Omar Zamzam

About Me

I am currently a Ph.D. student in the Biomedical Imaging Group at University of Southern California (USC), advised by Prof. Richard Leahy.

My research primarily focuses on applying machine learning techniques—particularly generative models, computer vision, and semi-supervised learning—to medical imaging. I prioritize working on problems that require creative and critical thinking, and I value being part of a dynamic, collaborative team where the environment fosters both innovation and productivity.

My Research

My research focuses on applying machine and deep learning techniques, particularly generative models, computer vision, and semi-supervised learning, to medical imaging. I have worked on projects ranging from positive-unlabeled (PU) learning for image data as well as time series EEG signals to developing diffusion models for 3D brain image synthesis and inpainting. Additionally, I have applied quantile regression for anomaly detection and image translation, as well as explored image registration techniques, and image captioning using vision transformers. Currently, my work is centered on creating a multimodal foundational model to handle diverse medical data modalities for unified analysis.

Most Recent Work Experience

• Jan 2021 – Present: Research Assistant
  • University of Southern California
  • Duties include: Research on the intersection of machine learning and medical imaging
  • Advisor: Professor Richard Leahy
• May 2024 – August 2024: Machine Learning Intern
  • Insitro
  • Duties include: Diffusion Models for Image-to-Image Translation
• Jan 2021 – Jan 2024: Teaching Assistant
  • University of Southern California
  • Course: EE 503 Probability for Electrical and Computer Engineers

Education

Publications

Powered by Jekyll and Minimal Light theme.