17179 research outputs found
Sort by
A Machine Learning Approach to Predict Muscle Fatigue Among Novice Roofers
No full textRoofing is an exceptionally high-risk occupation where severe ergonomic demands lead to Work-Related Musculoskeletal Disorders driven by muscle fatigue. Traditional ergonomic assessments are static and subjective, failing to capture a worker's dynamic physiological state. This thesis addresses critical gaps by developing generalizable fatigue detection models for varied tasks and introducing a quantitative metric for systemic, whole-body fatigue.
This study used a multi-modal sensor system (12-Cometa surface electromyography (sEMG), 17-Xsens Awinda inertial measurement unit (IMU)) to collect data from novice participants performing simulated roofing postures (standing, stooping, kneeling). A two-stage methodology using Weak Monotonicity (WM) trend analysis generated high-confidence fatigue labels from sEMG data. A suite of machine learning models, including deep learning (CNN-LSTM-Attention) and novel hybrid architectures, were evaluated using a Leave-One-Subject-Out cross-validation.
Results show a generalized CNN-LSTM-Attention model achieved 79.5% accuracy for localized fatigue classification across all postures. For systemic fatigue, this study introduces the Multi-Muscle Fatigue Score (MMFS), a novel 0-12 scale index. IMU-only regression models effectively predicted this score, achieving a Mean Absolute Error of approximately 1.63.
This research concludes that sEMG-based models are superior for localized fatigue classification, while IMU-only models offer a practical strategy for assessing systemic fatigue. The primary contributions a comparative analysis of specialized versus generalized models, and the evaluation of novel hybrid architectures.Engineering Technolog
Expert Consensus on the Nutrition Care Process in Guatemalan Hospitals: Findings from a Delphi Study of nutritionDay 2022 Participants
No full textBackground: Disease-related malnutrition (DRM) remains an underdiagnosed condition in Latin American hospitals, with substantial clinical and economic consequences. The global nutritionDay initiative, promoted by ESPEN, provides a standardized audit to evaluate and improve hospital nutritional care. This study aimed to develop expert consensus recommendations to optimize the nutritional care process in Guatemalan hospitals, based on the findings from nutritionDay 2022. Methods: A modified Delphi study was conducted, including three meetings held before and after each round to discuss the results. Sixteen clinical nutrition professionals from eleven hospitals participated. A total of 89 items were assessed, with a predefined consensus threshold of 70%. Data were analyzed using R software (version 4.5.0) and Kendall’s W coefficient was applied to evaluate inter-round agreement. Results: Consensus was achieved for 51 key recommendations covering nutritional screening, clinical assessment, anthropometry, body composition, functional assessment, biochemical monitoring, dietary intervention, and post-discharge follow-up. The proposed actions are aligned with international guidelines (ESPEN, ASPEN, GLIM) and adapted to the Guatemalan healthcare context. Conclusions: This consensus provides a comprehensive and context-specific framework for standardizing and improving hospital nutritional care in Guatemala and similar settings. Its implementation could help reduce DRM prevalence and foster the development of quality indicators and digital tools for clinical nutrition management.Division of Researc
Macrophyte Community Response to Restoration Efforts, Flow Conditions, and Recreational Pressure in the San Marcos River
No full textAquatic macrophytes play a vital ecological role in spring-fed river ecosystems, yet native submerged aquatic vegetation (SAV) in the San Marcos River, Texas, is threatened by invasive species, drought conditions, and recreational pressures. Since 2013, a full-scale habitat restoration program has aimed to reduce non-native SAV and promote recovery of federally endangered species, including Zizania texana and Etheostoma fonticola. This study evaluated restoration effectiveness using over a decade of biomonitoring data and spatial modeling, with a focus on species-level responses and environmental drivers. SAV coverage was assessed using a repeated grid-cell framework at two scales: full-system maps of the upper 2.4 km of river at 5-year intervals and annual Long-Term Biological Goal (LTBG) reach subsets. Restoration data were paired with channel morphology variables derived from high-resolution imagery. Hydrologic variation was also incorporated to account for flow fluctuations and recreation-associated disturbance. Bayesian and spatial mixed models were used to estimate cover change and diversity responses, aimed to provide insights that would enhance future management protocols. Results showed that non-native SAV removal effort was significantly associated with reductions in invasive coverage, though species-specific responses varied. Planting efforts produced mixed outcomes, with some native species showing limited success. Diversity patterns were related to spatial position in the channel, with higher diversity further from banks and access points, but diversity didn’t change much over time, suggesting more stable communities in those areas. While river morphology and environmental conditions showed some influence, limitations in monitoring design and available data likely reduced model performance. Overall, this research highlights how long-term, adaptive management has contributed to a major shift toward native SAV in the San Marcos River. Though challenges remain in using biomonitoring data for predictive modeling, the findings offer a spatially explicit approach that can inform future restoration planning. By using accessible and repeatable metrics, this study provides a practical framework for refining restoration strategies under the Edwards Aquifer Habitat Conservation Plan as conditions continue to change.Biolog
Consistency of Restricted Maximum Likelihood Estimators in High-Dimensional Kernel Linear Mixed-Effects Models with Applications in Estimating Genetic Heritability
No full textRestricted maximum likelihood (REML) estimators are commonly used to obtain unbiased estimators for the variance components in linear mixed models. In modern applications, particularly in genomic studies, the dimension of the design matrix with respect to the random effects can be high. Motivated by this, we first introduce high-dimensional kernel linear mixed models, derive the REML equations, and establish theoretical results on the consistency of REML estimators for several commonly used kernel matrices. The validity of the theories is demonstrated via simulation studies. Our results provide rigorous justification for the consistency of REML estimators in high-dimensional kernel linear mixed models and offer insights into the application of estimating genetic heritability. Finally, we apply the kernel linear mixed models to estimate genetic heritability in a real-world data application.Mathematic
Roadmap to Sustainability in Texas Wine
No full textThe Fermented Landscapes Lab at Texas State University, led by Dr. Colleen Myles, with significant contributions from Matt Roberts, Claryce Bosson, Mik Lopez, and Nicole Anguiano, has been actively engaged in a collaborative effort with Texas Hill Country winegrowers and winemakers for a few years now. In this “Roadmap to Sustainability in Texas Wine” project, the fermented landscape concept is utilized to better analyze sustainability efforts in the Texas wine industry, with the goal of understanding how to effectively address the multifaceted challenges and opportunities related to sustainability within the region.
Driven by industry partners’ motivations and armed with this deeper understanding of regional contexts within the Hill Country wine industry, collaborative efforts shifted towards establishment of a feasible sustainable framework for Texas wine. The end of this portion of the “Roadmap to Sustainability in Texas Wine” project has culminated to the submission of a proposal for an official sustainability certification for Texas vineyards and winemakers; attached in the appendix under Roadmap_Texas Proposal of SIP.pdf. However, any project striving towards sustainability is an ongoing process and this does not mark the end of this collaboration between Texas State University and Hill Country wine entities, as the Fermented Landscapes Lab continues to work closely with the local industry through various projects and sustainability prerogatives.Geography and Environmental Studie
SPARK! at TX State: A High-Impact Service Internship
No full textThis presentation discusses the growth outcomes of SPARK! for German program offered to German majors in the World Languages and Literatures Department. In this internship course juniors and seniors receive essential training in language pedagogy and lesson planning and go on to offer German lessons to junior-high children in Central Texas. The internship integrates classroom and experiential learning and combines features of traditional internships and service learning. SPARK! for German interns increase their self-efficacy, develope socially responsible leadership skills, and build social capital within the Central Texas community.Conference travel was made possible by the generosity of the Goethe InstitutWorld Languages and Literature
Measuring the Impact of Education on Waste Streams from Multi-Family Housing
No full textFood waste is a significant global issue with substantial environmental, economic, and social implications. This exploratory study aimed to evaluate the impact of an educational composting program on reducing food waste generation and promoting proper waste sorting practices within multi-family housing units in San Marcos, Texas. A comprehensive methodology was employed, encompassing pre- and post-intervention waste audits, educational interventions, weekly organic waste collection, and quantitative data analyses. Nine multi-family complexes, spanning student housing, conventional family units, low-income residences, and senior living facilities, were targeted through strategic recruitment efforts and incentivization. The treatment group, consisting of 43 participants, received ongoing education throughout the eight-week implementation period, facilitated through informational resources, feedback mechanisms, and door-to-door organic waste collection. Conversely, the control group did not partake in the educational component. Statistical analyses, including descriptive statistics and paired t-tests, facilitated comparisons across various dimensions, such as housing types, treatment versus control groups, and pre- versus post-intervention periods. The findings revealed significant reductions in organic waste and compostable materials within the treatment group’s weekly landfill trash, underscoring the effectiveness of the educational program. Furthermore, insights into contamination patterns and housing-specific waste characteristics were garnered, informing targeted intervention strategies and policy recommendations for optimizing multi-family composting initiatives.Agricultural Science
A Broader View of Conservation: Mapping Nature and Culture of Working Lands in the Texas Hill Country
No full textWorking lands are (agri)cultural landscapes with natural value that provide tangible and intangible benefits to people. The development and loss of working landscapes and other rural lands have been increasing, especially in regions experiencing rapid urban growth. To address these issues and conservation needs, we conducted a multi-scale study that involved (1) an in-depth case study of a working ranch focused on conservation, and (2) a geospatial inventory of ecosystem services and cultural values across a Central Texas county that is still dominated by working landscapes yet is experiencing increasing urbanization. The purpose of these two complementary scales was to add a broader view of working landscape values by mapping the multiple benefits they provide and assessing where provisioning, regulating, and cultural benefits overlap. The ultimate objective of our study was to identify target zones for conservation planning. Based on categorical and spatial variation in the mapped benefits inventory, we identified four conservation target zones: Headwaters, Historic Towns, Geologic Significance, and Tourism Corridor. Finally, we framed these conservation target zones within the context of land use decisions and tradeoff analyses.Geography and Environmental Studie
Oral history interview: Frank Breedlove
No full textTranscript file (.pdf) and closed captioning available.Interview with Veteran Frank Breedlove’s son reflecting on his father’s military service and Veteran life
From Triads to Tools: A Comprehensive Review of the Expanding Roles of G-Triplex Structures
No full textInterest in non-canonical DNA structures continues to grow, in part fueled by the recent discovery of a new structure, G-triplex DNA. Originally proposed as folding intermediates for G-quadruplex DNA, G-triplex DNA has more recently been shown to form from truncated G-quadruplex sequence oligonucleotides and other, specifically designed sequences. In this review, we provide the first, comprehensive survey of G-triplex DNA and RNA, covering the literature up to 2024. We include reports of G-triplex DNA from bulk solution and single-molecule approaches, the structural characterization of G-triplex DNA, and the breadth of oligonucleotide sequences that have been reported to form these structures. The formation of G-triplex RNA structures is also reviewed. The evolving understanding of sequence and environmental effects on G-triplex formation are presented together with challenges due to structural polymorphism and competing formation of multimeric G-quadruplex structures. Hints of the biological relevance of G-triplexes are provided by reports of protein recognition of these structures and their effects on DNA replication in vitro. Interaction of G-triplex DNA with a variety of ligands has been reported, although the search for selective ligands that can distinguish G-triplex from G-quadruplex is on-going. The vast majority of publications in the area have focused on the utilization of G-triplex in biosensing applications, which has shown some advantages compared to G-quadruplex-based systems. These results highlight the potential utility of G-triplex structures in a variety of domains and show its promise in applications in biotechnology, medicine, and research.Chemistry and BiochemistryMaterials Science, Engineering, and Commercializatio