Mason Journals (George Mason Univ.)
Not a member yet
3256 research outputs found
Sort by
Analysis on Extracellular Vesicles Using the 96.96 IFC with SNP Type Assays as Quality Control for Biobanks
No full textSingle Nucleotide Polymorphisms (SNP) are variations in the human DNA that change one letter in the codon. These variations change person to person, therefore, studying these changes can aid in personalization of medications and ensure that researchers have obtained the correct samples from biobanks. This study analyzes the possibility of utilizing SNP genotyping on extracellular vesicles (EVs). EVs are released by cells into the interstitial fluid and transport a variety of cellular components, including proteins, mRNAs, miRNAs, DNA, and lipids. Due to their essential role in intercellular communication, EVs hold promise as diagnostic biomarkers, therapeutic agents, and drug delivery systems. In cancer diagnostics, EVs could assist in early disease identification and biomarker development as their content closely resembles that of their parent cells. In this study, SNP investigations were enhanced through the application of microfluidics PCR and SNP assay methodologies, utilizing a 96.96 dynamic array IFC and Biomark HD PCR device. Our research encompassed multiple experiments, adjusting various parameters to observe differences in the confidence and call rates ofSNPs present in each sample. The results vary based on some loading errors during the process of pipetting, sometimes preventing SNPs from being called, as well as limited reagents for certain SNPs. Overall, SNP genotyping on extracellular vesicles is a possible solution in ensuring researchers have obtained the right samples and aiding in personalization of medications
Utilizing hand tracking technologies in serious games to teach American Sign Language
No full textSerious games gamify, or apply elements of games to, processes that are typically considered wearisome, such as education or training, to boost the user’s interest and participation. Recent advancements in virtual reality have allowed for widespread accessibility of hand tracking technologies, rapidly opening new avenues in both serious and non-serious games for the general public. ASL is the most common form of sign language in the U.S., used by approximately half a million people there1, but can be difficult to learn through online resources such as mobile apps that are unable to track the user’s hands. With the number of people relying on ASL only growing, an easy way to gain some level of comprehension of the language should be made available. This project utilizes the Unity game engine and OpenXR packages to track the user’s hands when using compatible technologies such as the Meta Quest 2, 3, or Pro. The VR game prototype recognizes when the user makes certain ASL signs, and the surrounding environment then reacts accordingly. Using this mechanic, the user can solve puzzles by figuring out what different signs mean and what effect they have. Though no studies have yet been performed to test the application’s effectiveness, as the player imitates nearby signing characters to learn new vocabulary, the end goal is for the step of associating motions with their resulting in-game actions to strengthen the player’s memory and help them effectively internalize the meanings of ASL signs
Low-Cost Geolocation for Laser Target Designators in Unmanned Aerial Systems
No full textMilitary Unmanned Aerial Systems (UAS) utilize high-cost Laser Target Designators (LTD) for intelligence andtargeting operations. This research attempts to fill the LTD access gap by producing a low-cost and opensourcedLTD solution for under $500. Using a multi-rotor drone, a GPS, and a LiDAR range finder, thisresearch will use flight data from these sensors to produce geolocation results (latitude and longitude) of wherethe LiDAR sensor is impacting the target object. The key variables derived from the flight controller are altitude,yaw, roll, and pitch. The LiDAR sensor produces distance from the UAS to the target object. Using thesevariables, an open-source python code is built to determine the location of the object classified. Virtual offsetsof the GPS, flight controller, and LiDAR sensor are incorporated into the code to allow for increasedgeolocation precision. Using Pythagorean theorem, we are able to solve for ground distance from the UAS'slocation to the target object. After ground distance is solved, yaw (direction) can be incorporated to determinethe target object’s location. The results demonstrate a 95% efficiency in solving the object’s location usingautomated geolocation compared to manual calculations. Further, automated geolocation allows for real-timetranscribing into a dataset which can further be plotted in a three dimensional plane to facilitate understandingof the locations of objects of interest. We consider this an acceptable amount of efficiency to allow real timetranscribing in a variety of scenarios, mainly for initial experimental code that can be further improved upon
Design of a Wearable System Providing Visual Biofeedback on Anterior Ground Reaction Forces to Increase Lower Limb Propulsion of Stroke Patients
No full textAfter stroke, individuals often have a limited ability to generate limb propulsive forces which have been shown to be correlated with increased fall risk and lower walking speed. Biofeedback on anterior-posterior ground reaction force (AGRF) is a promising approach for improving limb propulsion. However, collecting data on ground reaction forces typically requires the use of expensive instrumented treadmills which have limited clinical applicability. Instead, Inertial Measurement Units (IMUs) have been utilized to estimate ground reaction forces of gait, presenting a portable, and low-cost alternative. But to date this approach has primarily been used for prognosis of gait impairments and has yet to be employed in biofeedback for gait retraining. In this study, we present the initial design of a wearable system for AGRF biofeedback gait retraining using seven IMUs strapped to the foot, shank, thigh, and pelvis. The IMU data was processed by a MATLAB application in real time to output the AGRF values of the paretic leg which were transmitted via Wi-Fi to a single board computer fastened to the waist. AGRF values were graphed in Python using the Matplotlib library in reference to the target AGRF value. Wearable display glasses were connected to the single board computer via cable to project the graph into the lens in real time. This system represents an alternative to bulky, and expensive equipment. Further similar systems could be developed that incorporate auditory or haptic feedback on AGRFs. Further, future work will look to compare the efficacy of this approach in real world settings in addition to the efficacy of other feedback modalities on gait retraining
A Mixed-integer programming approach for identifying optimal levels of hydrogen fuel as a renewable energy source
No full textGreen hydrogen presents an opportunity to move towards a more sustainable and low-carbon energy future and haspotential to address the current challenges faced by the world in terms of climate change and adverse health effects ofusing fossil fuels. While new technologies are coming up every day for more efficient solutions to address the challengesposed by climate change, this research aims to look at the current cost of liquid and gas hydrogen, considering sugarcaneas the source. In this work, we model a real-world problem using a mixed-integer optimization approach and solve itusing Excel's built-in optimization solver, taking into account cost functions that need to be minimized and constraints asrestrictions related to the production, distribution, and storage of liquid and gas hydrogen fuel, considering sugarcane asa source. To identify the best course of action, data on demand, manufacturing costs, and storage costs were examinedas potential factors as well. The research establishes a relationship between the use of hydrogen and gasoline based ontheir current costs while limiting overall pollution. Our results provide important insights into making decisions on usinghydrogen fuel as a potential option for renewable energy
Dynamic Time-Expanded Topology Design for LEO Satellite Network
No full textThe dynamic nature of Low Earth Orbit (LEO) satellite constellations, exemplified by systems like Starlink, Amazon Kuiper, and OneWeb, presents significant challenges in network topology design, impacting parameters such as latency and capacity. This research focuses on exploring Deep Reinforcement Learning (DRL) algorithms for the design of dynamic topologies in LEO satellite networks. Additionally, it investigates the use of the SpaceNet platform and Mininet for conducting LEO satellite network experiments. The proposed approach utilizes DRL algorithms to dynamically adjust satellite links, optimizing for reduced latency and minimized link churn while maximizing network capacity. These algorithms are rigorously tested through real-time simulations on the SpaceNet platform and Mininet, providing a comprehensive evaluation of their performance compared to static and traditional methods. Preliminary results indicate that DRL-based topology design algorithms offer substantial improvements in network performance, including enhanced capacity and lower latency. The study also underscores the effectiveness of the SpaceNet platform and Mininet as valuable tools for simulating and experimenting with LEO satellite networks. This research contributes to the field by providing a detailed investigation into the application of DRL for dynamic topology design and demonstrating the utility of advanced simulation platforms for experimental validation. These insights are crucial for the development of efficient and scalable network topologies in next-generation LEO satellite constellations
Generating a Robust Dataset for Radar Detection in the CBRS Band
No full textOftentimes wireless spectrums can become saturated when the demand for the spectrum is too high leading to lower quality communication. We choose to address this issue by employing spectrum sharing, which involves letting various users share the same frequency band in order to deliver certain data as bits through this spectrum, and leads to more optimized communication. There is interest in allowing cellular networks to communicate through federal bands that have so far been reserved for the government. However, one prevalent issue is that these communications need to be paused whenever the federal bands are being used to send a radar signal by the government in order to not interfere.
In order to recognize when the radar is being sent we propose developing a machine learning software to recognize radar signals within the spectrograms that are generated from the network and stop the data from being sent in the interfering frequencies without cutting off the connection that the users have. We deploy this on Open Radio Access Network (O-RAN), which is a virtualized, disaggregated solution for 5G/6G cellular networks, and is extremely beneficial in that any other cellular operator that uses O-RAN can use the same solution. For example, when there are different signal-to-noise ratios and when there isn’t, as well as when the radar is shifted overtime in the spectrogram. In fact, creating spectrograms of radar signals at different delays can be achieved through shifting existing data within a smaller dataset. After that, the graph is then overlaid with a LTE signal with varying noise levels in order to make sure that the radar can be recognized in different environments. Once we introduce multiple spectrograms with different delays as well as different noise levels, we will be making the dataset more robust leading to the machine learning model to detect radar signals more accurately
Identifying Key Contributors to Regional Ozone Levels: A Machine Learning Approach Integrating Environmental and Socioeconomic Factors
No full textOzone levels exhibit significant regional variability, influenced by a combination of socioeconomicfactors, emissions, and meteorological conditions. This study aims to identify the primary industrial andsocioeconomic contributors to ozone pollution, as well as the impact of other emissions such as NO2 andCO. Utilizing explainable machine learning techniques, specifically Random Forest, XGBoost and SHAP(SHapley Additive exPlanations) values, we quantify and identify the major contributors affecting ozonelevels in different regions. By analyzing industry data and various economic sectors—including mining,agriculture, scientific services, arts, and real estate—alongside other relevant emissions, we uncoverpotential reasons and sources for ozone levels unique to each area. These insights provide actionableinformation for policymakers and environmental agencies, enabling targeted interventions to manage andreduce ozone pollution. These insights offer actionable information for policymakers and environmentalagencies, enabling targeted interventions to manage and reduce ozone pollution
Assessing the Accuracy of a Machine Learning Model for Wildfire Spread Prediction in Regional Air Quality Forecasting
No full textDue to its significant fire emissions, wildfire spread, characterized by its spatiotemporal variations, is crucial for regional-scale air quality forecasting. The process involves various physical and dynamic phenomena, making it complex and computationally intensive to simulate these interactions accurately at regional scales. Recent advancements in machine learning have shown promise in predicting fire spread, effectively capturing the spatial distribution of fires while being computationally efficient once trained.
In this study, we developed a machine-learning wildfire spread model at NOAA-ARL to enhance next-generation regional air quality forecasting (AQF) applications. The model seeks to increase the accuracy and efficiency of wildfire-related air quality predictions. Preliminary results indicate the validation of the model's predictions compared to data obtained in the March 2024 Page County, Virginia wildfires. These outcomes suggest that the machine learning model is a powerful tool for forecasting wildfire spread and its impact on regional air quality. Ultimately, this tool can be used as a short-term prediction for the growth of future wildfires
Implementation of Learning to Enhance Cost Reduction in Object Search
No full textMany object search algorithms in unknown household environments search greedily, since many algorithms, unaware of where task-critical objects may be located, naively search nearby containers until the target object(s) can be found, thus making them problematically inefficient. Efficient object search in unknown environments requires predicting what lies beyond the known space. In our work, we demonstrate that machine learning is an effective tool for making such predictions about where unseen objects are likely to be found in household environments. In our research, we trained and tested neural networks that we designed to complete complex tasks in household environments while minimizing the overall cost of running a certain algorithm. We train a graph neural network in similar household environments and then use the resulting model to predict the probability of the object’s existence at unseen locations. By using our learned models, the algorithm can anticipate where target object(s) are likely to be found and allow the robot to make more informed decisions, even in unfamiliar situations, enabling the system to adapt its decisions based on training done in similar environments. We finally report the learned model that results in the best object-search performance