22505 research outputs found
Sort by
Animating the Mechanism of Action of the Novel Glioblastoma Vaccine Survaxm
Get PDFGlioblastoma (GBM) is a deadly type of malignant brain tumor affecting over 11,000 people in the United States each year. Treatment options for GBM are limited, with the majority of patients receiving surgery and chemoradiation, and surviving around 15 months. New therapies are desperately needed. A novel peptide vaccine, SurVaxM, has shown promise in phase I and II clinical trials, and is currently under investigation in a Phase IIb trial (the “SURVIVE” trial) for newly diagnosed GBM. As SurVaxM moves closer to FDA approval, there is a need for education about the agent and how it works. In order to provide this information to various audiences (patients, caregivers, doctors, scientists, investors and pharmaceutical reps), an animated video was created. Assets were constructed in ZBrush and Cinema 4D, and molecular models were obtained from RCSB Protein Data Bank and modified in Chimera X. A human brain was modeled using MRI images reconstructed in Slicer software. Animation was done in Cinema 4D and rendered using Redshift, and compositing was done in Adobe After Effects. The finished product is a two minute-long video which provides a narrated overview of SurVaxM’s construction and mechanism of action in the body, with animations illustrating the narrated concepts. The completed video was posted on the home page of the main SURVIVE trial sponsor, Mimivax, www.mimivax.com, and also shared on Twitter and Linkedin by Mimivax’s CEO, Dr. Michael Ciesielski
Innovating Criminal Justice: Predictive Analytics for Effective Recidivism Management
Get PDFRecidivism—the tendency of previously convicted individuals to reoffend—poses significant challenges to criminal justice systems worldwide. In the United States, a study by the Bureau of Justice Statistics revealed that approximately 68% of released prisoners were rearrested within three years, and 83% within nine years (Bureau of Justice Statistics, 2018). These high rates underscore systemic deficiencies in rehabilitation processes and resource allocation. Traditional risk assessment tools often rely on static factors, failing to account for the dynamic and individualized nature of reoffending risks. This limitation highlights the need for innovative, data-driven methodologies to enhance offender management strategies. This study proposes a framework that leverages advanced data analytics and predictive modeling to address these challenges. By integrating historical and real-time data, the framework aims to improve the accuracy of risk predictions, support dynamic monitoring, and enable the development of tailored intervention strategies. Techniques such as regression analysis and machine learning algorithms will be applied to identify key predictors of recidivism and assess the effectiveness of interventions. The framework emphasizes scalability and adaptability, ensuring applicability across various correctional and community supervision settings. A mixed-methods research design will guide the study. Quantitative techniques, including statistical modeling and machine learning, will analyze offender data and validate predictive models. Complementary qualitative methods, such as expert interviews and case studies, will provide contextual insights and ensure practical applicability. Legal and ethical considerations, including data privacy and algorithmic fairness, will be addressed to align the framework with principles of justice and proportionality. Anticipated outcomes include the development of a proactive offender management system that shifts focus from reactive responses to preventive measures. By optimizing resource allocation and delivering customized interventions, the framework aims to reduce recidivism rates while enhancing public safety and offender rehabilitation. The integration of data-driven approaches has the potential to empower policymakers, rehabilitation practitioners, and community supervision officers with tools to improve decision-making and resource management. This research contributes to the broader goal of creating a more equitable and efficient criminal justice system. It bridges gaps in current methodologies by applying advanced computational methods to offender management while addressing associated ethical challenges. Ultimately, the findings aim to support the development of systems that balance technological advancements with social and legal imperatives, fostering long-term societal benefits
Modeling Femtosecond Laser Interaction with Glass for Optical Fabrication
Get PDFThe fabrication of precision optics is critical for a wide range of applications, including biosensors, virtual and augmented reality, medical imaging, and micro-electronics. However, it is challenging to meet the demands of these applications with conventional chemical and mechanical fabrication methods, as they can introduce chemical waste, mid-spatial-frequency errors, and subsurface damage that degrades image quality. Femtosecond lasers have emerged as a promising alternative, offering fast, non-contact, and chemical-free machining with single digit nanometer precision. This thesis presents a computational model designed to investigate the interaction process between a high-intensity femtosecond laser pulse and dielectric material. A pulse propagation model is implemented to simulate material response, predicting time and intensity dependent optical properties. The model incorporates the Keldysh theory of photoionization to accurately predict the generation of free electrons under laser irradiation, enabling precise estimates of material removal and crater formation. The predicted crater morphology shows strong agreement with experimental observations. To further understand the thermal effects of laser processing, a two-temperature model is implemented to simulate heat-affected zones and their impact. Additionally, the model is extended to multi-pulse simulations under gigahertz burst mode operation, allowing for analysis of plasma and thermal accumulation. These insights contribute to the optimization of femtosecond laser parameters for precision optical fabrication with minimized damage and improved control
Macroscopic Roughness Modeling of Satellite Multi-Layer Insulation Reflectance
No full textReflectance modeling of specular surfaces with irregular geometries remains difficult to solve for an array of applications. In the growing space industry, there are many spacecraft materials with high specularity and varying levels of roughness. Modeling these materials in simulations becomes difficult as any realistic amount of roughness is added, necessitating more accurate physical models to represent how light is interacting with these surfaces. While some work has been done to attempt to characterize the spectral signatures of the materials themselves using hyperspectral imaging systems, the scope of these efforts has been fairly limited and not representative of the conditions of on-orbit spacecraft geometries. Many models that attempt to characterize reflectance from roughness do not account for the effects of multiple-scattering of light due to the surface geometry, and the few that do only provide empirical solutions to the problem for specific scenarios/materials. There are multiple approaches that attempt to account for multi-facet scattering (typically Fresnel/specular models vs. those that consider diffusely scattered light), but none are exact. This has to do with each approach’s ability to account for both diffuse and specular scattering due to directional light. Using spacecraft Multi-Layer Insulation (MLI) materials with varying levels of roughness applied to each, the Bidirectional reflectance distribution function (BRDF) of each material in each roughness state was collected at multiple illumination angles in order to gather enough data to fit a recently developed model from the literature and to propose adjustments to said model. Through a discussion of the effectiveness of the model’s components accounting for single and multiple scattering, as well as the nature of the materials themselves in roughened conditions, a path forward for modeling such roughened specular materials is identified for the future capability of accurately modeling and simulating such scenarios
Mechanisms for oxygen vacancy defect migration in SrTiO3/NiO heterostructures
Get PDFPerovskite-based oxide heterostructures display promising properties resulting from interface phenomena, making them good candidates for next-generation solid oxide fuel cell electrolytes. Amongst the different features exhibited by these interfaces, misfit dislocations play an important role in influencing ionic transport, yet their role remains poorly understood, which is the case in rock salt-perovskite interfaces too. In SrTiO3/NiO heterostructures, to comprehend interface ionic transport, we investigate oxygen vacancy migration near misfit dislocations. To this end, we developed a high-throughput framework that integrates atomistic simulations with nudged elastic band method to predict migration energy barriers across disparate interface atomic environments. By comprehensively mapping activation energy barriers across different interfacial chemistries and asymmetric structural features, we explore how the dislocation structure, which is dependent on the local interfacial chemistry, modulates vacancy migration. This study aims to shed light on the role of dopants, oxygen vacancies, interfacial chemistry, and extended defects in shaping ionic migration at the atomic scale. Misfit dislocations are often considered thermodynamic sinks for oxygen vacancies, oftentimes hindering ionic conductivity at such interfaces. We report dynamic behavior at interfaces that are largely dependent on the local coordination environment, challenging this conventional perspective. The study further attempts to bridge the crucial gap in understanding interface-governed ionic transport mechanisms in complex oxide heterostructures while exploring novel computational techniques for characterization of misfit dislocations
Eviction Optimizations and Dynamic Lease Adjustments in a Single-Level Lease Cache
No full textIn a computer system, a hierarchical memory aims to give processing cores the illusion that they have access to the largest capacity memory, i.e., the disk, at the shortest access time, i.e., that of the memory level closest to them - the one-, two-, or three- levels of cache. Therefore, an efficient operation of the cache is key to lowering the average memory access time. Toward this end, lowering the miss rate, which is the principal metric of cache performance, is the goal of any cache implementation. Traditional caches are automatic and reactive. Several prior iterations of this work have introduced and explored a special kind of programmable and (implicitly) prescriptive cache: a lease cache. In this prior work lease values are static, i.e., would not change at run-time, and in the absence of expired leases, the target cache line of a forced eviction (victim) is selected pseudo-randomly. In this work, we introduce and present the hardware implementation and test results of two alternative eviction optimizations and dynamically (at runtime) adjusted lease values. The former two decide a eviction target by tracking and using the history of utilization of all blocks of data at word level. The latter augments static with dynamic information to adjust reference lease values. Test results show equal or lower miss rates than those with static leases, overall lower than PLRU, and in many cases close to OPT
Language Alignment and Career Outcomes in International Military Education: A Study of United States Military Academy Graduates
No full textHost country/sending country language alignment is an understudied topic within the field of international military education. While individual-level language barriers are recognized, this study asks how national-level language proficiency levels can influence international military graduates’ retention. International cadets from countries with lower proficiency in the host nation’s language may be selected primarily for their language skills rather than their military potential. Moreover, the acquisition of the host nation’s language and culture during international cadets’ studies may increase the attractiveness of private-sector opportunities upon their return home. At the same time, countries with lower proficiency levels in the host language may invest more heavily in cadets and provide them better retention incentives after graduation. To explore these logics, this study examines the impact of national-level language proficiency on the career prospects of international military students attending the United States Military Academy (USMA). Using average national TOEFL scores as a proxy for English proficiency, I compare the career attainment of international graduates from countries with varying levels of English proficiency. Quantitative analysis is complemented by interviews with international graduates and US officials involved in security cooperation. This research has significant policy implications. International military exchanges aim to foster future military cooperation, but premature departures from military service can undermine this goal. By understanding the factors influencing career outcomes, policymakers from both host and sending countries can implement strategies to enhance military retention rates and maximize the benefits of these exchanges
Test Data: Raised or Recessed? Finding the Optimal Gate Architecture for Improving the Static Performance of Graphene Transistors
No full textAs silicon CMOS technology approaches its scaling limits, graphene offers a compelling alternative as the active material channel in transistors due to its high carrier mobility and atomically thin profile, which provide strong electrostatic control and promise high-performance analog applications. However, roadblocks such as device-to-device variation, high contact resistance, poor dielectric interfaces, and non-uniform graphene quality have limited the adoption of graphene field effect transistors (GFETs). Hence, further investigations are required for mitigating these issues at a material, e.g., by improving graphene transfer, and device level, e.g., by finding an appropriate gate architecture. In this work, we directly compare two GFET structures through a controlled, side-by-side process split to evaluate the impact of gate stack architecture: raised vs. recessed buried local gate, in which both structures use hBN as the gate dielectric. Benchmarking is performed in terms of device performance and device-to-device variation. While the top-performing devices remain similar across the two proposed structures, significant statistical differences are seen in terms of device performance and yield in the two populations studied. A total of 256 identical devices from each gate architecture are electrically tested and characterized for a statistically significant comparison. The most significant difference is seen in the Dirac voltage, which is reduced from 1.2 V to 0.7 V with the recessed architecture, making it more suitable for low-power analog applications. Average hole mobility increases from 3,383 cm2V-1s-1 to 4,794 cm2V-1s-1, and device yield increases from 54.4% to 65.1%. Physical analysis, which includes spectroscopy and hysteresis measurements, indicates that these improvements are due to the proposed planarized gate architecture and reduction of interface defects. This study shows that direct statistical comparison studies of process conditions can help identify favorable process conditions to improve the manufacturability of graphene-based transistors
Spectroscopy Pre-trained Transformer (SpecPT): A Universal Spectroscopic Analysis and Redshift Measurement Framework
No full textSpectroscopic surveys are essential for measuring galaxy redshifts and probing the physical processes driving galaxy evolution. As datasets grow in scale and complexity, traditional analysis methods face increasing limitations, motivating the development of scalable, data-driven alternatives. This thesis presents SpecPT, a transformer-based deep learning framework for general-purpose spectroscopic analysis, with a focus on redshift estimation. The model is first trained on DESI Early Data Release spectra from the Bright Galaxy and Emission Line Galaxy samples, jointly performing spectral reconstruction and redshift regression while learning a latent representation that captures the intrinsic properties of galaxies. SpecPT is then extended to a unified model trained across BGS, ELG, and LRG samples, achieving robust performance across diverse galaxy types and redshifts without class-specific supervision. Finally, the DESI-trained model is fine-tuned on a small set of Keck/DEIMOS spectra from the COSMOS field, demonstrating strong transfer learning capabilities and accurate redshift predictions even in low-data regimes. SpecPT establishes a scalable and adaptable foundation model for spectroscopy capable of robust inference across instruments, redshifts, and data quality. The results lay the groundwork for future applications to space-based grism spectroscopy and downstream physical property estimation