Scholarworks
ScholarWorks is an open access repository for the capture of the intellectual work of Montana State University (MSU) in support of its teaching, research and service missions. MSU ScholarWorks is a central point of discovery for accessing, collecting, sharing, preserving, and distributing knowledge to the Montana State University community and the world.
Communities in DSpace
Select a community to browse its collections.
Recent Submissions
Machine learning surrogates for surface complexation model of uranium sorption to oxides
(Springer Science and Business Media LLC, 2024-03) Li, Chunhui; Adeniyi, Elijah O.; Zarzycki, Piotr
The safety assessments of the geological storage of spent nuclear fuel require understanding the underground radionuclide mobility in case of a leakage from multi-barrier canisters. Uranium, the most common radionuclide in non-reprocessed spent nuclear fuels, is immobile in reduced form (U(IV) and highly mobile in an oxidized state (U(VI)). The latter form is considered one of the most dangerous environmental threats in the safety assessments of spent nuclear fuel repositories. The sorption of uranium to mineral surfaces surrounding the repository limits their mobility. We quantify uranium sorption using surface complexation models (SCMs). Unfortunately, numerical SCM solvers often encounter convergence problems due to the complex nature of convoluted equations and correlations between model parameters. This study explored two machine learning surrogates for the 2-pK Triple Layer Model of uranium retention by oxide surfaces if released as U(IV) in the oxidizing conditions: random forest regressor and deep neural networks. Our surrogate models, particularly DNN, accurately reproduce SCM model predictions at a fraction of the computational cost without any convergence issues. The safety assessment of spent fuel repositories, specifically the migration of leaked radioactive waste, will benefit from having ultrafast AI/ML surrogates for the computationally expensive sorption models that can be easily incorporated into larger-scale contaminant migration models. One such model is presented here.
Comparison of Supervised Learning and Changepoint Detection for Insect Detection in Lidar Data
(MDPI AG, 2023-12) Vannoy, Trevor C.; Sweeney, Nathaniel B.; Shaw, Joseph A.; Whitaker, Bradley M.
Concerns about decreases in insect population and biodiversity, in addition to the need for monitoring insects in agriculture and disease control, have led to an increased need for automated, non-invasive monitoring techniques. To this end, entomological lidar systems have been developed and successfully used for detecting and classifying insects. However, the data produced by these lidar systems create several problems from a data analysis standpoint: the data can contain millions of observations, very few observations contain insects, and the background environment is non-stationary. This study compares the insect-detection performance of various supervised machine learning and unsupervised changepoint detection algorithms and provides commentary on the relative strengths of each method. We found that the supervised methods generally perform better than the changepoint detection methods, at the cost of needing labeled data. The supervised learning method with the highest Matthew’s Correlation Coefficient score on the testing set correctly identified 99.5% of the insect-containing images and 83.7% of the non-insect images; similarly, the best changepoint detection method correctly identified 83.2% of the insect-containing images and 84.2% of the non-insect images. Our results show that both types of methods can reduce the need for manual data analysis.
Neural network‑based strategies for automatically diagnosing of COVID‑19 from X‑ray images utilizing different feature extraction algorithms
(Springer Nature, 2023-07) Siddiqi Prity, Farida; Nath, Nishu; Nath, Antara; Aslam Uddin, K. M.
The COVID-19 pandemic has had an obliterating impact on the health and well-being of the worldwide populace. It has recently become one of the most severe and acute diseases and has spread globally. Therefore, an automated detection system should be implemented as the fastest diagnostic option to control the spread of COVID-19. This study aims to introduce neural network-based strategies for classifying and detecting COVID-19 early through image processing utilizing X-ray images. Despite the extensive use of directly fed X-ray images into the classifier, there is a lack of comparative analysis between the feature-based system and the direct imaging system in the classification of COVID-19 to evaluate the efficiency of feature extraction methods. Therefore, the proposed system represents introductory experiments of feature-based system using image Feature Extraction Algorithms [Texture, Grey-Level Co-occurrence Matrix (GLCM), Grey-Level Dependence Matrix (GLDM), Fast Fourier Transform (FFT), and Discrete Wavelet Transform (DWT)], Dimensionality Reduction Algorithms (Principal Component Analysis (PCA), Kernel Principal Component Analysis (KPCA), Sparse Autoencoder, and Stacked Autoencoder), Feature Selection Algorithms (Anova F-measure, Chi-square Test, and Random Forest), and Neural Networks [Feed-Forward Neural Network (FFNN) and Convolutional Neural Network (CNN)] on generated datasets consisting of Normal, COVID-19, and Pneumonia-infected chest X-ray images. Tenfold Cross-Validations are implemented during the classification process. A baseline model is created where no Feature Extraction Algorithm is implemented. Accuracy, sensitivity, specificity, precision, and F-measure metrics are utilized to assess classification performance. Finally, a comprehensive comparison of the success of the Feature-based system (Feature Extraction Algorithms, Dimensionality Reduction Algorithms, and Feature Selection Algorithms) in COVID-19 classification from X-ray images is made with the baseline model. The highest classification accuracy (95.44 ± 0.03%), sensitivity (96%), specificity (98%), precision (96%), and F-measure (96%) are achieved for Feature-based systems using Principal Component Analysis. The aim is to lay the foundation for the potential creation of a system that can automatically distinguish COVID-19 infection based on chest X-ray images using the Feature-based model.
An overview and policy implications of national nurse identifier systems: A call for unity and integration
(Elsevier BV, 2023-03) Chan, Garrett K.; Cummins, Mollie R.; Taylor, Cheryl S.; Rambur, Betty; Auerbach, David I.; Meadows-Oliver, Mikki; Cooke, Cindy; Turek, Emily A.; Pittman, Patricia (Polly)
There is a clear and growing need to be able record and track the contributions of individual registered nurses (RNs) to patient care and patient care outcomes in the US and also understand the state of the nursing workforce. The National Academies of Sciences, Engineering, and Medicine report, The Future of Nursing 2020–2030: Charting a Path to Achieve Health Equity (2021), identified the need to track nurses’ collective and individual contributions to patient care outcomes. This capability depends upon the adoption of a unique nurse identifier and its implementation within electronic health records. Additionally, there is a need to understand the nature and characteristics of the overall nursing workforce including supply and demand, turnover, attrition, credentialing, and geographic areas of practice. This need for data to support workforce studies and planning is dependent upon comprehensive databases describing the nursing workforce, with unique nurse identification to support linkage across data sources. There are two existing national nurse identifiers– the National Provider Identifier and the National Council of State Boards of Nursing Identifier. This article provides an overview of these two national nurse identifiers; reviews three databases that are not nurse specific to understand lessons learned in the development of those databases; and discusses the ethical, legal, social, diversity, equity, and inclusion implications of a unique nurse identifier.
Development of the Blackfeet Community Hospice Project: Pilot Workshop
(SAGE Publications, 2022-08) Colclough, Yoshiko; Brown, Gary M.
Taboo perception on talking about death and dying among American Indians/Alaska Natives is prevalent. This suppressive value makes hospice introduction difficult, leading hospice disparity. Working together by using a community-based participatory research approach over a decade, we conducted a 6-hour workshop including information sharing and group activities. The purpose of the study was to investigate the community readiness for end-of-life knowledge by conducting a public workshop. We used pre- and post-workshop surveys with Likert-type responses to five questions to assess the effect of workshop in end-of-life knowledge. Thirty individuals participated the workshop; 80% of them reported their knowledge increase on at least one question. While the survey had concerns, positive participant responses indicated readiness and appropriateness to use workshops to increase end-of-life knowledge.