Illinois Mathematics and Science Academy
Illinois Mathematics and Science Academy: DigitalCommons@IMSANot a member yet
9795 research outputs found
Sort by
3D Printing Failure Detection: Comparative Analysis of Deep Learning Architectures and Class Imbalance Techniques
No full textThis research aims to develop systems capable of real-time monitoring and alerting users of potential print failures before they result in material waste and loss of time. We present a novel approach towards early detection and classification of 3D printing failures using various machine learning techniques. Using the existing CAXTON dataset, which focused on optimizing 3D printing parameters, we synthesize a specialized dataset to detect signs of printing failures and trained deep learning models on this new dataset. We compare the effectiveness of various residual network and vision transformer architectures by evaluating both fully trained networks and regression on extracted features from frozen backbones. We also explore class imbalance techniques including oversampling and class weighting. Initial results demonstrate promising results, with a ResNet50 achieving an 89% accuracy. This research contributes to the growing field of applying intelligent systems to additive manufacturing, ensuring more efficient and reliable performance
Genetic Modification of Adenovirus Vectors for Bone Cancer Treatment
No full textThis paper aims to examine the complex interplay of bone morphogenetic protein 9 (BMP-9) with osteogenesis, osteosarcoma, and its many other functions in the arena of research and translation into clinical therapeutics. Six recombinant adenovirus lines containing different heparin-binding (HB) domains that can increase the osteogenic activity of BMP-9 were constructed and then separated into two groups. One group of the six lines had a red fluorescent protein (RFP) cassette inserted and the other group had a green fluorescent protein (GFP) cassette inserted. The fluorescence helped with precise titer determination. Multiple rounds of cell infection were conducted by adding the recombinant adenovirus lines to human embryonic kidney (HEK) 293 cells to create a high-titer stock and establish a virus bank that can be used for virus production. It was discovered that the titer with higher cell density and GFP showed more consistent infection of the cells within seven days than that with lower cell density and RFP. The titer with higher cell density consistently infected about 20% of cells in 24 hours and 30% after 36 hours with replating of the cells. This work demonstrated that the parameters needed, such as cell density, can be optimized for virus preservation and titer stability. In the future, stable genes of interest, such as HB domain and BMP-9, can be produced using the recombinant adenoviruses studied above, and then be effectively and efficiently delivered into mammalian cells. HB domain has shown to be able to increase the osteogenic activity of BMP-9, which can slow down the growth of osteosarcoma. Therefore, recombinant adenoviruses with HB domain and BMP-9 are very promising for future osteosarcoma treatments
Computational Optimization of Airfoil Aerodynamics via Machine Learning
No full textAerodynamic optimization is a critical aspect of airfoil design, traditionally relying on computational fluid dynamics (CFD) simulations to predict lift, drag, and pressure distributions. While effective, CFD methods are computationally expensive and time-intensive, limiting their practicality for rapid design iterations. This study explores the integration of machine learning, specifically convolutional neural networks (CNNs), to enhance aerodynamic predictions while significantly reducing computational costs. By training CNN models on datasets derived from CFD simulations of NACA 0024 and 0012 airfoils, this approach enables near-instantaneous predictions of aerodynamic properties, such as lift, drag, and pressure. The study examines the trade-offs between accuracy and efficiency, underscoring how machine learning can complement traditional CFD techniques. While challenges such as data quality and model generalization remain, this research demonstrates the potential of machine learning to streamline airfoil design, making aerodynamic optimization more accessible and efficient
Quantum Chess AI
No full textResearch into AI models and game theory involving extensive form games has been applied to economic models, computer science (Ikeda, K, 2023), and current reinforcement learning models such as AlphaZero. There is potential for advances in AI understanding quantum principles to have the same effect on quantum computing and technology. Quantum Chess is one of the first “quantum extensive form games” which involve the phenomena of superposition, entanglement, and phase (Cantwell, C, 2019). In this article we explore the application of reinforcement learning to the value and policy systems of Monte Carlo Tree Search. We then evaluate the basic understanding of quantum principles by having the AI complete chess puzzles requiring a quantum move solution and recording both the time taken and win rate
IMSAloquium 2025 Event Booklet
Get PDFCover designed by Aimanohi Imoukhuede, IMSA student,
Welcome to IMSAloquium, our annual, school-wide celebration of learning! This booklet contains abstracts for over 225 outstanding projects conducted by IMSA students in the 2024-25 academic year. The topics range from biomedical research to artificial intelligence, from history to the social sciences, and from business to entrepreneurshiphttps://digitalcommons.imsa.edu/archives_sir/1036/thumbnail.jp
Clientele Researching with Ivy League Potential
No full textIvy League Potential is a college admissions company that works with adolescents and their families to guide their process in achieving career goals. The focus of this intemship has been clientele research to help expand the company\u27s knowledge of all different interests to aid any client in gaining experience in certain fields. The research was centered around accessible career exploration. Over the past school year, the company introduced different Steps to create a final document of valuable resources and information in gaining experience as an interested adolescent. The goal is to create an Ivy League Potential sponsored accessible program to assist in gaining experience. While the project has only passed phase one of the program, the goal of this business project is to explore different researching techniques
Search Engine Optimization and Analytics
No full textIvy League Potential LLC is a college consulting company based out of Naperville, Illinois. Ivy League Potential specializes in helping students gain admittance to top-tier universities like Harvard, Stanford, and the University of Pennsylvania. Their track record of success has landed them features in the Chicago Tribune, Chicago Sun-Times, and U.S. News Education. One strategy Ivy League Potential uses to market their services is search engine optimization, or SEO. The goal of SEO is to rank on the first page of search engine results pages, making it more Visible to users when they search for relevant keywords. As an SEO intern, I have employed a variety of SEO tactics to increase its digital visibility and attract prospective clients. These strategies include, but are not limited to, optimizing website content with relevant keywords, fortifying domain authority, ensuring their site is mobile-friendly and fast loading, and competitor analysis. In addition to SEO, my tasks include researching for market partners. One way this was achieved was by reaching out to HOA organizations across the United States, partnering with them in return for advertising in their local communities. With these strategies and partnerships, we hope to grow the client base and solidify our reputation
Utilizing Artificial Intelligence to Help Neurodivergent Children with Abstract Operators
No full textAbstract Operators is a startup focused on developing a platform targeted at aiding companies by employing Al agents. In our project, we used Artificial Intelligence and web development with an emphasis on deployment and integration of various technologies. Neurodivergent children often struggle with issues such as overstimulation, concentration, and having a social filter. This can cause them to misbehave and accidentally harm others without meaning to. To counteract this problem, we created a website incorporating an Al chatbot that could give neurodivergent (and socially awkward people looking to improve their social skills) personalized social scenarios and help improve how they would act. Over the course of five months, the company offered one-on-one mentorship and resources to develop the Al-powered web application. This involved using multiple libraries and algorithms, optimizing an Al algorithm, implementing a React-based interface, and deploying the application on a cloud web server. Throughout the internship, the company benefited from the project through the creation of a functional application able to solve a critical problem
HBCU Experience Panel 2025
No full texthttps://digitalcommons.imsa.edu/dei_panels_gallery/1013/thumbnail.jp
HBCU Experience Panel 2025
No full texthttps://digitalcommons.imsa.edu/dei_panels_6/1003/thumbnail.jp