26400 research outputs found
Sort by
Integrating Machine Learning And Simulation For Resource Planning Of Hospital Systems Based On Predicted Length Of Stay
No full textRecently Hospital Systems faced a high invasion of patients generated by several events such as health crisis related epidemic (COVID, FLU) or seasonal flows. Hence, managing hospital bed availability and efficiency with proper care is obligatory for addressing the challenges associated with the overburden of patients. However, the Length of stay (LOS) is often increased due to the high patient influx and overcrowding problem occurs within the Hospital. It resolves these issues, it is essential for hospital authority to predict the Patients LOS which is the crucial indicator for the use of medical resources (allocation, utilization of providers and resource) and assessing the overcrowding within the hospital premises. Thus, accurate LOS and proper resource management is indispensable for hospital authority to ensure the maximum profit with optimized system utilization. This Study proposes a Machine Learning driven approach integrated with Simulation Software for the prediction of LOS and resource management within the hospital System. Artificial Neural Network is used for predicting the LOS in the simulation environment that learns the pertinent information from nonlinear and linear processes without prior assumption on data distribution and substantially boosts the prediction accuracy. Dynamic resources Planning is also integrated within this model that allows the management to make plans on the very first day for hospital resources based on Predicted LOS for the next few days as required. Most importantly, in this proposed Machine Learning Based Data-driven simulation method, Patients are generated randomly from a Dynamic simulation environment with different disease attributes and the simulation Predicts the LOS as per trained model in Brain. Based on this LOS, authorities can make decisions regarding the Bed allocation, Doctors requirements and Patients satisfactions for proper treatment in the hospital System. Hence this model significantly helps healthcare professionals and patients care to manage their resources and increases the patientâ??s satisfaction allowing early treatment and dynamic resource planning alongside with finance
Parametric Optimization Of Collaborative Robots For Pick-And-Place And Assembly Operations In Manufacturing Industries
No full textCollaborative robots or â??Cobotsâ?? are a type of robot that works in the same workspace as a human operator to assist in repetitive, hazardous tasks with accuracy and efficiency under the supervision of a human operator. In this age of Industrial 4.0, with the remarkable development of robotics, collaborative robots are introduced into smart manufacturing and industrial applications. Cobots are incorporated into several industries such as automotive, aerospace, electronics, technology, medical and cosmetics, manufacturing, etc. Cobots that work in a fixed coordinate system, are faster, and more productive. However, for industries with a variety of items, it is not cost-effective or feasible to produce fixtures and molds for each item and its sub-items. Hence, the vision system will provide more flexibility to the system. This study aims to work on the vision system to determine the optimum conditions for the Cobot to work effectively. The experiment was carried out by utilizing a setup with a UR3e Cobot with a ROBOTIQ Adaptive Gripper (end-effector), a ROBOTIQ wrist camera, an Amprobe Digital Light meter and a reflectance measurement setup with an integrating sphere. The factors of the experiment are color, and shape of the parts, ambient light and snapshot position (scanning height). This research will help to create the pathway for cobots usage in manufacturing and reduce its inconsistencies to improve the collaboration with a human operator to work simultaneously for applications like pick-and-place and assembly operations
Using Cognitive Dual-Task to Evaluate Individuals With Mild Traumatic Brain Injuries: A Pilot Study
Get PDFBackground: Traumatic Brain Injury (TBI) presents significant challenges, with cognitive and motor deficits often persisting long after the initial injury. Current assessment methods may fail to detect subtle impairments, highlighting the need for more comprehensive evaluation techniques. This pilot study aims to investigate the efficacy of cognitive dual-task assessments in detecting subtle gait deficits associated with concussion/mild Traumatic Brain Injury (c/mTBI), with a focus on identifying which cognitive dual task would provide the most cognitive cost and afford motion capture sensors the ability to detect subtle gait changes.
Methods: Twenty-two participants without prior concussion or mTBI history underwent dual-task assessments combining normal and tandem walking with verbal and arithmetic cognitive tasks. Gait parameters were recorded using motion capture sensors, and cognitive costs were calculated to evaluate task performance.
Results: Differences were observed in gait speed between single task and dual-task tandem walking conditions noted by weak correlation. However, cognitive tasks (verbal vs. arithmetic) did not significantly impact cognitive costs when combined with tandem walking. Motion capture sensors exhibited limitations in detecting differences between cognitive and motor tasks.
Conclusions: While tandem walking dual-task conditions revealed more pronounced deficits in gait speed, cognitive tasks did not significantly affect cognitive costs of tandem walking. Motion capture sensors demonstrated limitations in detecting subtle gait deficits associated with c/mTBI. Sample size constraints, potential learning effects, sensor placement variability, and human variability in task prioritization were identified as limitations. Future research should focus on expanding sample sizes, refining sensor technology, and exploring the interplay between cognitive tasks and motor performance to enhance sensitivity for detecting subtle gait deficits in c/mTBI populations
Body Parts in Spanish - I Have... Who Has?
No full textThe present activity serves as a follow-up to a general review of body parts in Spanish; the activity includes inner (organs) and outer body parts.https://scholarworks.utep.edu/humanities_tools/1007/thumbnail.jp
Microwave-Assisted Pyrolysis of Hydrocarbons Using Iron-Based Alumina Catalysts Obtained via Solution Combustion Synthesis
No full textThe demand for hydrogen is growing which makes the development of clean and efficient H2 synthesis technologies imperative. Microwave-assisted, thermocatalytic, dehydrogenation of hydrocarbons has demonstrated the ability to generate H2 with high yield/selectivity and leaving behind valuable solid carbon byproducts. However, this microwave-assisted process is unoptimized which prevents it from being utilized in industry. A critical component of optimization is the development of a catalyst that is catalytically active, a good microwave absorber, and can be regenerated for repeated dehydrogenation cycles. Previous studies that focused on plastic waste decomposition have used iron-based alumina (FeAlxOy) made via solution combustion synthesis (SCS). Unexplored is the effect of tuning SCS parameters on dehydrogenation performance, the use of these materials in hydrocarbon decomposition to H2, and the regeneration of these catalysts. This dissertation has three objectives: (1) characterize the relationship between SCS parameters and the material properties of FeAlxOy, (2) determine how differences in the material properties of FeAlxOy influence their performance as catalysts during microwave-assisted pyrolysis of fossil fuels, and (3) investigate the Boudouard reaction to regenerate the FeAlxOy post-dehydrogenation
Mechanical Testing Of Laser Beam Metal Powder Bed Fusion Tensile Specimens: Insights Of Factors Affecting Part Quality And Material Properties
No full textAdditive manufacturing (AM) processes include Laser-Based Powder Bed Fusion of Metals (PBF-LB/M), which enables the making of complex metal parts, making it an attractive prospect to industries such as aerospace and automotive. However, PBF-LB/M faces challenges regarding defect detection and part quality inconsistencies, often due to hidden parameters such as polygon delay that impact component quality beyond the standard quality control capabilities. Traditional PBF-LB/M qualification processes use machined specimens, including tensile, fatigue, and chemical coupons; these usually cannot pick up key features like scanning control errors, surface roughness, and geometric accuracy.
This thesis explores using as built tensile coupons from the GTADExP program’s quality test artifact (QTA) to address these limitations. As built coupons maintain inherent surface roughness and near-edge defects, better representing process-induced variations with lower cost input. This is supported by the specific tensile testing methodology used in this research on cylindrical-shaped sub-size samples by emphasizing 2.5 mm gauge diameter condition according to ASTM E8 Type 5 and 4 mm according to ASTM E8 Type 4. Examples of analyses included but were not limited to ANOVAs on YS, UTS, and strain at break to assess what process variables affect the mechanical properties of the tensile coupons.
Lack of fusion (LoF) defects present within UNS N07718 QTAs (UTEP 18.01 and UTEP 19.01) resulted in a strain at break reduction of 5% for UTEP 18.01, indicating ductility was reduced as a consequence of laser power ramping; however, YS and UTS values did not significantly vary between QTAs, suggesting that tensile testing can detect defects but lacks the specificity to identify them as LoF defects in PBF-LB/M coupons. Further experiments were conducted with Sc-modified AlSi alloy coupons in fractional factorial design experiments (DOE). For non-heat-treated samples, the most influencing factors in tensile properties were the hatch distance and laser power. A power setting of 370 W yielded an ultimate tensile strength (UTS) of 394 ± 11.7 MPa, in contrast to the UTS of 382 ± 15.2 MPa observed at a power setting of 300 W. The yield strength (YS) and elongation parameters were found to be influenced by the hatch distance. Specifically, a hatch distance of 0.06 mm resulted in a YS of 300.4 ± 11.9 MPa and an elongation of 25.7 ± 2.6%. Conversely, a hatch distance of 0.12 mm produced a YS of 287.8 ± 8.6 MPa and an elongation of 21.6 ± 3.3%.
In the published work presented in Chapter 4, the applicability of tensile testing towards machine qualification is evaluated via a full factorial experiment examining coupon size variation, surface roughness variation, and scanner control variation. Machined and as built conditions, standard-size coupons ASTM E8 Type 3, and sub-size Types 4 and 5 were tested. Amplified polygon delay 300 µs introduced defects that increased defect density an order of magnitude of 10 while maintaining porosity below the NASA-STD-6030 limit (0.25%). Standard coupon geometry (Type 3, 6 mm gauge diameter) in both as built and machined conditions presented a statistically significant difference in UTS under an idealized model (R^2\u3e0.9) of less than 1%, which was too small to be considered relevant. Subsequent alternative tests showed that Type 5 as built samples were susceptible to keyhole defects with pairwise t-tests indicating a significantly lower strain at break 22 ± 4% compared to the nominal parameters of 28 ± 3%.
Despite these findings, sub-size coupons showed high variance in UTS due to surface roughness, introducing uncertainty in load-bearing area measurements. Consequently, tensile testing is limited as a qualification metric and is better suited for assessing material properties after qualification. Further research is recommended to evaluate rough sample geometry, explore alternative qualification methods based on geometrical accuracy, and improve subsystem testing and scanner control
The Last Xochitl, An American Story, Book 1: The Chicken Bone Wars
No full textThe following preface will demonstrate the power of female storytelling along the Texas border from the Tejano consciousness within the framework of my novel, The Last Xochitl, An American Story, Book 1: The Chicken Bone Wars. I will be examining oral and written stories within the genre of Magical Realism, from Latin-American, to Mexican and Chicano, and finally, a newly proposed genre, Tejano Magical Realism
An integrated study of aerosol concentrations and meteorological conditions in El Paso airshed using modeling and instrumentation
No full textAtmospheric aerosols are solid or liquid particles suspended in the atmosphere. Aerosols can be primary, which are directly emitted into the atmosphere, or secondary, which are formed when emitted gases undergo complex chemical reactions. Atmospheric aerosols can range from a few nanometers to tens of microns in diameter. Atmospheric aerosols play a significant role in climate; they can absorb or reflect heat, influencing temperature and weather patterns. Air quality is also affected by aerosols, which can have detrimental effects on human health, contributing to respiratory issues and other health problems. El Paso Texas is located at the south of the New Mexico state line and is surrounded by the Chihuahuan Desert. It is also adjacent to Ciudad Juarez - one of Mexico\u27s largest industrial cities - making it an ideal hub for studying atmospheric aerosols. This dissertation comprises several studies focused on aerosols within the El Paso airshed. Chapter 3 examines the behavior of meteorological parameters during dust events, and their impact on the dynamics of the planetary boundary layer. Chapter 4 employs convergent cross mapping and time series data to investigate the causality relationships between aerosols and meteorological parameters. Chapter 5 delves into the impact of particle coating on aerosol optical properties which is critical for understanding the radiative effects of aerosols. Collectively, these studies contribute to a deeper understanding of aerosol behavior in the El Paso region
Efficient Anomaly Detection Driven By Different Machine Learning Architectures And Models
No full textThe rapid growth and ubiquitous adoption of the internet and cyber-physical systems (CPS) have fundamentally transformed modern communication, work, and human-system interactions. While networks now form the backbone of critical digital ecosystems, enabling seamless data transmission across diverse, interconnected systems, this increased connectivity also expands the attack surface, making real-time detection of network intrusions and anomalies a pressing challenge. Detecting unusual activities within network infrastructure requires advanced data traffic analysis to differentiate between legitimate and malicious interactions. Traditional approaches to network anomaly detectionâ??such as rule-based and signature-based systemsâ??often depend on predefined patterns to identify known anomalies, limiting their effectiveness against emerging, stealthy, or previously unseen threats. These conventional methods suffer from high false alarm rates and fail to adapt to the ever-evolving nature of network traffic, particularly in large-scale, decentralized environments where data volume, velocity, and variety are constantly increasing.
This dissertation presents artificial intelligence (AI)-driven approaches to anomaly detection that leverage graphics processing unit (GPU)-enabled high-performance computing (HPC) platforms for processing massive network traffic data and monitoring the components of cyber-physical systems (CPS) for potentially hazardous conditions. The research advances several key contributions: (1) Designing efficient machine learning techniques for CPS condition monitoring and anomaly detection; (2) enabling federated learning (FL) frameworks that enable distributed detection while preserving data privacy and system resilience; (3) exploring graph-based methodologies combining graph neural networks (GNN) and graph machine learning (ML) approaches for the Internet of Things (IoT) and automotive network security, and (4) performing distributed edge computing optimizations that integrate FL with scalable technologies for reduced communication overhead. Through extensive experiments, these methodologies demonstrate that complex anomaly detection and condition monitoring tasks can be achieved while balancing computational efficiency and detection accuracy through fine-grained network information processing.
The frameworks developed in this research establish a robust foundation for network anomaly detection, providing scalable, adaptive, and privacy-preserving solutions for safeguarding CPS and IoT networks in an increasingly interconnected digital landscape. The practical implications of these research findings are significant, as they can inform the development of next-generation network security systems and contribute to the protection of critical infrastructure against sophisticated cyber attacks
Ultra-High Temperature Ceramics: Materials, Manufacturing, And Machine Development
No full textThis research aimed to advance the 3D printing of ultra-high-temperature ceramic matrix composites (UHTCMCs) and develop a custom Direct Ink Writing (DIW) system with multimaterial and in-line mixing capabilities. The study first focused on creating ZrB₂–SiC composites reinforced with aligned silicon carbide fibers (SiCf) using paste extrusion. By formulating suspensions with a preceramic polymer (SMP-10), ZrB₂, and SiCf, the study assessed how fiber alignment influenced key properties such as thermal and electrical conductivity and mechanical strength. Green curing and pyrolysis transformed the printed parts into UHTCMCs, with results showing substantial improvements in thermal conductivity—nearly double that of non-aligned parts—and a 10-fold increase in electrical conductivity when fibers were aligned. Despite the presence of pores that limited final conductivity, increased fiber content significantly enhanced fracture strength, suggesting that a fully dense, aligned composite could achieve even higher performance. The second focus of this research involved developing a custom DIW system to process yield-pseudoplastic ceramic inks. This system enables exploration of functional designs for composite ceramic armor and supports testing of biologically inspired armor structures, such as fish scales and abalone nacre. Future DIW system enhancements will include heating for viscosity control, adaptation for magnetically responsive inks, and UV curing capabilities. Overall, this research highlights the potential of tailored DIW-based additive manufacturing for creating UHTCMCs with optimized thermal, mechanical, and electrical properties, facilitating applications in high-speed and high-stress environments