University Honors

In ethical philosophy, the ìkilling and letting dieî argument discusses whether a moral distinction exists between killing someone and letting them dieóan extension of the ìdoing and allowingî argument. There is some sense of intuition in which we understand that this distinction exists between doing harm and allowing harm to be done to another. However, this intuition disappears when shifting the discussion from two parties to one partyóin other words, it is not intuitive whether there is a distinction between killing oneself and letting oneself die. This capstone project is concerned with determining whether a possible distinction between killing oneself and letting oneself dieóor in synonymous terms, an ìactiveî and ìpassiveî suicide. In order to examine a possible distinction between the two, an intention-based account of suicide was created, under which suicide is defined using terminology such as want, means, end goal, and intention. Utilizing this new framework for suicide, we revisited our notions of active and passive suicide in order to determine if a distinctive nature between the two could possibly exist. Discussion highlights that, with the want and end goal being the same in both cases, a possibility for a distinctive nature of active and passive suicide exists on the basis of intention and action.

Breast cancer survivors often experience significant psychosocial burden after treatment, including depression and anxiety. Although post-mastectomy breast reconstruction (PMBR) has been associated with improved body image and quality of life, existing literature on psychiatric outcomes after reconstruction remains mixed, with some studies reporting psychosocial benefits while others finding no significant differences. Additionally, much of the current literature focuses on patient-reported psychosocial well-being rather than incident psychiatric diagnoses reported in large clinical datasets. Using the TriNetX Collaborative Network, this study evaluated incident depression and generalized anxiety disorder (GAD) among women undergoing PMBR within five years of surgery, with marital status serving as the primary factor of interest. Multivariable Cox regression and propensity score-matched subgroup analyses were performed to evaluate predictors of psychiatric outcomes and compare PMBR patients with mastectomy patients without reconstruction. Married PMBR patients demonstrated significantly lower risk of depression and GAD compared with divorced and never-married patients, while divorced patients exhibited the highest relative psychiatric risk across analyses. Age and race/ethnicity were also associated with psychiatric outcomes. In direct matched comparisons, PMBR did not significantly differ from mastectomy with respect to depression or GAD risk, suggesting that reconstruction itself may not independently reduce psychiatric burden. Overall, these findings suggest that psychiatric outcomes following breast cancer surgery are more strongly influenced by social and demographic context, particularly marital status, than by reconstruction alone. These results highlight the importance of psychosocial screening and targeted mental health support for higher-risk patient populations following breast cancer treatment.

Opioid-related events are a leading and preventable contributor to maternal mortality in the United States, with the postpartum period representing a high risk yet under-studied interval. Obstetric patients frequently receive opioid prescriptions following delivery; however, the incidence and drivers of postpartum opioid-related risks remain poorly defined within the available anesthesiology literature. This study aims to quantify postpartum opioid-related risk using real-world electronic health record data to inform opioid stewardship in obstetric anesthesia.

Deepfakes have become increasingly realistic with the rise of diffusion models, vision transformers, and multimodal generators, posing major risks to security, trust, and authenticity in digital media. While existing detectors achieve strong benchmark accuracy, their performance deteriorates under real-world distortions such as compression, transmission loss, or bitrate fluctuations. This literature review traces the evolution of deepfake detectionófrom early spatial and temporal CNNs to frequency-, latent-, and transformer-based architecturesóand examines robustness-oriented frameworks including ADD, QAD, BZNet, and DPL. To address the gap between laboratory conditions and deployment environments, this study introduces two complementary research directions. First, an adaptive detection framework, the MDN-Guided Noise-Aware Modular Ensemble, models test-time uncertainty using a Mixture Density Network to dynamically select noise-specific fine-tuning modules. Second, a new dataset, DeFND (Deepfake Forensics under Network Degradation), captures authentic Wi-Fi and cellular transmission artifacts by streaming deepfake videos through real WebRTC pipelines. Together, these contributions provide a foundation for evaluating and improving deepfake detectors under realistic network and corruption conditions, advancing the pursuit of reliable, uncertainty-aware detection systems for practical deployment.

Asian Americans in the United States are underrepresented in both politics and public administration. One of the reasons could be the type of education that Asian Americans obtain leading them to a political career. In fact, nearly all members of Congress have a bachelor's degree, coming in with 94% of representatives; most also have a graduate degree, with 64% of the elected officials holding one. Furthermore, the most common subjects that legislators study are political science, history, and economics in their undergraduate degrees. This shows that there is a ìtypicalî career pathway in order to work in politics or become a policymaker through public administration. I hypothesize that the underrepresentation of Asian Americans in politics is due to lower enrollment rates in typical pathway majors/degree programs that lead to careers in politics and public administration. These lower enrollment rates are an individual-level explanation of lower political participation, which is possibly affected by familial influence. By looking at the connections between the traditional educational pathway to becoming involved in politics, the underrepresentation of the Asian American community could be explained.

Liquid argon (LAr) has wide applications in nuclear and particle physics, particularly in dark matter research, where it enables the detection of ionization and scintillation signals while minimizing radioactive background. Doping liquid argon with xenon can enhance ionization and increase photon yield, lowering the energy threshold and improving signal strength. This amplification of light output enhances detector sensitivity, increasing the likelihood of detecting weak interactions, such as those from potential dark matter candidates. However, the system presents challenges due to the narrow liquefaction temperature range and the potential for xenon to separate from argon or to form ice, which can introduce instabilities. This paper discusses a cryogenic system designed to increase liquid argon production while maintaining a stable liquefied state to address these challenges. This research involves developing a stable, functional cryogenic condenser to ensure a continuous supply of high-purity liquid argon, optimizing storage conditions, and refining the argon-xenon doping process. Through this research, I aim to advance the capabilities of liquid argon detectors for experiments such as DarkSide-LowMass and contribute to future applications in medical, engineering, and scientific research, including aerospace applications in satellites and rocketry. Preliminary testing has demonstrated successful nitrogen condensation and partial argon liquefaction. Ongoing work will address the challenges associated with stabilizing the system and continue investigating scalable xenon doping for applications in experiments such as DarkSide-LowMass. This research will contribute to the development of next-generation liquid argon detectors, broader cryogenic engineering, and particle physics research.