) a comprehensive multi-scalar evaluation of spatial variability. We quantify the unique spatial signature of four specific events: a major snow accumulation, a natural avalanche, a calm period, and a significant wind event. We find a non-linear relationship between elevation and snow depth, with upper elevations proving to be the most variable. We also verify that significant storm events result in the largest snow depth change variability throughout our study area, as compared to other meteorological events. The synthesis of these findings illustrate the dynamic spatial and temporal snow depth distribution patterns observed in complex mountain terrain during the course of a winter season. These findings are relevant to avalanche forecasters and researchers, snow hydrologists and local water resource managers, and downstream communities dependent on snow as a hydrologic reservoir.Item Extracting abstract spatio-temporal features of weather phenomena for autoencoder transfer learning(Montana State University - Bozeman, College of Engineering, 2020) McAllister, Richard Arthur; Chairperson, Graduate Committee: John SheppardIn this dissertation we develop ways to discover encodings within autoencoders that can be used to exchange information among neural network models. We begin by verifying that autoencoders can be used to make predictions in the meteorological domain, specifically for wind vector determination. We use unsupervised pre-training of stacked autoencoders to construct multilayer perceptrons to accomplish this task. We then discuss the role of our approach as an important step in positioning Empirical Weather Prediction as a viable alternative to Numerical Weather Prediction. We continue by exploring the spatial extensibility of the previously developed models, observing that different areas in the atmosphere may be influenced unique forces. We use stacked autoencoders to generalize across an area of the atmosphere, expanding the application of networks trained in one area to the surrounding areas. As a prelude to exploring transfer learning, we demonstrate that a stacked autoencoder is capable of capturing knowledge universal to these dataspaces. Following this we observe that in extremely large dataspaces, a single neural network covering that space may not be effective, and generating large numbers of deep neural networks is not feasible. Using functional data analysis and spatial statistics we analyze deep networks trained from stacked autoencoders in a spatiotemporal application area to determine the extent to which knowledge can be transferred to similar regions. Our results indicate high likelihood that spatial correlation can be exploited if it can be identified prior to training. We then observe that artificial neural networks, being essentially black-box processes, would benefit by having effective methods for preserving knowledge for successive generations of training. We develop an approach to preserving knowledge encoded in the hidden layers of several ANN's and collect this knowledge in networks that more effectively make predictions over subdivisions of the entire dataspace. We show that this method has an accuracy advantage over the single-network approach. We extend the previously developed methodology, adding a non-parametric method for determining transferrable encoded knowledge. We also analyze new datasets, focusing on the ability for models trained in this fashion to be transferred to operating on other storms.