Tag: Poster Presentation

Briauna Davis¹
Faculty Mentor: Szu-Ping Lee, Ph.D.^2
1School of Integrated Health Sciences, Department of Kinesiology and Nutrition Sciences
²School of Integrated Health Sciences, Department of Physical Therapy

ABSTRACT
It is difficult to stop accidental falls and maintain balance after leg amputation. The goal of post-amputation rehabilitation is to improve mobility function, reduce fall risks, and improve safety. Although there have been advancements in prosthetic technology, individuals with leg amputation are still experiencing frequent falls. We believe this is partially due to the lack of scientific knowledge on prosthetic skill learning after amputation. Post-amputation rehabilitation involves learning and relearning complex motor skills, such as walking and quick stepping to stop falls. This process becomes intense as Individuals perform these tasks with a prosthesis. Our goal was to find a way to make rehabilitation training after amputation more effective. This study examines the effectiveness of incorporating peer-based training during post-amputation rehabilitation. Participants with lower limb amputation will be instructed to watch video demonstrations of balance and recovery tasks performed by an amputee peer or non-amputee. The performance of participants will be tracked before, throughout, and after training. Knowledge from this study will benefit individuals with lower limb amputation by speeding up the learning of prosthetic skills lower limb amputation.

This research was funded by UNLV’s TRIO McNair Scholars Institute, which is housed within UNLV’s Center for Academic Enrichment and Outreach and funded under the TRIO Ronald E. McNair Postbaccalaureate Achievement Program by a grant (P217A170069) from the U.S. Department of Education. 

Jevons Wang¹
Andrew Ortiz²
Karen Alcazar³
Faculty Mentor: Jefferson W. Kinney, Ph.D.^4
1College of Sciences, School of Life Sciences
²College of Liberal Arts, Department of Psychology
³College of Sciences, Department of Chemistry and Biochemistry
⁴ School of Integrated Health Sciences, Department of Brain Health

ABSTRACT
Alzheimer’s disease (AD) is a progressive neurodegenerative disease that results in tissue damage and cognitive impairment. There are three pathological hallmarks seen in AD: 1) senile plaques, composed of the accumulation of A-beta protein, 2) neurofibrillary tangles, composed of hyperphosphorylated tau protein, and 3) a sustained immune response in the brain (neuroinflammation). A major non-genetic risk factor for AD is type 2 diabetes (T2DM). T2DM confers up to a 1.5 to 4 times more likelihood of developing AD; additionally, 80% of individuals who have AD, have T2DM or glucose intolerance. T2DM Initiates chronic neuroinflammation that leads to an exacerbation in AD pathology. The interrelationship between T2DM and the AD hallmarks are not yet fully understood. The focus of this study was to investigate if the AD deficits seen from T2DM are caused by hyperglycemia or neuroinflammation. We examined both tissue and behavioral data; however, for this project we focused on measuring learning and memory by utilizing the Barnes Maze. The preliminary data from the Barnes Maze video analysis indicated that there were no significant differences in “total latency” between any of the groups. Although there were no significant learning deficits with the STZ drug, preliminary data did show that the PZ group had significantly less “total error”.

This research was funded by UNLV’s Title III Part F Asian American and Native American Pacific Islander Serving Institution (AANAPISI) program, which is housed within UNLV’s Center for Academic Enrichment and Outreach and funded under grant (P382B160008) from the U.S. Department of Education.

Edgar Toro¹
Lindsay Chiquoine¹
Faculty Mentor: Scott Abella, Ph.D.^1
1College of Sciences, School of Life Sciences

ABSTRACT
Measuring the soil organic carbon (SOC) is of vital importance to soil science ecology. With the approaching challenges that rapidly changing weather patterns and temperatures bring, it is becoming more important to be able to measure the organic carbon levels in soil quickly, accurately, and cheaply. Currently, existing methods can very accurately deduce organic carbon levels but are lacking in speed and cost. Using alternate methods such as Loss-On-Ignition (LOI) can make up for lacking speed and affordability and have been shown to be very accurate under the right conditions. Ultimately, having a secondary method to quickly and easily get rather accurate results could be useful to preliminary investigations. The primary objective is to use LOI to accurately estimate the soil organic compound levels of low-elevation Mojave Desert soil. Using soils that were collected from wildfires spanning 15-20 years old, LOI procedures were tested using a factorial design including crucible size, temperature, and time. Using 5 mL crucibles, temperatures ranging from 300 to 600°C and time periods ranging from 2 to 8 h were used in different combinations to obtain the best results. The r2 values were examined for these factorial combinations to determine the most convenient and accurate combination. The results showed that 600 °C at 6 hours had the best results, with an r2 of 0.602. As these are just preliminary results, running a set with a much larger sample size must be done to ensure that the results are consistent.

This research was funded by the Southern Nevada Northern Arizona (SNNA) Louis Stokes Alliance for Minority Participation (LSAMP), which is housed within UNLV’s Center for Academic Enrichment and Outreach and supported by a grant (HRD – 1712523) from the National Science Foundation (NSF). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the NSF. 

Willaine Mae Kahano¹
Faculty Mentor: Sean A. Neiswenter, Ph.D.^1
1College of Sciences, School of Life Sciences

ABSTRACT
Domestic dogs (Canis familiaris) are one of the most well-known carnivorous species on the planet. Despite our familiarity with them, their effects on native fauna in protected areas is still unclear; however, many studies warn that dogs are a potential threat to wildlife. To understand the relationships between domestic dogs and native species, we conducted a preliminary carnivore survey at the Tule Springs Fossil Beds National Monument (TUSK) in Las Vegas, Nevada. Camera traps were placed in 14 different locations for an average of 28 trap nights. They were affixed to approximately 50 cm above the ground and were adjusted parallel to the ground. The cameras were programmed to run continuously for 24 hours/day and captured three photos every minute when triggered. 44,294 images were recorded across the cameras. The most dominant species recorded were humans (155 sightings), followed by domestic dogs (61 sightings), black-tailed jackrabbits (58 sightings), coyotes (46 sightings), and a few other desert species. Results reveal coyotes and jackrabbits were largely present in areas where both domestic dogs and/or humans were seen, and domestic dogs and coyotes were located at sites near residential areas. Our study indicates that coyotes do not respond negatively to the presence of domestic dogs. This is a preliminary survey, so more deployments are needed to form more definitive conclusions regarding the relationships between domestic dogs and native fauna.

This research was funded by UNLV’s Title III Part F Asian American and Native American Pacific Islander Serving Institution (AANAPISI) program, which is housed within UNLV’s Center for Academic Enrichment and Outreach and funded under grant (P382B160008) from the U.S. Department of Education.

Bijoux Cheun¹
Maria Martinez¹
Faculty Mentor: Sean Neiswenter, Ph.D.^1
1 College of Sciences, School of Life Sciences

ABSTRACT
The objective of this study is to investigate the effects that lead left at shooting ranges have on local rodent populations. Shooting ranges have been shown by previous research to have a great influence on the level of lead present in nearby soils and plants. This lead contamination has also been shown to have serious consequences for fauna residing near these ranges, ranging from lead toxicity to death. Many shooting ranges exist in Southern Nevada, and we have sampled a small number of these ranges to explore the possible effects they have on nearby rodent populations. The livers of Dipodomys merriami were sampled. We intend to sample additional species and consider life histories in relation to levels of lead toxicity.

This research was funded by UNLV’s Title III Part F Asian American and Native American Pacific Islander Serving Institution (AANAPISI) program, which is housed within UNLV’s Center for Academic Enrichment and Outreach and funded under grant (P382B160008) from the U.S. Department of Education.

This research was funded by the Southern Nevada Northern Arizona (SNNA) Louis Stokes Alliance for Minority Participation (LSAMP), which is housed within UNLV’s Center for Academic Enrichment and Outreach and supported by a grant (HRD – 1712523) from the National Science Foundation (NSF). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the NSF. 

Erick Serrano¹
Faculty Mentor: Bernard Zygelman, Ph.D.^2
1Howard R. Hughes College of Engineering, Department of Computer Science
²College of Sciences, Department of Physics and Astronomy

ABSTRACT
Neural Networks are commonly known for their applications across AI vehicles, smart appliances, robotics and more; however, they are limited by their non-polynomial time complexity. This report proposed to replace the feed-forward neural network (FFNN) with a quantum hybrid model and possibly reduce the complexity to linear time. We simulated Adiabatic Quantum Computing (AQC) for any two hamiltonians using python, and we showed that the accuracy scales with time. Then we use AQC to create and classically simulate a “trivial” AQC-QNN model; we state that it is trivial because the model replaces the back-propagation calculation for a perceptron. By testing against labeled 5-bit binary data, we found that the AQC-QNN model yielded 96.875% accuracy on the training set, and we also found that the model increased the time complexity of the neural network overall. While the AQC-QNN fails to reach the run-time goals, this sets up the foundation for an implementation on a quantum computer using the well-known quantum approximate optimization algorithm (QAOA). We expect to yield better run-times with a larger hamiltonian (which would minimize multiple weights, w1,…,wN, at a time) and with QAOA.

This research was funded by the Southern Nevada Northern Arizona (SNNA) Louis Stokes Alliance for Minority Participation (LSAMP), which is housed within UNLV’s Center for Academic Enrichment and Outreach and supported by a grant (HRD – 1712523) from the National Science Foundation (NSF). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the NSF. 

Karen Gonzalez¹
Vivian Flaum¹
Dustyn Weber¹
Faculty Mentor: Cory A. Rusinek, Ph.D.^1
1College of Sciences, Department of Chemistry and Biochemistry

ABSTRACT
Due to industrialization and globalization in the late centuries, environments have become increasingly compromised by pollutants. Heavy metals are a common environmental pollutant; they refer to a naturally occurring element having a high atomic weight and high density. Heavy metals, such as lead (Pb), tend to be toxic and present in trace amounts. Pb is a naturally occurring element that can affect virtually every function in the human body upon exposure, including severe damage to the kidneys and nervous system[2]. Interest in detecting lead (Pb2+) in water samples has risen in recent years due to several incidents of community-wide exposure around the world. Cloud point extraction (CPE) is a green chemistry technique used to extract and preconcentrate metals. Limited reports exist coupling CPE to anodic stripping voltammetry (ASV), an electrochemical method that can be used for trace detection of a variety of toxic metals. ASV is an inexpensive and easy to miniaturize technique that can achieve detection limits in the picomolar (10-10 M) range. In this work, Pb2+ was extracted by CPE and analyzed by ASV. The CPE matrix yielded a 38x increase in sensitivity over a traditional acetate buffer matrix with a 1-minute deposition time. The limit of detection (LOD) and quantification (LOQ) were 0.660 and 2.201 ppb, respectively, using CPE. Utilizing acetate buffer at pH 4.65, the LOD and LOQ were 3.202 and 10.670 ppb. Thus, the CPE yielded a 5x improvement in the LOD and LOQ. Overall, this method further exemplifies the wide applicability of electroanalytical methods.

This research was funded by UNLV’s Title III Part F Asian American and Native American Pacific Islander Serving Institution (AANAPISI) program, which is housed within UNLV’s Center for Academic Enrichment and Outreach and funded under grant (P382B160008) from the U.S. Department of Education.

Vanessa Nava-Camal¹
Faculty Mentor: Jorge Fonseca Cacho, Ph.D.^1
1Howard R. Hughes College of Engineering, Department of Computer Science

ABSTRACT
Augmented Reality (AR) technology can be used with indirect mediums to allow virtual interaction to have a greater impact on society. This research is significant because it contributes to giving users a better experience interacting with their surroundings. The purpose of this study was to test if items can be overlayed on an environment using a phone camera. Additionally, we investigated the type of software required for mobile phone platforms. Researchers used Unity Engine software equipped with a Vuforia mod to overlay a digital world onto a physical environment utilizing fiducial markers. For pathfinding, we used traditional algorithms such as A* and Dijkstra’s shortest path. Thus far, results indicate that when development is complete it will be feasible to develop cloud anchor-based navigation in ULABS navigation application as long as the device used is compatible with AR Core. The application will achieve AR navigation by overlaying on top of the physical environment with a fiducial tracker outline to pull virtual objects into reality using a phone’s camera. Further research into this subject will continue after the project moves into production.

This research was funded by UNLV’s Title III Part F Asian American and Native American Pacific Islander Serving Institution (AANAPISI) program, which is housed within UNLV’s Center for Academic Enrichment and Outreach and funded under grant (P382B160008) from the U.S. Department of Education.

Yessenia Henriquez¹
Faculty Mentor: SJ Kim, Ph.D.^1
1Howard R. Hughes College of Engineering, Entertainment Engineering and Design

ABSTRACT
Virtual reality (VR) and augmented reality (AR) habitually seek to construct new environments to produce virtual experiences. Virtual experiences are unimaginable to establish in real life, but not in a digital context. These virtual experiences could range from medical procedures in a virtual hospital setting to bizarre worlds in entertainment environments. Nonetheless, the rudimentary necessity to fabricate genuine virtual experiences is immersion. This literature review centers around discovering various aspects that formulate both augmented reality (AR) and virtual reality (VR). This study attempts to comprehend the possibilities and objectives that AR and VR are most compatible with. The upcoming results heavily touch upon immersion (genuine virtual experiences) and cost (physical or software cost) in VR/AR. Overall, VR structures are more immersive to the user in integrating an “out of body” experience but are often more costly due to the equipment needed to perform these computations. AR structures are more available to cheaper alternatives but do not create a serious virtual experience (VE) as VR structures. Both VR and AR environments and objectives, however, combine interdisciplinary areas for a dynamic VE for users.

This research was funded by the Southern Nevada Northern Arizona (SNNA) Louis Stokes Alliance for Minority Participation (LSAMP), which is housed within UNLV’s Center for Academic Enrichment and Outreach and supported by a grant (HRD – 1712523) from the National Science Foundation (NSF). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the NSF.