Events
Spring 2025 Computing Showcase (C-Day) Winners
Undergraduate Capstone Winners
First Place
Abstract: Draw The Night Sky is a game project made in collaboration with Carter鈥檚 Lake to make their constellation viewing program more accessible. The stars in the sky are quite difficult to see without the perfect conditions, so an alternative would assist with this greatly. By creating a fun and interactive experience through a game, it should teach the visitors of the nature center to be able to search for stars even outside of the game. Utilizing an accurate star map based on the Yale Bright Star catalogue, we have an accurate star map that mirrors the real world which adds to the immersion of the players. In addition to this, very simple real-world tools are provided to the player to find the constellations. Exploring space is quite lonely, so a fun companion in the form of Stella is there to give the player all the right tools to get the job done.
Department: Software Engineering or Game Design and Development
Supervisor: Dr. Lei Zhang
|
Second Place
UC-111 Accessible Interactive Map by , Ingram, Megan, , ,
Abstract: Finding that walking campus gets you out of breath? We did too! Using React and Flask,
we are building a web application that directs 食色视频 students to the path with the lowest
elevation and shows the shifts in between. It also displays accessible doors. The
purpose of this app is to develop a more inclusive application so people with asthma,
cardiovascular issues, and wheelchairs at 食色视频 can safely traverse campus.
Department: Computer Science
Supervisor: Prof. Sharon Perry
| |
Third Place
UC-092 Cookly.io - Advanced Recipe Generator by , ,
Abstract: Cookly.io was a passion project started during the AI Club Hackathon where it was
awarded 3rd place. Cookly is an AI powered recipe assistant that helps users use available
ingredients into delicious meals. Users can input ingredients manually or upload a
photo of the pantry or fridge where Cookly will use computer vision to identify the
ingredients and SBERT to match the ingredients with the perfect recipe.
Department: Computer Science
Supervisor: Dr. Femi Ojo & Dr. Chen Zhao
|
Graduate Capstone Winners
First Place
GC-008 MediVault: An AI-Powered Secure Medical Image Sharing Platform by , ,
Abstract: MediVault is a secure, cloud-native platform that empowers patients and healthcare
providers to upload, view, and share medical images like X-rays and MRIs with confidence.
Built using Next.js and Node.js, and deployed on AWS Free Tier (S3, RDS, KMS), the
system implements role-based access, two-factor authentication, and end-to-end encryption
to ensure privacy and HIPAA compliance. MediVault features an intuitive interface
and integrates AI-enhanced automation to streamline metadata tagging and detect potential
anomalies in scans. Designed for scalability, usability, and compliance, this project
showcases real-world expertise in full-stack cloud development and healthcare cybersecurity.
Department: Information Technology
Supervisor: Dr. Ying Xie
Project Sponsors: Gennadiy Kemelmakher, Arpna Aggrawal, Richard Windlandy
| |
Second Place
Abstract: Cancer pathology reports are important for diagnosis and treatment planning, yet their complex language poses a significant challenge for patients and nurses to understand. This communication barrier often results in confusion, anxiety, delayed decisions, and reduced care quality. To address this, OncoClarify, an AI-powered tool, has been developed to simplify cancer pathology reports and provide role-specific explanations tailored to doctors, nurses, and patients. By leveraging the Gemini API, the system extracts critical details from pathology reports and generates customized summaries. Doctors receive technical insights highlighting biomarkers, mutations, and prognostic indicators. Nurses are provided with actionable guidance for managing side effects and care protocols, supporting effective treatment implementation. Patients benefit from plain-language summaries, enabling a better understanding of diagnoses, treatment options, and potential side effects. This clarity fosters informed decision-making and enhances communication among all parties involved. The tool integrates Tally Forms for report submission and role identification, PDF.co for text extraction from various file formats, and Make.com for seamless automation. The user interface, designed with Gamma.app, ensures ease of access and usability for non-technical users. Output is delivered via email or downloadable PDF. Validation was performed using real-world cancer pathology reports from The Cancer Genome Atlas (TCGA). Accuracy, usability, and processing speed were measured, and feedback from healthcare professionals confirmed its effectiveness. The tool supports scanned images and complies with HIPAA and GDPR standards. Future plans include support for multiple languages, tumor visualization using medical imaging AI, EHR integration, and application to other medical fields such as radiology and cardiology.
Department: Computer Science
Supervisor: Dr. Sanghoon Lee
|
Third Place
GC-039 ClinicPix: Secure Medical Image Sharing Web Application by , , ,
Abstract: ClinicPix is a cloud-based system designed to streamline the management of medical
images such as X-rays and MRIs. It offers healthcare providers and patients a secure,
intuitive interface to upload, view, and share medical images across institutions
and devices. The platform ensures full compliance with HIPAA through robust security
measures, including role-based access control, end-to-end encryption, and comprehensive
audit trails. Its scalable architecture supports growing data needs while maintaining
high performance and reliability. By enhancing accessibility and safeguarding sensitive
health information, the platform aims to improve clinical workflows, patient engagement,
and collaborative care.
Department: Information Technology
Supervisor: Dr. Ying Xie
|
Undergraduate Research
First Place
Abstract: This research focused on the implementation of modern computing systems by designing and simulating a 16-bit RISC-based ISA computer. The computer is built on a Von Neumann memory architecture with 1024脳16-bit word-addressable space and a 6-bit ISA with 36 implemented instructions. The central processing unit (CPU) includes a control unit (CU) that automatically drives the fetch-decode-execute (FDE) cycle, four addressable general-purpose registers (GPRs), and an Arithmetic Logic Unit (ALU) comprising 21 operations and producing four flags. We validated the system by executing Euclid's GCD algorithm, generating the binaries with a custom assembler written in Python.
Department: Computer Science
Supervisor: Prof. Waqas Majeed
|
Second Place
Abstract: Mental health is an essential part of living a balanced and fulfilling life, but it is often overlooked compared to physical health. While physical health is important for performing daily activities, mental health plays a crucial role in how we manage stress, build connections, and make decisions. Previous research studies have shown that nearly 60 million Americans experienced a mental illness in 2024, yet there were only 340 people for every one mental health provider in the U.S. Furthermore, young adults aged 18鈥�25鈥攚ho are the most digitally connected generation鈥攕uffer from the highest rates of severe mental illness yet are the least likely to seek or receive treatment. These findings highlight a growing crisis where more people are struggling with mental health issues, but the resources available to help them remain insufficient. This study presents a comprehensive investigation into predictive models and datasets for early mental health intervention, combining a systematic literature review with empirical research. We examine a range of existing machine learning algorithms and datasets that focus on behavioral and physiological indicators, including heart rate variability, sleep patterns, device usage, and social interaction metrics. Through critical evaluation of these models, we identify key features and data types most effective for predicting early signs of mental health conditions. Based on these insights, we design a predictive system architecture, including form-matching tables that align symptom inputs with appropriate risk levels and recommended actions. To translate the system into an accessible user experience, we develop mobile application wireframes and conduct usability research on features that support early detection and intervention. This work aims to bridge the gap between technical innovation and user-centered design, offering a holistic and proactive approach to empowering mental wellness through early intervention.
Department: Computer Science
Supervisor: Dr. Maria Valero
| |
Third Place
Abstract: Whole Slide Images are used to capture details of patient cells. Hospitals and clinics have different processes and methods to create WSIs resulting in WSIs not being standardized. Different file formats are used and different colors are used to represent different features. The normalization process helps set up the WSI into a format that the current model can easily process.
Department: Computer Science
Supervisor: Dr. Sanghoon Lee
|
Master's Research
First Place
Abstract: Alzheimer鈥檚 disease is a growing public health issue due to its progressive nature and increasing prevalence. Large language models (LLMs) offer promising avenues for non-invasive cognitive assessment through natural language understanding. In this study, we evaluate DeepSeek鈥檚 general-purpose model V3 and reasoning-enhanced R1 variant鈥攆or identifying Alzheimer鈥檚 dementia (AD) and Cognitively Normal (CN) individuals using transcripts derived from spontaneous speech. Two baseline prompting strategies (zero-shot, chain-of-thought ) were applied to both model types and an additional query (self-consistency prompting) was applied to assess better predictions. Accuracy was the primary performance metric. When positively identifying AD, the general-purpose DeepSeek V3 model produced the highest true positives at 88%, but tended to misclassify CN as AD. In contrast, the DeepSeek-R1 model achieved the highest true negatives at 90% for CN classification. Overall, DeepSeek models surpass chance-level classification, but further refinement is needed before clinical applicability can be ensured.
Department: Software Engineering or Game Design and Development
Supervisor: Dr. Seyedamin Pouriyeh
|
Second Place
Abstract: Large Language Models (LLMs) are widely used in Alzheimer's disease research to classify speech patterns. However, there is no standardized framework to ensure the reliability of prompts used in these classifications. This study investigates the sensitivity of Alzheimer鈥檚 disease classification prompts to small variations and finds that these prompts are indeed sensitive, leading to inconsistencies in model performance. To address this, we implement an automatic prompt optimization framework to refine the base prompt. Experimental results demonstrate that the optimized prompt improves classification accuracy by 12.83% compared to the baseline, underscoring the significance of systematic prompt engineering in enhancing the reliability of LLM-based Alzheimer鈥檚 disease detection. Although the optimized prompt remained sensitive to variations, it consistently showed improved overall accuracy.
Department: Data Science and Analytics
Supervisor: Dr. Seyedamin Pouriyeh
|
Third Place
Abstract: This project investigates the presence of artificial intelligence (AI) competencies across Georgia鈥檚 higher education curriculum using university course catalogs as the primary data source, supplemented by OpenALG materials. We applied large language models, including OpenAI鈥檚 ChatGPT and embedding APIs, to analyze over 34,000 courses summarizing content, classifying AI relevance, and mapping to global frameworks (AI4K12 and UNESCO). Techniques such as topic clustering, semantic similarity analysis, and geographic distribution mapping were used to uncover patterns in AI integration. Findings reveal that AI content is concentrated in computing disciplines and research universities, with limited coverage in community colleges, MSIs, and non-technical fields. The results highlight the need for more equitable and interdisciplinary AI education across Georgia鈥檚 institutions.
Department: Information Technology
Supervisor: Dr. Ying Xie
|
PhD Research
First Place
Abstract: This study presents the design of an AI-powered caregiver support app aimed at personalized mental health and burden management for individuals caring for Alzheimer鈥檚 patients. The design is grounded in insights drawn from a comprehensive analysis of 28 recent studies on AI-driven mental health interventions. These findings informed the implementation of key features, including machine learning models such as Random Forest, clustering, and supervised learning to create adaptive care plans tailored to patient and caregiver profiles. The system dynamically adjusts task schedules based on engagement data and provides interpretable recommendations through SHAP. With built-in emotional check-ins, mood tracking, and caregiver-centric interventions, the app promotes well-being while reducing the cognitive and emotional load often experienced in dementia caregiving.
Department: Computer Science
Supervisor: Dr. Nazmus Sakib
|
Second Place
Abstract: This study predicts postprandial glucose peaks and spike durations using 10-day multimodal data from 10 participants. Glucose, meals, workouts, and insulin doses were logged via the Nutrilyzer web app. Macronutrient content carbs, fats, and proteins was extracted using GPT-Vision, a highly accurate food analysis tool. These tuples were normalized to baseline glucose and aligned with a 3-hour window post-meal. Three models were tested: LSTM, Time Series Transformer, and ARIMA. LSTM performed best with 83.78% accuracy, followed by Transformer (71.43%) and ARIMA (62.41%). Results show the promise of AI-based food logging and time series modeling for personalized glucose forecasting.
Department: Computer Science
Supervisor: Dr. Maria Valero
| |
Third Place
Abstract: Electric vehicles (EVs) are widely considered a cleaner alternative to internal combustion engine vehicles. But their growing use creates indirect emissions via two main channels: more traffic congestion from more vehicle activity and more demand on power plants providing electricity for EV charging, usually depending on fossil fuels. This work offers a comprehensive, data-driven framework to forecast greenhouse gas (GHG) emissions connected to road traffic as well as EV-related power generation. Based on vehicle and speed characteristics, we forecast vehicle-level energy consumption and emission rates using a multi-model architecture that includes a Feed Forward Neural Network (FNN). While the Meta Prophet time series model is used to project power plant emissions under different energy demand conditions, macroscopic traffic flow models are used to estimate tract-level speed-density relationships. An Integrated Emission Model combines these elements to allow assessments particular to each area. Capturing vehicle mix, traffic dynamics, and energy grid variations, our study covers four major U.S. states鈥擟alifornia, Georgia, New York, and Washington. Results show notable spatial and temporal variations in emissions, therefore stressing the need of thorough models taking into account the intricate interactions between energy infrastructure, traffic patterns, and EV adoption.
Department: Computer Science
Supervisor: Dr. Mahyar Amirgholy
|
Contact Info
食色视频 Campus
1000 Chastain Road
食色视频, GA 30144
Marietta Campus
1100 South Marietta Pkwy
Marietta, GA 30060
Resources For
Related Links