Generative Modeling of Synthetic Echocardiography Videos using Diffusion Processes
Ahmed Alaa, Professor
Electrical Engineering and Computer Science
Closed. This professor is continuing with Fall 2024 apprentices on this project; no new apprentices needed for Spring 2025.
Institutional constraints and concerns over patient privacy impedes sharing of clinical data among researchers. High-fidelity synthetic data that mimics the real data distribution without revealing real data for individual patients can help empower and democratize research applying machine learning models to clinical problems. In this project, we will use diffusion-based generative models to synthesize echocardiographic studies (cardiac ultrasound clips) using 4 publicly available data sets.
Qualifications: Successful applicants will work with publicly-accessible data sets to train diffusion models to generate synthetic echocardiograms as described above. The student should have knowledge of Python and strong background/interest in machine learning and computer vision, and will be expected to meet with the faculty mentor twice a week. The student will be trained to conduct literature review, formulate research problems and engage in scientific writing.
Hours: 12 or more hours
Related website: https://github.com/ahmedmalaa/ETAB
Digital Humanities and Data Science Engineering, Design & Technologies