LSU College of Engineering High School Summer Research Program

HSSR Program

High School Summer Research Program

About

As part of its strategic plan, mission, and vision, the LSU College of Engineering is dedicated to developing students into the next generation of transformative problem solvers for the local region, the state, and beyond. The High School Summer Research (HSSR) Program is an outreach initiative aimed at engaging high-achieving high school students in real research in the fields of engineering, computer science, and construction management. HSSR interns have opportunities to learn from faculty research groups and understand how they work, what inspires them, and how to continue in fields related to engineering in college and beyond.

In addition to their work on a research team/project, HSSR interns also attend workshops that include trainings on safety, research best practices, ethics in research, and communication. The program culminates in a poster presentation for students to present their research projects, which is mandatory for the completion of the internship.

Program Goals

  • To give high-achieving, highly motivated, and hard-working local high school students meaningful experiences in engineering, computer science, and construction management research during summer.
  • To develop students’ curiosity, research methods, and intellectual abilities before they have completed high school or made decisions about which college to attend and what bachelor’s degree to pursue.
  • To develop students’ abilities to communicate technical content using both written and oral modes of communication.
  • To teach students about the engineering design process and how it can be applied to both fundamental and applied research.
  • To introduce these students to the opportunities available at the LSU College of Engineering and showcase the impressive faculty and research available to them at the state flagship university.
  • To support faculty in their research projects and outreach efforts.

Information Sessions

Three in-person information sessions will be held, as well as one virtual session:

  • October 26, 2023 at 6:00 p.m. (in-person)
  • November 30, 2023 at 6:00 p.m. (in-person)
  • December 6, 2023 at 6:00 p.m. (virtual)
  • January 16, 2024 at 6:00 p.m. (in-person)

Registration is required to attend an information session. Meeting details will be provided via email after registration.

Register for an Information Session

Program Details

Students selected as HSSR interns are matched with a College of Engineering faculty member's research team. They will receive guidance from the professor, as well as graduate-student and undergraduate-student mentors, as they work on a project related to the research team's ongoing research. Here are examples of past research projects and an article about a student/project from summer 2020.

HSSR Interns will not be paid for their work.

HSSR Interns will be held accountable for their work responsibilities by the College of Engineering and will be expected to compete in regional science fair competitions. The program will provide information and training regarding science fair participation.

HSSR Interns will have to complete detailed safety paperwork and training through the course of the spring 2024 semester in order to begin work on a project in summer 2024 and beyond.

HSSR Interns will be required to work about 15-20 hours per week during summer 2024 for a total of about 120-140 hours. Weekly schedules can be flexible depending on summer travel/activity schedules, however students should not miss more than 4 working days.

The HSSR Intern application and selection process will be highly competitive due to high interest and a limited number of available positions.

Apply

Application Deadline: February 2, 2024

The LSU College of Engineering is seeking qualified local high school students to apply for a limited number of High School Summer Research (HSSR) Intern positions available in summer 2024. Please see the program details and eligibility before applying. If you have questions about eligibility and program details, please contact Raynesha Ducksworth at rducksworth@lsu.edu.

To be eligible for this program, you must:

  • Be at least 15 years of age.
  • Be currently enrolled as a 9th, 10th, or 11th grader.
  • Have a 3.5 (or equivalent) high school GPA (as listed on a current high school transcript).
  • Complete and submit this application by end of day on February 2, 2024.

To apply for this program, you must:

Current Projects

Project Title Abstract Program Student and Mentor
The Use of Glass Sand to Prevent Erosion in Coastal Louisiana The Louisiana coast experiences rapid land-loss due in part to both man-made and natural phenomena. Glass Half Full, a private non-profit, has begun incorporating recycled glass into projects that would usually use natural sand; their goal is to directly deposit glass sand into wetlands and coastal areas that experience high rates of erosion. The company is still relatively new, however, and little research has been done into how the glass sand mimics natural sand in a natural environment or if it would have any unexpected effects on the surrounding ecosystems. In order to determine these properties, experiments were conducted on a coarse glass sand, a fine glass sand, sand from the Mississippi River, and sand from Grand Isle, Louisiana, a barrier island of the state to test grain size distribution, settling velocity, density, angle of repose, incipient motion, and chemical composition. The primary goal of these tests was to compare the artificial sands to the natural ones, looking for any property that would warrant the glass sand an unusable supplement to the natural sand in a natural environment. While many differences did present themselves, primarily in grain size distribution and settling velocity, nothing implied that the glass sand was too unlike the natural sands to be used as a supplementary material. The only test from this round of experiments that researched the possible effects of the glass sand on the environment were the chemical composition tests. Using a standard drinking water test kit to find the amount of sixteen different common chemicals in the water, only the fine glass sand showed any variation, causing the water to have a higher pH, alkalinity, and carbonate. Further research can and should be done into the glass sand’s ability to impact the natural environment, including any effects it may have on the flora and fauna that interact with it. Civil and Environmental Engineering Adele Aldridge, Clint Willson

Quantifying the Benefits of Freeboard Policy on Louisiana Parishes’ Class in the Community Rating System

The Community Rating System (CRS) is a federal program, sponsored by the Federal
Emergency Management Agency (FEMA) and in partnership with the National Flood Insurance
Program (NFIP). Participating communities receive a discount on their Flood Insurance
Premiums based on the efficacy and scope of their flood-related/CRS-based policies. This project aimed to quantify the cost benefit of a specific factor of CRS: freeboard. Freeboard is added height to the lowest level of a housing unit, above the Base Flood Elevation (BFE). This
quantification was done specifically for Jefferson Parish, due to it being the highest performing
CRS community in Louisiana, with a Class 5 Rating. Previous CRS literature and research has
been generalized to reach an overall understanding of CRS’s past effectiveness, whereas this
project builds off of the established effectiveness and isolates a specific factor and community in
order to calculate future benefits. The isolation of the freeboard factor was done using 9 potential
freeboard policies, each one being a combination of: 1, 2, or 3 feet of required freeboard height
and: no fill requirements (NFR), compensatory storage required (CSR), or fill prohibited (FS).
These policies were then customized by finding the impact adjustment ratio, annual growth
adjustment, and county growth adjustment of Jefferson Parish. The aforementioned factors were
calculated using the amount of the parish (in acres) in a Special Flood Hazard Area (SFHA), the
suggested number of acres in the parish which should be left for Open Space Preservation (OSP),
the predicted number Dwelling Units in 2028 (DU+5, which were calculated based on the
structural percent growth change from 2018 to 2023) and the number Dwelling Units in 2018
(DU-5). The results indicated that any of the 9 freeboard policies would improve Jefferson
Parish’s status in CRS and create a multimillion-dollar discount on their Flood Insurance
Premiums. However, of these 9 policies, CSR and FS with 3 feet of required freeboard would
have the greatest cost benefits.
Biological and Agricultural Engineering Keya Chanda-Rastogi, Zachary Kaupp, Carol Friedland

Studying the Effects of Distortion on the Hydrodynamics in the Lower Mississippi River Physical Model

Human intervention has caused many environmental effects, such as climate change, sea level rise, and saltwater intrusion. These, along with other manmade structures like levees and dams, have led to a lack of sediment deposition along Louisiana’s coastline. As a result, land loss has become a prominent problem in Louisiana that not only affects ecosystems, but also families and businesses. One of Louisiana’s plans for the current coastal crisis is sediment diversions. Diversion projects are gates built into levees that allow the flow of sediment and freshwater back into the marshes to rebuild lost land. The Lower Mississippi River Physical Model (LMRPM) is a moveable-bed model used to test and research coastal restoration projects such as sediment diversions, in order to optimize operation. However, the physical model has a distortion of 15 which can affect factors such as centrifugal forces, turbulence, and helical flow. In this study, particle image velocimetry (PIV) was used to analyze the flow of fluorescent particles in the LMRPM in order to quantify different parameters in the LMRPM. 3 different flow rates - ~550k cfs, ~700k cfs, and ~1000k cfs - and 3 different locations – a straight section, a moderate bend, and a sharp bend - were tested. The velocity magnitudes, vorticity, and shear rate color maps were generated and analyzed. The data was then compared to a numerical model of the prototype (the actual Mississippi River), and a numerical model of the LMRPM in order to determine the effects of the distortion. From the color maps, the LMRPM shows a much smaller separation zone compared to prototype, which could be caused by the weaker centrifugal force in the LMRPM due to the steeper channel walls. Since bed shear stress is proportional to velocity, the maximum bed shear stress in the LMRPM is more towards the inner bends compared to the prototype. The smaller separation zone and higher shear stress indicates less sediment deposition in the inner bends in the LMRPM. Understanding the effects of model distortion is crucial for coastal restoration efforts and research with the LMRPM. In the future, a 3-dimensional software can be created and compared with the data from this research to create a more detailed analysis on the 3-D hydrodynamics in the LMRPM. Civil and Environmental Engineering Emily Chen, Clint Willson

Coordination of Fully Automated Vehicle Platoons for Crossing Non-stop Intersections

Autonomous vehicles have developed rapidly in recent years, now being commercialized to the public. As roadways become increasingly more autonomous, the use of V2X communication will become more viable as well, possibly replacing current visual traffic signs with instant inter-vehicle coordination and non-stop traffic. With non-stop intersections, urban travel time would decrease, and fuel/energy wasted at intersections can be eliminated. The goal of this project was to develop a method of coordinating vehicle platoons through a perpendicular, non-stop intersection. The trajectory was computed on MATLAB and visualized on MATLAB’s driving scenario designer which uses waypoints and instantaneous velocity to plot routes. The initial model was a simple 2-platoon intersection where the initial calculations were formed. The spacing between vehicles was determined so that each platoon’s vehicles would enter the intersection at the same rate as the other platoons with the spacing being great enough to avoid collision. Platoons also implement a forward shift to ensure that the vehicles of one platoon cross the intersection in between the vehicles of the other platoon. The program also checks if the change in spacing and forward shift is needed and if not, the platoon would continue with the same trajectory. The program was then recreated with a 2-lane intersection where the initial calculation for spacing was changed to be based on the road length rather than the length of each vehicle. Afterward, four platoons were implemented, and the conditionals and calculations were adjusted accordingly. In the final model, the calculations and changes in velocity were optimized by having the change in spacing and forward shift act in the same period. Vehicles also began spacing at the same time, but ended it at the same distance from the intersection. This allowed for more cars on each platoon without any vehicle traveling in reverse when spacing. Finally, the velocity-time graphs of vehicles on the non-stop intersection model were compared those of vehicles at a traditional stop-and-go intersection, decelerating at distances of 70-120 meters from the intersection and ignoring time spent waiting at the intersection. The comparison was made with platoons of different velocities and amounts of cars, each showing a similar time spent to cross the intersection.

Electrical and Computer Engineering

Alvin Cheng, Jones Essuman, Xiangyu Meng

Optimal Path Planning with Applications to Automatic Parking

Automatic parking is an important problem to solve, as it provides convenience and safety to inexperienced drivers and contributes to unmanned/autonomous driving. One important aspect of automatic parking is path planning, where a vehicle plans a trajectory that it will take. When creating a path, it is important to create a physically feasible and collision-free trajectory. However, a more interesting aspect of path planning is to optimize this trajectory by minimizing distance as well as avoiding sharp acceleration and steering. This project aimed to investigate different path-planning methods and implement them in various parking scenarios. Two major path-planning algorithms were used: RRT* (Rapidly Exploring Random Trees) and NLMPC (Non-Linear Model Predictive Control). RRT* converts geometric space into a tree of randomly selected points, then adjusts the tree to make points that minimize the distance from the starting position to the goal. Model Predictive control numerically solves for the optimal path by minimizing a function that consists of both following a reference path and minimizing changes in inputs. Various parking situations were created in Matlab, and executed using both algorithms. The algorithms themselves were adapted from existing examples for alternate parking situations. A combination of both algorithms was also investigated, specifically where RRT* planned the path and an MPC provided inputs to follow the path. All algorithms were able to find feasible and reasonable trajectories. Model predictive control minimized the change in steering angle and velocity, resulting in a smoother trajectory. On the other hand, RRT* was much faster, and more versatile in usage. Still, MPC took only about 15 seconds to park, which is fine since parking is not time-sensitive. Still, in certain scenarios, MPC takes about 15 seconds, which is acceptable given that parking is generally not a time-sensitive procedure. Both algorithms have their merits, and combining their strengths enhances solving more challenging parking problems in both current and future applications.

Electrical and Computer Engineering

Varun Gala, Xiangyu Meng

Utilizing Machine Learning Classification Algorithms to Detect Pancreatic Cancer with Fluorescence Spectrum Data

Pancreatic cancer is one of the hardest types of cancer to detect early, since there are few symptoms of pancreatic cancer in the early stages. Because early detection improves rates of survival, this study aimed to use random forest machine learning to classify pancreatic cancer tumors with fluorescence spectrum data collected from mice, in order to help with early detection and increase survival rates. During the imaging process, indocyanine green (ICG) was used as a fluorescent agent. 400 samples of fluorescence spectrum data were imported into Excel, with 200 non-cancerous samples and 200 cancerous samples, and the data was processed, transposed, and analyzed. A random forest algorithm was then implemented in Python, and the model was trained and tested, with 80% of the data used to train the model and 20% used for testing. With an accuracy rate of 98%, the model was highly effective in classifying whether a sample was cancerous or not. The results indicate that random forest machine learning algorithms can be effective in classifying pancreatic cancer tumors using fluorescence spectrum data. This framework could be utilized in future studies in order to detect other types of cancer. This study could be another step in the further development of a non-invasive pancreatic cancer detection tool in the long run.

Electrical and Computer Engineering 

Vaishnavi Kumbala, Huaizhi Wang, Jian Xu

Weighted Gene Co-expression Network Analysis (WGCNA) Analysis of the Genotype-Tissue
Expression (GTEx) data from the left ventricle

Cardiovascular disease is the leading cause of mortality in the United States. To better understand cardiovascular disease, we require deeper understanding of the genetic networks involved in heart function and pathology. We used Weighted Gene Co-expression Network Analysis (WGCNA) to analyze human heart tissue samples, aiming to unravel the complex network of gene co-expression and identify hub genes. Determining hub genes in a dataset can aid in elucidating various biological pathways, or diseases associated with a given tissue. Our gene expression data was sourced from the established Genotype-Tissue Expression (GTEx) portal (https://gtexportal.org)2 , an online repository maintained by the National Institutes of Health (NIH), which offers comprehensive gene expression data across various tissues. Specifically, we utilized data from the left heart ventricle and constructed large-scale signed networks with a set of 386 genes. Upon constructing the network, we began an in-depth analysis of these genes, focusing on their properties and the identification of hub genes within each cluster or module. This process resulted in the creation of 63 clusters. The hub genes within these clusters were identified based on their significance in correlation with other genes in the same module. The “turquoise” module of the data was examined and further analyzed using Cytoscape to better understand the networking of genes.  The gene “ENSG00000197885.10” is the top hub gene of the turquoise module. Results from this current analysis were not conclusive and research is still ongoing.

Biological Engineering

Micah McKanstry, Jangwook P. Jung

Pancreatic Cancer Detection by Artificial-Intelligence-Assisted Raman Spectroscopy

 

Pancreatic cancer is one of the most lethal types of cancer with a five-year survival rate of 9% and a mortality rate of 79% mainly due to the difficulties in the early detection of pancreatic cancer. The survival rate increases for pancreatic cancer the earlier it is diagnosed because it can be cured before it has metastasized. This project intends to increase the five-year survival rate by aiding in pancreatic cancer detection at early stages by incorporating a machine learning algorithm in accordance with Raman Spectroscopy. Raman spectroscopy is an emerging cost-effective method of cancer diagnosis that can yield high accuracy. In this project, we utilized Raman spectroscopy to capture data from the pancreases and tumors extracted from a mouse model. We then retrieved 400 spectra (200 cancerous spectra and 200 non-cancerous spectra) from captured data which was then normalized via MATLAB and saved in Excel allowing it to be used as a data input for the Random Forest machine learning algorithm. The Random Forest algorithm was then implemented into Python and the data was trained while optimizing the parameters of the algorithm for high-performance results. For this project, we focused on observing the accuracy, sensitivity, specificity, as well as the ROC curve. The Random Forest algorithm performed very well resulting in 98.5% accuracy, 98% specificity, and 99% sensitivity, and a ROC curve area near 0.9984. Overall, the results indicated high performance of the Random Forest machine learning algorithm meaning Raman Spectroscopy can be used in the future with machine learning algorithms such as Random Forest for pancreatic cancer detection at earlier stages potentially increasing the survival rate.

Electrical and Computer Engineering

Neha Mothilal, Ya Zhang, Jian Xu

Development of Earthen Building Materials Inspired by the Nest Construction Techniques of Mud Dauber Wasps

 

Using concrete and steel to build structures hurts the environment since doing so emits significant amounts of carbon dioxide. Due to the high carbon emissions from producing steel and concrete, engineers are now constructing more buildings from earth. However, earth lacks strength, which results in lower safety levels compared to concrete and steel. This disadvantage hinders the common use of earth. This research seeks to increase the strength of earthen building materials (soil) by applying techniques used by mud dauber wasps in hopes of popularizing the building technique. Mud dauber wasps construct nests out of soil and repeatedly tap their nests with their legs and mandibles, producing a vibration that compacts the nest. It is hypothesized that soil strength increases by applying vibration to soil like mud dauber wasps in the experiment, the vibratory compaction was replicated by passing layers of 3D printed soil samples between two metal plates placed in the jaws of a wrench. In order to determine the effects of vibratory compaction of soil strength, penetration resistance tests and density tests were conducted on 3D printed soil samples compacted with vibration and uncompacted soil samples. The initial penetration resistance tests and density tests showed that vibratory compaction has no effect on the penetration resistance of soil and soil density. This may be because the vibration applied to the soils was not strong enough. Thus, vibratory compaction will be re-examined by applying the same force on a smaller area of soil (thinner layers of soil). In addition to increasing the strength of soil through vibratory compaction, another goal of this study is to model the cavities inside mud dauber nests in the design of earthen walls to decrease the amount of soil needed in the construction of earthen buildings. Since mud dauber nests are strong despite containing cavities, it is hypothesized that having a hollow column in an earthen wall can potentially reduce the soil needed for a wall while resulting in a strong wall. This research aims to find the optimal cavity size in an earthen building by determining the ratio between tensile strength and the amount of soil used cavity size. Models of earthen columns with various cavity sizes were designed using Fusion and Cura Ultimaker and were 3D printed. Currently, tensile strength tests are being conducted on the earthen columns with varying cavity diameters to determine the optimal cavity diameter.

Civil and Environmental Engineering

Ahebwa Muhumuza, Joon Soo Park, Hai Li

Machine Learning-Based Colloidal Self-Assembly Phase Identification

Colloidal self-assembly systems are systems containing colloids that can be guided into different states using electromagnets. These states are important to classify as the system can then be guided slowly into the right state. These states have previously been determined by order parameters; however, order parameters are hard to derive and are system specific. Recently, Machine Learning has been used to classify these systems, however the current algorithm will still mispredict images, when images are clustered. In this project a machine learning model is created to detect and enhance cluster purity. A Siamese Convolutional Neural Network (SCNN) was chosen to solve the misprediction of images as it allows for small amounts of data to be used to train the model and still give cutting edge performance. The SCNN is primarily based off ResNet-50, with a dense layer attached.  The model’s hyperparameters are tuned utilizing Bayesian optimization to allow for optimized hyperparameters. The triplets for the SCNN are generated with online triplet mining, and consist first of semi-hard negative triplets, then hard-negative, and finally easy triplets. These triplets allow the model to learn as much as it can in the shortest time possible. After the final model was trained, it was tested on the clusters to see if the proficiency was achieved. The model generated an embedding for each image in each cluster, then found the Euclidean distance between each image. This pairwise distance matrix was then used to find by summing each row and comparing the value. The images that were outliers in each cluster were then compared with other clusters and reclassified. The SCNN correctly reclassified _ images, however it incorrectly reclassified _ images, and did not reclassify _ incorrect images. These results show the potential for a SCNN generated embedding layer to be able to correctly classify images with a large reduction of features used from over 20,000 to 16. SCNNs have the potential to accurately identify features in colloidal systems and classify or reclassify these images based on these features.

Chemical Engineering

Benjamin Namikas, Andres Lizano, Xun Tang

Unraveling the potential of ChatGPT and AI in optimizing the Average High Schooler’s Daily schedule

High schoolers lead hectic lives, juggling academics, athletic commitments, social interactions, and personal responsibilities. This project delves into the transformative potential of ChatGPT and AI in shaping the daily schedules of average high schoolers. I used ChatGPT to uncover ways to improve on time management, productivity, and the overall well-being of the student. The incorporation of ChatGPT into high schooler’s lives revolutionizes the planning of daily routines. AI-powered schedule optimization takes into account personal preferences, academic performances, and extracurricular activities in order to create personalized schedules. The purpose of this study is to investigate and understand the potential benefits and challenges that arise when integrating AI technology into the lives of a high school student. This is accomplished by using ChatGPT and asking it a series of increasingly specific questions in order to create the perfect agenda for the student to live by. I also added unique situations that would only apply to me in order to see how effective it would be as a personal assistant and a planner. I found that the chatbot was able to give seemingly adequate schedules, however it took many attempts and specifications to reach a decent plan. The schedule provided by ChatGPT had increased my efficiency and productivity whether it be in studies or athletics. Additionally, I found that ChatGPT could also be used to help me learn new skills and different techniques. In conclusion, this project showcases the remarkable potential of ChatGPT and AI in optimizing the daily schedules of an average high schooler. By streamlining time management, enhancing academic support, and promoting a good balance, AI can empower students to navigate their daily lives with efficiency and purpose. As AI continues to grow and evolve, it will become essential for students, parents, and all sorts of people to utilize this technology to help better society and lead ourselves towards a brighter future.

Computer Science and Engineering

Joshua Ng, Hao Wang

Examining Drivers’ Behaviors to Connected and Autonomous Vehicles

Truck platooning is a transportation method that consists of a manually operated lead truck followed closely by a platoon of automated, connected trucks. These follower trucks follow the actions of the leading truck; if the leader brakes, vehicle-to-vehicle communications will tell the following vehicles to brake, too. Interest in this field of research comes mostly from the benefits offered by truck platooning. According to previous research, truck platoons reduce the drag experienced in contrast to individual trucks, subsequently increasing fuel efficiency by ~10-12%. When taking into consideration that trucks and trailers accounted for around 7% of all greenhouse gas emissions in the U.S. in 2020, that translates to around a 0.7% decrease in carbon dioxide emissions annually. Though seemingly small, this would be a step in the right direction. However, one major problem needs to be addressed before truck platoons can be fully implemented: how drivers of manually operated vehicles will respond to these automated truck platoons. This study aims to do that with a total of 80 participants of differing ages, education levels, and years of driving experience. They were tasked with driving two separate scenarios in a driving simulator, in which they would merge and diverge with automated truck platoons on a two-lane highway. From here, data such as the time to diverge (TTD) and time to merge (TTM) were collected and subsequently analyzed to see the effect of autonomous and connected truck platoons on driver behavior. Using a single factor ANOVA test for the participants’ years of driving experience vs. time indicated no significant difference between years of driving experience and TTM, but a significant difference between years of driving experience and TTD. On the other hand, using a single factor ANOVA test for the participants’ age vs. time indicated a significant difference of both age vs. TTM and age vs. TTD.

Civil & Environmental Engineering

Andy Ou, Mohamed Mohamed, Hany Hassan

Comparison of Bone’s Natural Microstructure to Applied Speckle Patterns

Finding different ways to examine a bone’s microstructure can allow for a better understanding of how bones fracture. Digital image correlation (DIC) is a method used to assess deformations in a material under mechanical loading and requires a speckling pattern on the material surface to track the deformation of the material.[1] In prior work, a speckle pattern was applied to the bone’s surface, and the bone was placed into a three-point bending test and deformed to fracture. The drawback of using a speckle pattern that was applied on top of the bone is that the pattern covers the bone’s microstructure and prevents the viewing of fracture propagation as it moves through the bone’s microstructures. The goal of this work is to determine if the natural microstructure of a bone can be used as a speckle pattern to observe fractures as they move across the bone. In order to make the microstructures visible, dyes must be applied to the bone. The dyes used in this research are black India ink, and toluidine blue dye. These dyes will enhance the visibility of the microstructures and potentially allow for use in DIC analysis. To allow a better understanding of how the dye affects the microstructures, a software called ImageJ will be used.[2] ImageJ will give specific parameters of the bones that are dyed. These parameters include size of the natural speckles and what percentage of the image the speckle pattern takes up. These parameters will then be compared to the proper speckle patterns that were previously applied to the bone. 

Biological and Agricultural Engineering

Grant Pendergraft, Beatriz Garcia, Alexander Lee, Kevin Hoffseth

A Machine Learning Approach to Analyze Energy Burden in U.S. Low-Income Households

44% of households in the U.S. are low-income, and these households often occupy affordable homes, the lowest quality housing unit. As a result, these homes tend to be less energy efficient with a higher energy burden (the percentage of household income spent on energy services). This poses a problem for these low-income households (LIHs) that must pay more for energy. This is especially problematic in the South, where the energy burden can reach over 10% for LIHs. The goal of this project is to analyze the energy characteristics of affordable homes and the energy burden of U.S. LIHs in response to this problem. In this study, the Residential Energy Consumption Survey (RECS) data is used from the U.S. Energy Information Administration (EIA). This data was collected by surveying energy usage and household characteristics from 5,600 homes in the U.S. It has been cleaned by removing imputation flag columns and unavailable data points (marked by a -2), changing certain variables for easier use (ex. converting the range of income into an estimated income), and adding columns for future analysis (energy burden, energy index, poverty line, etc.). The data is then analyzed through JMP using correlation analysis to observe the relationship between the entire data set and three dependent variables: the energy index, the energy burden, and the total energy usage in BTU. A predictive model is also developed using Artificial Neural Network (ANN) to estimate the energy index and energy burden in LIHs.

Construction Management

Cesar Rico, Amirhosein Jafari

Toluene Production Capacity of a Microbial Community Derived from Colorado River Sediment

Bacteria synthesizing toluene is a recent discovery that has the potential to provide a fuel source that, unlike current methods of producing toluene, is renewable. Toluene is a gasoline additive that boosts the octane rating, or how well the fuel resists premature combustion inside of the engine. While a previous study was based in California, this study revolves around the Colorado River. Samples were collected from a section of the Colorado River that flows through the city of Austin in Texas. After inoculating two groups of bottles, one with the samples collected from the Colorado River as the biotic group and the other with deionized water as an abiotic control group in order to ensure that preexisting toluene was not present to affect toluene and phenylacetic acid content measurements. In order to measure toluene and phenylacetic acid content, a gas chromatograph and ion chromatograph were used. After compiling data presented by a chromatograph, toluene production was shown in one out of the three bottles samples weekly. A pattern that was noticed was that the bottles containing toluene all came from the same original growth medium bottle, which could mean that the other bottles were prepared incorrectly. Producing toluene renewably through bacteria would have a profound effect on the amount of greenhouse gasses emitted during the production of nonrenewable energy sources such as fossil fuels, natural gases, and petroleum.

Civil and Environmental Engineering

Amaan Shafi, William M. Moe

The Photobleaching Effect of Fluorescent Proteins for Cell-Free Biosensor Development

Light exposure and photobleaching affect the fluorescence of green fluorescent proteins (GFP) extracted from cell-free protein systems (CFPS). GFPs allow identification of subcellular components. All GFPs are the result of extraction from a CFPS; E. coli CFPSs are used for this research. CFPSs allow for direct access to the contents of the cytoplasm without blockage from the cell wall or membrane. Most GFPs do not have a high fluorescence detection. A commonly used GFP, sfGFP, has a high fluorescence detection, but does not have a high dynamic range. A relatively new GFP, StayGold, maintains a longer dynamic range and has a relatively high fluorescence detection. Studying subcellular components without consistent identification proves difficult, but StayGold allows for longer research periods. The goal of this research is to utilize StayGold’s high fluorescence detection limit and dynamic range and study its ability to maintain a dynamic range with photobleaching effects. The presence of the protein correlates to fluorescence detection. Protein presence of a negative control, deGFP, sfGFP, and StayGold were measured using an SDS-Page with 4-12% Bis-Tris protein gel. sfGFP was the most present protein in the gel, while StayGold followed. Fluorescence was detected using a plate reader (Em: 485 nm / Ex: 510 nm). The plate readings were done in one- and three-minute cycles. Neither cycle duration led to significant decrease in dynamic range of fluorescence. The cycles of the plate reader did not allow for consistent photobleaching of the proteins.

Biological Engineering

Camille Starkovich, Khoa Doan, Yongchan Kwon

3D Printed Co-culture Platform to Study Bacteria Induced Endocrine Resistance in Breast Cancer

The tumor microbiome has been suspected to play an important role in cancer progression and metastasis, with breast tissue-resident microbiota potentially driving resistance to endocrine therapies. Unfortunately, there is still a limited understanding on how bacteria alter cancer cell behavior due to few available technologies to study interactions between cancer cells and bacteria in vitro. This project aims to simultaneously co-culture cancer cells and bacteria in the presence of an estrogen modulator metabolite (4-OHT) using a 3D-printed plate insert in 6-well plates that physically separates, yet chemically connects, the two cell types to determine if there is a cancer-promoting effect. The bacteria used in this experiment was Escherichia coli (E. Coli) because recent studies have found it to be present in the breast tumor microbiome and it is easy to culture which allows for optimization of the system. The estrogen receptor positive (ER+) breast cancer cells, MCF-7s, were grown in the well plate while the bacteria were placed on an agar slab supported by the 3D printed insert. An Alamar blue assay was performed to determine how the presence of the bacteria altered the growth rate and 4-OHT susceptibility of the MCF-7 cells. Preliminary findings support the use of the technology to co-culture the two cell types without contamination and suggest that the bacteria do alter growth and the response of MCF-7 cells to 4-OHT after 24 and 48 hours of co-culture.

Chemical Engineering

Anna Terrell, Victoria Hart, Emmaline Miller, Adam Melvin

Tracking Augmented Reality/Virtual Reality (AR/VR) Users' App Usage Duration through Push Notifications

Augmented Reality (AR) and Virtual Reality (VR) technologies have emerged as powerful tools that revolutionize the way we interact with digital content and our physical surroundings in recent years. As these immersive technologies gain widespread adoption, understanding user behavior and engagement within VR/AR applications becomes increasingly vital. One significant aspect of user interaction that demands investigation is the duration of application usage. In this project, we develop an AR application for the NREAL AR Glasses. To enhance user engagement and manage usage duration effectively, we implement push notifications within the application. Users will be promptly notified when their overall usage duration nears or surpasses predetermined thresholds. This proactive strategy promotes responsible user behavior by encouraging regular breaks and mindful screen time management. By setting limitations on both AR and VR usage, potential adverse effects such as eye strain, fatigue, and potential impacts on mental health can be minimized. The careful management of usage duration fosters a healthy equilibrium between immersive virtual experiences and real-life obligations. Moreover, exercising control over AR and VR usage emphasizes the significance of face-to-face interactions and the development of crucial social skills. This aspect holds particular importance for children, whose cognitive faculties are still in the process of maturation. Unrestricted use of AR and VR may lead to decreased productivity and hinder educational or professional growth. Therefore, employing measures to regulate and monitor AR and VR engagement becomes imperative for ensuring overall well-being and sustained personal development. The AR application is developed on Unity in C# programming language and experiments show promising results in accurately collecting app usage time and sending out timely push notifications. Bottom of Form This project presents a contribution to the field of AR and VR technologies by providing an innovative way to address problems such as physical and mental well-being, balance with real life, social interaction, productivity, and learning difficulties brought on by the extended usage of AR and VR technologies. To get the most out of AR and VR experiences, the application encourages users to adopt healthier screen habits by using the app's ability to track and manage usage duration through push notifications. Overall, this project contributes to creating a more user-friendly and sustainable environment for the rapidly expanding AR and VR technology industries.

Computer Science and Engineering

Katherine Toncrey, Chen Wang, Ruxin Wang

Improving Low Temperature DRM by Deposition of CeO2 Overlayers on Ni/Al2O3 Catalysts

Carbon Dioxide (CO2) and Methane (CH4) are the most abundant greenhouse gases in the atmosphere and are largely responsible for global temperature increases as well as other ecological problems like acidification of the world’s oceans. The dry reforming of methane (DRM) is a process used to reform atmospheric CO2 and CH4 into an optimal ratio of 1:1 H2/CO (syngas). Syngas can be used as a fuel and to make industrial chemicals using the Fischer-Tropsch process. Nickel is commonly used as a catalyst for DRM because of its cheapness as compared to noble catalyst metals (Pt, Pd, Rh) while having similar DRM activity. However, nickel deactivates quickly under realistic DRM conditions (>700°C) due to sintering of the active metal. Rare earth oxides like Ceria (CeO2) have been shown to mitigate these issues due to their redox behavior, allowing them to easily reduce and oxidize between Ce3+/Ce4+. This forms oxygen vacancies, which help both to oxidize coke and to activate CO2. This has been shown to greatly reduce the activation energy, allowing for lower temperature regimes (700°C >). A drawback of using CeO2 is that it encourages the reverse water gas shift (RWGS) reaction, which produces water instead of H2, decreasing the H2/CO ratio. In this work, we tested both an incipient wetness and urea strong electrostatic absorption method to create a CeO­2 overcoat on Ni/Al2O3. These catalysts were then tested to determine their activity using a TGA/DSC reactivity screening test. XRD and BET were used to confirm the deposition of CeO2 on the Ni/Al2O3 base sample and RAMAN spectroscopy was used to further identify the phase of the overlayer. Our sample made using urea deposition activated without reduction, but it took a long time for this to happen (~400 min). Our sample made using incipient wetness did not activate whatsoever, implying that reduction is needed for nickel activation. We confirmed the deposition of ceria and found that a ceria overlayer improved the metal-support interaction.

Chemical Engineering

Paul Toups, Johnathan Lucas, Kerry Dooley

Microfluidic 3D Co-culture of Estrogen Receptor Positive (ER+) Breast Cancer and Stromal cells Study Endocrine Resistance

 

One in eight women develop breast cancer within their lifetime with 80% of cases being estrogen receptor positive (ER+) allowing for patient treatment using endocrine therapies. Unfortunately, many patients develop a resistance to endocrine therapies; however, this mechanism is currently unknown. One potential driving factor for resistance is cell-to-cell communication between cancer and healthy cells in the breast tumor. The tumor microenvironment (TME) is a complex system containing multiple cell types interacting with each other and not only cancer cells, for example, fibroblasts, monocytes, stem cells, and T cells. Many initial co-culture studies have relied on two-dimensional (2D) culture platforms which fail to reproduce in vivo conditions including the cell-to-cell interactions found in the TME. New methods have been developed mimicking the 3D environment of real tumors in which using hydrogel scaffolds is one of the most promising approaches. In this study a thiol-acrylate (TA) hydrogel is used in a microfluidic droplet generator for the high-throughput production of small tumors (called 3D spheroids) ~250 um in diameter. This tool was used to evaluate and interrogate the interaction of Adipose Derived-Stem Cells (ASCs) with two different estrogen receptor positive (ER+) cancer cell lines (MCF-7 and ZR-75) in the TME and how they may enhance endocrine resistance. Additionally, this study looked at the role of patient age and BMI to find if these are features that enhance endocrine resistance in tumors resulting in greater cell proliferation. Co-culture 3D spheroids are grown for three days and then treated for two days with an endocrine therapy (ICI) followed by interrogation for proliferation levels using immunostaining with Ki67 and measuring the fluorescence intensity using Fiji. After image processing is completed, clustering methods and graphs were used to visualize trends across the spheroids with ASCs from donors of different ages and BMIs. Preliminary findings indicate that the presence of ASCs in the TME enhances levels of proliferation between the drugged populations and the control populations which suggest that ASCs may inhibit the purpose of the therapeutic.

Chemical Engineering

Tejasvi Tyagi, Braulio Andres Ortega Quesada, Adam Melvin

Backdooring AI Models with Data Poisoning

Deep learning techniques have made remarkable strides in achieving state-of-the-art
performance across various recognition and classification tasks. However, training these
networks is often computationally intensive, demanding extensive GPU resources and weeks of
computation. As a consequence, many users opt to outsource the training process to cloud
services or rely on pre-trained models that can be fine-tuned for their specific needs. This project
seeks to investigate the potential security risks introduced by outsourced training. Specifically,
we observe the possibility of creating a maliciously trained network, termed a "BadNet," which
exhibits exceptional performance on the user's training and validation data but behaves
maliciously when confronted with certain attacker-chosen inputs. This project aimed to create a
live working demo of a BadNet through the implementation in both a toy example involving
handwritten digit classification and a real-world scenario involving U.S. street sign classification.
We achieve a remarkably high backdoor accuracy without compromising a significant drop in
baseline accuracy of our models. Moreover, we accomplish this while only poisoning a relatively
small portion of the training data. Our findings underscore the potency of backdoors in neural
networks and their stealthy nature due to the intricacies of neural network behavior, making them challenging to detect.

Computer Science and Engineering

William Wei, Hao Wang

Improving Superwood by Optimizing the Delignification Process

The creation of superwood requires wood to be delignified and compressed at high temperatures. However, as the thickness of the wood increases, boiling the wood in the delignification solution becomes less effective. Due to this, it becomes more important to optimize the delignification process to remove as much lignin and hemicellulose as possible without damaging the cellulose too much. During this project, wood samples were delignified for different amounts of time and in different concentrations of solution and were then tested for cellulose, hemicellulose, and lignin contents. The samples were grinded into a powder with a ball mill grinder before testing. The two tests we ran were Fourier-transform infrared spectroscopy and x-ray diffraction. Specific wavenumbers of the FT-IR results correspond to compounds in cellulose, hemicellulose, and lignin and were checked. The x-ray diffraction test gives a crystallinity index to each of the samples which is greater the less hemicellulose and lignin is in the sample. The crystallinity index also decreases when cellulose is reduced, providing evidence of cellulose being damaged. Data from the FT-IR and XRD tests suggest that boiling the wood samples for more than three hours can cause notable damage to the cellulose. This is shown through the reduced peaks at wavenumbers 1098 and 1160 on the FT-IR tests and the reduced crystallinity index on the XRD tests as boiling time increased. Additionally, increasing the concentration of the delignification solution to 2.5M NaOH and .8M Na2SO3 increases degradation of cellulose.

Civil and Environmental Engineering

Hayden Willett, Hussein Alqrinawi, Hai Lin

Past Projects

Project Title Program Mentor
Optimization of Cell-Free Protein Synthesis Biological Engineering Samuel X. Adjei, Parker Hannan, Yongchan Kwon
Evaluation of Staining Method for Analysis of Cortical Bone Geometry Biological Engineering, Civil and Environmental Engineering Akshay Basireddy, Simone Muir, Beatriz Garcia, Alexander Lee, Kevin Hoffseth
3D-Printed Co-Culture Platform to Study Bacteria-Induced Chemotherapeutic Resistance in Breast Cancer Chemical Engineering Rocio Larenas Bustos, Stephanie Price, Emmaline Miller, and Adam T. Melvin
Solute Movement in Surface Water With Different Stream and River Geometries Civil and Environmental Engineering Emily Chen, Clint Willson
Keystroke Privacy Leakage From Zoom Meetings Computer Science Collin Clement, Long Huang, Chen Wang
Artificial-Intelligence-Aided Laryngeal Cancer Identification  Electrical Engineering Mariana Cuadra, Zheng Li, Huaizhi Wang, Jian Xu
3D-Printed Soil Bricks Inspired By Mud Dauber Nest Civil and Environmental Engineering Josephine Day, Joon S Park, Hai Lin
Single Cell Analysis of Deubiquitinating Enzyme (DUB) Activity Using a Droplet Microfluidic Trapping Array Chemical Engineering Veda Devireddy, Alireza Rahnama, Adam Melvin
Accelerating Reinforcement Learning Computer Science Ryan Ding, Hao Wang
The Role of the Genus Azospira in Transforming Arsenic-Containing Compounds Civil and Environmental Engineering Andi Hayes, Kali Martin, Bill Moe
Designing RNA Gene Circuits 
With Coherent Feedforward Loops
Chemical Engineering Benjamin Hogg, Xun Tang
Demonstrating UAV Propulsion Using an Aircraft and Flight Model With Hardware in Loop Approach Mechanical Engineering Nicole Lin, Shyam K. Menon
An Investigation into the Role of Fluid Shear Stress on Enhanced Cancer Extravasation during Metastasis Chemical Engineering Josie Ostrowe, Braulio Ortega Quesada, Adam T. Melvin
Nanoengineering Balsa Wood for Resilient Superwood Civil and Environmental Engineering Addison Schempf, Hussein Alqrinawi, Hai Lin
Reinforcement Learning in Flappy Bird Computer Science Kaitlyn Smith, Hao Wang
Steel Fiber Reinforcement in 3D Construction Printed Concrete  Construction Management Kaiser Stentiford, Ilerioluwa Giwa, Hassan Ahmed, Ali Kazemian
Detecting Hidden Security Threats With a Thermal Camera  Computer Science Kenzie Stentiford, Ruxin Wang, Chen Wang
The Impact of an Integrated Local Fan in a Central Cooling System on Occupant Thermal Comfort in Working Environments Construction Management Sarah Thomasa, Seddigheh (Tala) Norouziaslb, Amirhosein Jafari
Designing RNA Gene Circuits With Incoherent Type-1 Feedforward Loop Chemical Engineering Ahan Zaman, Xun Tang
Multimodal Label-Free Monitoring of Stem Cell Differentiation: Confocal Microscopy  Mechanical Engineering Laura Zapata, Sreyashi Das, Ram Devireddy
Geotechnical Analysis and Comparison of Recycled Glass Sediment for Coastal Restoration Environmental Engineering Louisa Zhu, Julia Mudd, Clint Willson

Project Title Program Mentor
Application of PCR to Detect Aromatic Hydrocarbon Producing Bacterial Populations in Sediment Samples from South Louisiana Civil and Environmental Engineering Bill Moe
Role of the Genus Azospira in Biological Nutrient Removal Civil and Environmental Engineering Tamara K. Martin, Bill Moe
Investigation of Physical and Mechanical Properties of a Mud Dauber Wasp Nest Civil and Environmental Engineering Joon S. Park, Hai Lin
Hurricanes vs. Oil Storage Tanks Civil and Environmental Engineering Sabarethinam Kameshwar
Effect of Sand Content on Metakaolin Based Geopolymers Construction Management Ruwa AbuFarsakh, Gabriel Arce
A Data-Driven Approach to Improving Energy Efficiency in Buildings Construction Management Amirhosein Jafari
Crystal Phases of Metal Oxide Materials Chemical Engineering Yuming Wang, James Dorman
Optimization of Hydrogel Identity and Composition in an Open-Air 3D Printed Microfluidic Device to Study 3D Cell Migration Chemical Engineering Kalena Nichol, Adam Melvin
Development of a Modular Microfluidic Device to Study the Effects of Fluid Shear Stress on ER+ Breast Cancer Chemical Engineering Blake Nassar, Adam Melvin
3D Bio-Printing of Tumor Phantom in the Larynges for Tumor Resection Training Applications Biological and Agricultural Engineering Kaushik Sunder, Michael E. Dunham, Jangwook P. Jung
The Effects of Bone Dye Techniques on Numerical Microstructural Analysis Biological and Agricultural Engineering Kevin Hoffseth
Droplet Interaction with Propagating Shockwaves Mechanical Engineering Shyam Menon
Colorimetric and Spectroscopic Sensing of Biomarker for Cystic Fibrosis Using a Smartphone Mechanical Engineering Elnaz Sheik, Manas Ranjan Gartia 
Preventing Handheld Device Distraction for Drivers Using Smartphone Motion Sensors Computer Science Chen Wang
Preventing Driver Distractions Via Acoustic Sensing Computer Science Long Huang, Chen Wang
Machine Learning Methods on Raman Spectroscopic Cancer Data for Early Diagnosis Electrical Engineering Zheng Li, Jian Xu

Project Title Program Mentor
Simulating Cortical Bone Structure in Large Vertebrates Biological Engineering Kevin Hoffseth
Microstructural Geometry and Damage Detection in Cortical Bone Images Biological Engineering Kevin Hoffseth
Characterization of Fluorescent Proteins Produced in the E. coli Cell-Free Protein Synthesis System Biological Engineering Yongchan Kwon
Meta-Analysis of Cardiac Extracellular Matrix Proteins: Information Extraction for 3D Bio-printing Biological Engineering Philip Jung
Dynamic Photoluminescence Response of Dipole-Modulated Rare Earth Doped Core-Shell Nanoparticles to Local Changes in Temperature and Solution pH Chemical Engineering James Dorman
Machine Learning-Based Feature Analysis and Classification for ICG-Assisted Vibrational Spectroscopic Data of Pancreatic Carcinoma Electrical Engineering Jian Xu
3D Tumor Spheroid Generation Using a Droplet Microfluidic Device Chemical Engineering Adam Melvin
Circulating Microfluidic Co-Culture Device for the Dynamic Analysis of the Tumor Secretome Chemical Engineering Adam Melvin
Development of a Modular Microfluidic Platform to Investigate the Role of Fluid Shear Stress on Cancer Cell Phenotype Chemical Engineering Adam Melvin
Using Pulsed UV Light for Enhancing Advanced Oxidation Water Treatment Environmental Engineering Samuel Snow
Using Pulsed UV Light for Enhanced Water Disinfection Environmental Engineering Samuel Snow
Shockwave Induced Droplet Breakup Mechanical Engineering Shyam Menon
Characterization of Animal Nest-Building Geomaterials Civil Engineering Hai Lin
Breath Monitoring: Analyzing Breathing with Wireless Bluetooth Earbuds Computer Science Chen Wang
Evaluation of the Field Performance of Stabilized and Non-Stabilized Asphalt Overlays in Louisiana Construction Management Momen Mousa
The Use of RAP and WMA Mixtures in South-Central States: Challenges & Limitations Construction Management Husam Sadek
Variability and Uncertainty of Overlay Tester Testing Data, Analysis, and Results Construction Management Husam Sadek

Photos

HSSR Presenter and Board HSSR Presenter and Board HSSR Presenter and Board HSSR Presenter and Board
HSSR Presenter and Board HSSR Presenter and Board HSSR Presenter and Board HSSR Presenter and Board
HSSR Presenter and Board HSSR Presenter and Board HSSR Presenter and Board HSSR Presenter and Board
HSSR Presenter and Board HSSR Presenter and Board HSSR Presenter and Board HSSR Presenter and Board
HSSR Presenter and Board HSSR Presenter and Board HSSR Presenter and Board HSSR Presenter and Board
HSSR Presenter and Board

HSSR Finalists

Finalists: 

Josie Ostrowe, St. Joseph’s Academy
Kaitlyn Smith, Baton Rouge Magnet
Veda Devireddy, Baton Rouge Magnet

   

Contact

The program administrators are responsible for the facilitation of the program from start to finish by creating the policy/structure, providing regular communication to all stakeholders, serving as the key liaisons between all stakeholders, and generally supporting/directing the program throughout each cycle.

Program Administrator Contact Info:

Raynesha Ducksworth
Assistant Manager
225-578-5335
rducksworth@lsu.edu

Corina Barbalata, PhD
Assistant Professor of Mechanical Engineering