Stephanie Murray

Hi, I’m Stephanie Murray, a second-year Master’s student in Computer Science at the University of Washington Bothell. My research focuses on machine learning, data analysis, and quantum computing, with a particular interest in applying these techniques to healthcare and neuroscience.

My thesis explores the use of quantum machine learning for classifying EEG signals, aiming to improve accuracy and efficiency in processing complex datasets. I’m also working on an independent project that combines quantum feature generation with deep learning for medical imaging.

I enjoy working on meaningful problems, learning from new ideas, and exploring creative solutions in computer science that can have a positive impact.

• Quantum Computing: Exploring hybrid quantum-classical algorithms for applications in machine learning and medical imaging.
• Machine Learning: Developing models for classification, prediction, and feature extraction in complex datasets.
• High-Performance Computing (HPC): Using parallel computing techniques and GPU acceleration (CUDA) to improve computational efficiency for large-scale problems.
• Data Analysis: Investigating methods to handle small or imbalanced datasets, including synthetic data generation and feature engineering.
• Neuroscience Applications: Applying computational techniques to analyze complex signals like EEG data.

Masters Thesis: "Exploring Quantum Machine Learning Enhanced Models for EEG Data Classification"

My thesis investigates the application of Variational Quantum Circuits (VQCs) for classifying EEG signals. This work integrates quantum computing methods to enhance signal analysis, focusing on feature extraction and classification performance for neuroscience applications.

Methods Used: Variational Quantum Circuits, classical preprocessing (e.g., Fourier transforms), and gradient-based optimization algorithms for training quantum models.

Independent Project: "Quantum-Enhanced Convolutional Neural Networks for Liver Tumor Classification: An Exploratory Hybrid Approach"

This project investigates a hybrid approach combining quantum feature generation with Convolutional Neural Networks (CNNs) for liver tumor detection using CT scans. The focus is on improving classification performance in small, imbalanced datasets.

Methods Used: Quantum kernel feature extraction, CNN-based classification, synthetic data augmentation using SMOTE to address class imbalance, and stratified splits for training and testing while maintaining class distribution.

BlurEffect: GPU-Accelerated Gaussian Blur Optimization

Designed a CUDA-based Gaussian blur algorithm that achieves significant performance gains for image processing. Integrated the implementation into a custom-built Qt-based GUI for real-time visualization and batch processing.

Methods Used: CUDA parallel programming, GPU kernel optimization, and real-time rendering with Qt framework.

SightShift: VR Assistive Technology

Worked in a team in conjunction with MXTReality to create a VR application simulating LHON (Leber Hereditary Optic Neuropathy) disorder using Unity and Oculus SDK. Focused on accessibility and assistive technology, promoting empathy and understanding of visual impairments through immersive experiences.

Methods Used: Unity game engine, VR development (Oculus SDK), and accessibility design principles.

I attended and volunteered at Supercomputing 2024 (SC24), a premier international conference on high-performance computing, quantum computing, networking, and storage.

Email: contact at stephaniemurray dot org

LinkedIn: linkedin.com/in/s-a-murray