Timothy Droddy
Dr. H. Lee Sawyer
The Black-Scholes model is a mathematical model used to provide specific details for pricing European Options. The model describes European Options through a specific partial differential equation. This paper focuses on the derivation of the equation, the many different ways to solve the equation and the interpretation of solutions of the model.
Clayton Fry
Willis-Knighton Cancer Center
Dr. Joseph Dugas
Dr. Yohan Walter
Dr. Terry Wu
Prostate cancer patients undergoing radiotherapy are at risk of rectal toxicity due to the close proximity of the rectum to the prostate. Rectal spacers have been introduced to increase prostate–rectum separation and reduce rectal dose; however, differences in dosimetric performance among spacer types remain an area of active investigation. This study performs a dosimetric comparison of Bioprotect and SpaceOAR rectal spacers in proton therapy for patients with prostate cancer. Anonymized treatment plans including prostate and seminal vesicle (PSV) target coverage were analyzed using clinically relevant planning techniques. Dose–volume histogram metrics for the prostate target and organs at risk, particularly the percentage of rectal volume receiving specified dose levels, are evaluated and compared across both spacer types while maintaining consistent prescription dose and target coverage among patients. Primary endpoints include rectal dose reduction and target dose coverage. This study provides quantitative insight into the dosimetric advantages and limitations of different rectal spacer options, with the goal of informing clinical decision-making in prostate cancer radiotherapy.
Emma Lamoureaux
Dr. Pedro Derosa
A memristor, named for the blending of memory and resistor, is the non-linear connection between electrical charge and magnetic flux linkage by means of a two terminal electrical component. Despite being proposed as a theoretical concept alongside resistors, capacitors, and inductors in 1971, the memristor lags behind progress related to its fellow electrical components. One of such aspects features its design, with physical models first being produced in 2008. By using Density Functional Theory (DFT), a computational method of quantum mechanical modeling, the design process could be optimized for the production of such models, computing band gap structure diagrams as well as the total energy to compare the proposed materials considered for the design. Programs such as QuantumATK and MobaXterm, utilized to build potential materials and process jobs, respectively, performed the calculations of the extensive list of proposed materials with the potential of creating an effective design. In conjunction with the results of the testing of physical models made from materials studied through DFT, ideal patterns in the computations provide information regarding the effectiveness of materials chosen and guide the production team on the most optimal choice leading forward.
Laurel Larramendi
Willis-Knighton Cancer Center and LSU Health Shreveport
Dr. Joseph Dugas, Yohan Walter, Olivia Moncrief, and Emily Warren
Previous studies have shown that kidney tumors are vulnerable to changes in the lungs,
bowel, and skin. The changes that can occur in these places have affected the depths
of the tumors, causing issues with the initial treatment plans. We present a study
on how adaptive radiotherapy (ART) benefits X-ray treatment of kidney tumors by adjusting
the treatment plan to the current changes of the patient’s tumor size, shape, and
positioning. Data from a previous study about ART for proton therapy was re-arameterized
for treatment on a linear accelerator. The results have yet to be discussed.
Dylan Mitchell
Christopher Lin
Consider a configuration space Ω whose boundary is made up of an asymptotically straight curve and its reflection or its rotation about its axis of symmetry (for two and three dimensions respectively). We prove that by imposing an infinite net-area condition on the profile curve along just one end of Ω, the Dirichlet Laplacian −∆ on Ω has at least one isolated eigenvalue (hence at least one bound state) below a natural threshold. The varying eigenvalue of the cross-section along the axis of symmetry plays a central role in our analysis, and integrals of the Bessel function are also of crucial importance.
Landon Moore
Dr. Arden Moore
A Physically Uncloneable Function (PUF) is a hardware security method for creating
random and unique digital "fingerprints". A PUF leverages random physics phenomena
in the silicon chip manufacturing process to create a "black box” function. The structure
of the chip is prompted in a specific way, called a challenge, and the output of the
challenge, called the response, is measured and kept secure. A collection of challenge
response sets create a unique and hopefully irreplicable encryption key. Our PUF uses
chips with silicon-compound coatings as the "black box", where lattice mismatches
at the boundary create random patterns of mechanical strain on the surface. A polarized
infrared beam is incident on the coating, and the strain causes a birefringence effect
on the beam. The optical transmission through the interface is greatly affected by
the alignment between the polarization and the strain. The challenge is then a collection
of elliptical polarization states, and the response is a 2-dimensional grid of outgoing
optical intensity, called a speckle pattern. A brief summary on the history and need
for PUFs is discussed, as well as categorizing
them and outlining notable current methods. Some similar work is discussed, which
will then motivate the concepts of the proposed PUF. The PUF is experimentally tested
at a small scale, "proof-of-concept" level. As such, we end with a discussion on the
strengths and weaknesses of the PUF and future work to improve, test, and implement
it.
John Sibley
LaSpace
Dr. Elisabeth Fatila
Scintillating materials are widely used for radiation detection in high-energy physics, security screening, and medical imaging. In this work, molecular Ce3+ complexes were examined as a possible approach for combining the high light output of inorganic scintillators with the fast response of organic systems. Traditional inorganic crystals are often hygroscopic and difficult to process, while molecular Ce3+ complexes offer improved solubility and chemical tunability. These properties allow for detailed photophysical characterization and suggest potential compatibility with polymer hosts. The optical behavior of selected Ce3+ complexes was studied using UV–vis absorption spectroscopy, steady-state fluorescence, and time-resolved luminescence techniques. Data collected included absorption and emission spectra, emission lifetimes, and relative quantum yields. The observed emission exhibited fast decay components consistent with Ce3+-centered luminescence, with lifetimes on the nanosecond timescale. Variations in emission intensity and lifetime were associated with differences in ligand environment. Results from this study can inform the design of future molecular Ce3+ scintillators by clarifying relationships between structure and photophysical behavior.
Ethan Thomas
Dr. Rakitha Beminiwattha
Scintillation is a process in which photons are emitted from a material after absorbing energy from a charged particle. There are many applications of scintillation in the medical and nuclear fields of study. In this project, we study the time variation of the light yield using different configurations of Wavelength Shifting (WLS) fibers. The objective is to study different WLS fiber-based designs and compare the time variation of the light yield. Our goal ideal time variation is a few nanoseconds. The measurements are collected using WLS fibers, which are then coupled to a photomultiplier tube (PMT).
Jesse Webb
Eric A. F. Reinhardt and Sergei Gleyzer (University of Alabama)
In high-energy physics (HEP), training models to classify and regress physical properties of particle collision data is highly data intensive, with training sets generally produced using Monte Carlo simulation. Machine learning models like Generative Adversarial Networks (GANs) can significantly reduce compute cost, but an open problem is whether Quantum GAN models will outperform their classical counterparts. Quantum data encoding techniques like amplitude embedding allow quantum machine learning (QML) models to scale exponentially with available resources, providing a promising avenue for achieving larger effective model sizes and representing physically complex datasets. Separately, QML models have shown potential to capture implicit correlations and features of both limited and quantum datasets. Exploring these principles with a physically-motivated training paradigm, we implement hybrid classical-quantum GAN models which respect detector symmetry and physically-seeded generator output. We present a comparison of the classical and hybrid quantum GAN models for the simulation and classification of low-level jet events in a hadronic calorimeter, showing that the QGAN performs comparably to its classically equivalent GAN in image realism while expressing trade-offs between statistical matching and discriminator event classification.
David Yu
Dr. Rakitha Beminiwattha
The Measurement of a Lepton-Lepton Electroweak Reaction (MOLLER) is a planned fixed-target experiment at Jefferson Laboratory that will measure the parity-violating asymmetry in electron electron (Møller) scattering. By determining the weak mixing angle from the parity-violating asymmetry parameter in a lower energy scheme (Q ≈ 0.1 GeV) with unprecedented precision, MOLLER will provide a stringent test of the Standard Model and sensitivity to new physics. Achieving this goal requires a highly reliable simulation framework, remoll. Currently, in the literature, gaps remain in the availability of high-quality reference data and a standardized procedure for benchmarking. Benchmarking refers to comparing simulation outputs to well-established theoretical or experimental results and is important in establishing confidence in remoll as a competent tool. We develop scripts that compare simulation results across multiple versions of remoll. Differential cross sections and various quantities for electrons, neutrons, photons, positrons, and background electrons are selected as benchmark observables across various locations in the MOLLER experimental setup, enabling regular checks for remoll updates. Ultimately, these efforts ensure the reliability of the current remoll simulation and any future iterations.
Benjamin Eunice
Dr. John Shaw
The emission-line spectra of Ba and B[a] stars present a unique set of footprints such that their analysis presents insight into the late stage main sequence evolution of massive stars, and the nature of their interior structures. Spectroscopic signatures were taken of a catalog of stars, and their spectra were measured against known emission lines of various compounds. It was found that the chemical makeup of these stars correspond with the existing literature of their types. Some of these lines, however, did not line up with such findings and should be further investigated as to the origin of these discrepancies.