22505 research outputs found
Sort by
On-The-Go Handbag: Sustainable & Biodegradable Handbag Design
Get PDFHandbags have been an ever-present accessory in society for centuries, coming to symbolize not only fashion trends but status and femininity. They have evolved to hold extreme significance, being able to show a snapshot of who a woman is by taking a simple look inside. Purses have become a companion for consumers in many ways, accompanying them throughout important days and events, providing talking points, and even compliments for the user. However, these constant companions come at a cost. The fashion industry and the accessories produced within are an immense cause of environmental pollution. This provides an ever-present internal fight for the consumer in which sustainability and fashion concerns battle with one another for purchasing power. In seeking to create a consumer product that would provide an alternative to this vicious cycle of production and its resulting waste, the On-The-Go bag was born. A completely biodegradable handbag that still addresses the fashion needs of the consumer. The current market lacks products which combine these two ideals, tending to either focusing on environmental impact or style. Using locally sourced wool as the base material with vegetable-tanned leather accents allows this bag to be completely composted at home after its decades-long life as a bag. Principles of slow fashion are explored to further increase sustainability and long-term impacts of the bag. The findings of this research support and address the prevalent issues of sustainability in numerous ways through a low production number, no chemical processes, and the ability to return nutrients to the earth at the end of a product’s life. The On-The-Go bag enters the fashion market in direct opposition to current practices, hoping to give more sustainable purchasing power to consumers
Robust Multimodal Fusion for Oncology
No full textMultimodal deep learning models have the potential to significantly improve survival predictions and treatment planning for cancer patients. These models integrate diverse data modalities using early, intermediate, or late fusion techniques. However, many existing multimodal models either underperform or show only marginal improvements over unimodal models. To establish the true efficacy of multimodal survival prediction models, it is critical to demonstrate consistent and substantial advantages over unimodal counterparts. In this thesis, we introduce the Robust Multimodal Survival Model (RMSurv), a novel discrete late fusion model that leverages synthetic data generation to compute time-dependent weights for various modalities. RMSurv utilizes up to 6 distinct data modalities from the Cancer Genome Atlas Program (TCGA) non-small cell lung cancer and the TCGA pan-cancer datasets. In our experiments, RMSurv outperforms the best unimodal model’s Concordance index (C-Index) by 0.0273 on the 6-modal TCGA Lung Adenocarcinoma (LUAD) dataset. Existing late and early fusion methods improved the C-index by only 0.0143 and 0.0072, respectively. RMSurv also performs best on the combined TCGA non-small-cell lung cancer dataset and the TCGA pan-cancer dataset. The key innovations of RMSurv are the calculation of time-dependent late fusion weights using a synthetically generated dataset and a new statistical feature normalization technique to enhance the interpretability and accuracy of discrete survival predictions. These advancements underscore RMSurv’s potential as a powerful approach for survival prediction, establishing robust multimodal benefits and setting a new benchmark for survival prediction models in pan-cancer settings
An Investigation of Electrical Surface Impact as a Three-Dimensional Low-Energy Defibrillation Method
Get PDFVentricular fibrillation is a medical emergency that leaves the heart unable to beat properly, and is fatal within minutes if left untreated. Furthermore, it’s the leading cause of sudden cardiac arrest according to the NIH, killing over 400,000 Americans every year. Previously, we’ve proposed a new low-energy defibrillation method to restore the heart to its normal rhythm during fibrillation. Our method required further investigation due to the presence of several important dynamical processes. In this dissertation, we report the results of these investigations. First, we will present our results related to a novel behavior called “flopping,” which occurs when a 3-D rotating electrical wave thought to be present during fibrillation reorients its axis of rotation, leading to our method’s failure. Second, we will investigate the effects of the alternation of the electrical properties of successive beats of the heart. Third, we will analyze optical mapping voltage data related to our method, generated in-lab at Georgia Institute of Technology. We will be looking for evidence of dynamical changes in wave propagation patterns consistent with how we expect our method to work. These studies will help direct future development of our low-energy defibrillation method
Developing MELD-accelerated Molecular Dynamics Protocols to Simulate the Binding of the P53-Derived Ligand to the MDM-2, X Protein
Get PDFIn this study, we focus on developing computational methods to predict protein-ligand binding affinities, with applications in peptide drug discovery. Molecular Dynamics (MD) simulations can capture the complex conformational behavior of proteins, but their high computational cost limits their efficiency. MELD, or Modeling Employing Limited Data, is a Bayesian approach that integrates external information to accelerate sampling of low- energy, high-probability conformations. Building on previous work by Morrone et al., which successfully applied MELD to P53-MDM2 complexes, we hypothesize that we can effectively compute the relative binding affinities while reducing steric clashes and mitigating the effect of slowed diffusion on simulation convergence time. We present optimized MELD protocols that reproduce Morrone’s results within 1% of the target value, supporting the method’s accuracy and efficiency for peptide-based drug discovery
Modelling and Optimization of a Desiccant Cooling System for Industrial Applications in Dubai
Get PDFThe study examines the modelling and optimization of solar-assisted desiccant cooling systems (SADCS) specifically designed for industrial applications in Dubai. Four system configurations were evaluated under Dubai\u27s extreme climate using TRNSYS 18 simulation software: classic ventilation, variable percentage recirculation, ventilation with a sensible heat exchanger that utilizes exhaust air to preheat the feed of the auxiliary heating air, and recirculation with a sensible heat exchanger that also utilizes exhaust air for preheating the auxiliary heating air feed. A parametric study comprising 60 simulation cases was performed, examining variations in desiccant wheel effectiveness, airflow rates (3–5 ACH), regeneration temperatures (50–80 °C), and supply air mixing ratios of (0–100%). The goal was to achieve thermal comfort (T₅ ≤ 25 °C, RH₅ ≤ 68%) while minimizing energy usage and emissions. A parametric investigation was performed by adjusting desiccant wheel effectiveness, air flow rates (3-5 ACH), regeneration temperatures (50-80 °C), and fresh air mixing ratios (0-100%). The optimal configuration was determined to be the Recirculation Cycle with Sensible Heat Recovery, functioning at 50% fresh air ratio, a regeneration temperature of 50 °C, a supply flow rate of 3 ACH, and a high-performance desiccant wheel with (εF1 = 0.05, εF2 = 0.95). The configuration achieved a suitable mix of comfort and energy economy. It achieved comfort Zone A coverage of 70.0%, Zone B of 78.14%, and Zone C of 86.52% while utilizing just 1.39 MWh/year. While Case 3.0 at 70 °C provided somewhat more comfort coverage, Case 4.0 was chosen for its energy efficiency and Zone D compliance. . The SADCS demonstrates substantial decreases in electricity consumption and lifecycle CO₂ emissions when compared to a traditional 520 kW rooftop HVAC system required to cater for the traditional HVAC needs of the factory. The findings indicate the system\u27s feasibility for sustainable industrial cooling and its compatibility with Dubai\u27s Clean Energy Strategy 2050 and objectives for energy demand reduction
Memorable Whale Watching Tourism Experiences: Insights from the Azores
Get PDFThis study examines the roles of learning, experience co-creation, and experiential satisfaction as antecedents of a memorable whale-watching tourism experience, as well as the relationships between experiential satisfaction, memorable whale-watching tourism experience, hedonic well-being, place attachment, and pro-environmental behavior. Using data collected from 247 tourists who had taken a whale-watching trip in the Azores, partial least squares structural equation modeling was then applied to the dataset. The results indicate that the higher the levels of learning, experience co-creation, and experiential satisfaction, the more memorable is the whale-watching tourism experience. Experiential satisfaction and memorable whale-watching tourism experience are, in turn, significant determinants of hedonic well-being, place attachment, and pro-environmental behavior. The paper therefore calls for greater efforts by whale-watching trip providers to enhance their learning and interpretation programs and to ensure that the overall trip experience is as interactive as possible
Reusing Underutilized Warehouses: Impact of IoT Integration on Water Conservation in Hydroponic Farming within Mina Zayed Warehouses in Abu Dhabi
Get PDFUrban farming is seen as an essential approach to sustainable development, particularly in arid regions like Abu Dhabi, United Arab Emirates, where limited arable land and water resources pose significant challenges. With an emphasis on reusing underutilized urban spaces like the industrial warehouses in Mina Zayed, this thesis explores the transformative potential of shifting from conventional agricultural methods to advanced smart farming approaches. The main objective of this project is to optimize the effective use of urban areas and improve water conservation in agricultural operations by integrating Internet of Things (IoT) technology into hydroponic systems within these warehouses. The research proposes a solution that can be adapted to different scales and aims to develop an urban agricultural model that addresses key issues related to food and water security in Abu Dhabi’s urban environment. This is achieved through deployment of advanced sensors that monitor environmental factors like temperature, humidity, water flow, and light, enabling real- time adjustments within simulated indoor hydroponic setups. This strategy is expected to impact different sectors by enhancing water usage, boosting agricultural productivity, and improving resource management. Data for this study has been gathered and examined from a selection of current literature to support the validation of the suggested model. The study uses a mixed methods strategy to collect and analyze data, combining quantitative and qualitative techniques. Also, it demonstrates the effectiveness of IoT technology for real-time monitoring in hydroponic farming by using a simulation model that replicates the real environment. The outcomes from this simulation will provide valuable insights and strategic recommendations for the agriculture sector, urban planners, and policymakers, highlighting the potential of IoT technology to transform urban agriculture
Utilizing Natural Language Processing to Optimize Business Processes
Get PDFThis Capstone project explores the potential of Natural Language Processing (NLP) techniques in optimizing decision-making and organizational workflows across various domains. The Capstone project uses three case studies to demonstrate how sentiment analysis, frequency analysis, and topic modeling can analyze unstructured textual data to provide insightful findings. The first case study evaluates employee satisfaction using sentiment analysis, uncovering trends across departments and roles to guide targeted organizational interventions. The second case study focuses on student feedback at a higher education institution, using sentiment and frequency analyses to identify key areas for improvement in academic programs and services. The third case study leverages advanced topic modeling techniques to analyze thematic trends in artificial intelligence (AI) research over a decade, providing strategic insights into emerging innovations and priorities. The findings highlight the efficiency and scalability of NLP techniques, with automated processes completing tasks in seconds or hours that would otherwise take weeks or months manually. The research emphasizes the importance of selecting models and techniques, including embedding models, clustering methods, and preprocessing approaches that are specifically tailored to the task and organizational needs to ensure detailed, relevant, and easily interpretable outcomes. While limitations remain, including the need for end-to-end pipelines to make these techniques accessible to non-technical users, this Capstone demonstrates the transformative role of NLP in enabling organizations to harness the power of unstructured data for strategic planning, resource allocation, and innovation
Investigating the Performance Variation for Graphene-Based Field-Effect Transistor
Get PDFAs silicon CMOS technology approaches its scaling limits, two-dimensional (2D) materials such as graphene offer promising alternatives due to its atomically thin material property and high carrier mobility. Graphene field-effect transistors (GFETs) are especially suited for high-frequency and RF applications; however, large-scale integration is hindered by substantial device-to-device variability. Primary contributors include inconsistent graphene transfer, contact resistance, poor dielectric interfaces, and underlying substrate topography. This work presents a comparative study of three GFET architectures to address variability: (a) raised Al gate with 15 nm Al2O3, (b) raised Al gate with monolayer hexagonal boron nitride (hBN), and (c) recessed Al gate with monolayer hBN. By replacing Al2O3 with atomically flat hBN (~0.45 nm) and introducing a chemical-mechanical planarized (CMP) recessed gate, we significantly improved both performance and uniformity across 750+ fabricated devices. Results show the average Dirac voltage shifts from ~7.8 V (Al2O3/raised Al gate) to 1.2 V (hBN/raised Al gate) and further to 0.7 V (hBN/recessed Al gate). Contact resistance drops from 2.59 kΩ∙µm to 0.67 kΩ∙µm, and hole mobility variability reduces from 75% to 18%. Raman spectroscopy and SEM analysis confirm smoother surfaces and improved graphene integrity in the recessed gate structure. Mobility performance remained comparable across various structures, but the recessed gate configuration exhibited tighter distribution and greater reproducibility. Overall yield increased from 14.8% for the Al2O3/raised gate devices to 65.1% for the hBN/recessed gate structured devices. Heatmaps of yield, mobility, and Dirac voltage distribution illustrate that the planarized gate and improved dielectric interface led to spatially uniform and statistically consistent device behavior. Presented work highlights the importance of gate dielectric selection and geometry in minimizing GFET variability. The integration of monolayer hBN with a recessed gate provides a reproducible and scalable fabrication platform, reducing the impact of surface roughness and interface defects. This work evaluates complete device distributions, revealing statistically significant improvements in performance and reliability. Our results establish a practical pathway for wafer-scale GFET integration using hBN dielectrics and planarized gate structures for advancing two-dimensional based transistors toward high-yield, high performance electronics