About Me
If you've made it this far, welcome! Thanks for taking the time to learn a little more about my background and interests. I have three passions: 1) AI, 2) SciFi, and 3) my family. Well, I don't think they're necessarily in that order (LOL). But, you get the gist.
My key research interests are generative adversarial networks (GANs), diffusion models (DDPMs), multispectral imagery in healthcare to include thermal imagery, and multimodal deep learning for pain detection. I'm on TikTok as well where I post random musings and progress in my AI research: https://www.tiktok.com/@mybuddyskynet
I also spend many a weeknight, early 5AM morning, and weekend afternoon, maximizing the word count on my scifi trilogy which I hope to independently publish by April 2025.
After graduating from Georgia Tech with a B.S. in Applied Biology, I started my career at the Centers for Disease Control and Prevention (CDC) in Atlanta studying injury epidemiology from blast explosions, to DARPA, to national security work, leading me to Booz Allen where I currently work as an Executive Advisor in AI. Over the past decade I've led projects for multiple federal agencies ranging from the Food and Drug Administration (FDA) to the Department of Veterans Affairs to the U.S. Army. I also support an AI clinical trial with the National Institute of Health and Booz Allen Hamilton called Intelligent Sight and Sound (ISS) for cancer pain detection where we use multimodal data such as audio, video, facial landmarks, and text, to predict different levels of chronic pain. See my papers for a list of publications associated with this research.
I have a Masters in Public Health from Emory University, an MBA from George Washington University, and Ph.D in Information Systems from the University of Maryland Baltimore County (UMBC) Department of Information Systems under Dr. Sanjay Purushotham and Dr. Edward Raff.
My doctoral dissertation is entitled Multimodal Deep Generative Models for Cross Spectral Image Analysis. One of the most challenging aspects of my dissertation research was modernizing a classical computer vision task called image registration using GANs. Doing so allowed me to align images across visible and thermal spectra without the need for a reference, or ground truth between pairs. The task is complex due to the variance of rotation, shear, and translation, requiring the model to predict a deformation matrix.
Lastly, but just as important(!)... I'm on the newly Board of Advisors for the Mark Cuban Foundation AI Bootcamp. This is an incredible organization that hosts free AI bootcamps every fall for underprivileged high school students. The curriculum is rigorous and the people who pull the material together, curate it, and develop the interactive labs are brilliant.
Talks and Presentations/Awards
- Catherine Ordun shared insights into a pioneering project with the National Cancer Institute, focusing on the use of AI to detect chronic cancer pain. This endeavor, she explained, is not just beneficial but necessary for advancing healthcare, employing multimodal AI models to analyze comprehensive data from cancer patients.
- A Light Walk Through AI Visible-Thermal Research // Applied AI Meetup October 2023 on YouTube
- Unstoppable Podcast on "A More Equitable AI" - Jennie Brooks, host of the Unstoppable Together podcast chats with Booz Allen executive advisor and AI engineer, Catherine Ordun, on completing her PhD in generative artificial intelligence. Tune in as they discuss how to make access to the technology itself and education about AI more equitable, as well as how diversifying the generative AI workforce leads to more equitable outcomes
- Society of Asian Scientists and Engineers (SASE) Pro Information Technologist of the Year 2023
- Stay tuned - my IEEE International Conference on Image Processing (ICIP 2023) talk will be posted in the winter.
- 28th AAAI 2023 Doctoral Consortium - Multimodal Deep Generative Models for Remote Medical Applications
- ICML Women in Machine Learning 2022 Workshop - Thermal-Face-Contrastive GAN (TFC-GAN): A Framework for Visible-to-Thermal Face Generation
- ICML 1st Covid and Healthcare 2022 Workshop - Fourier-Based Strategies to Explore Ethnic Feature Generation during Visible-to-Thermal Facial Translation
- NeurIPS Datasets and Benchmarks 2021 Track - ISS Dataset
- NeurIPS Women in Machine Learning 2021 - favtGAN
- Honorable Mention for the 2021 PhD UMBC IS Research Award
- Presentation at AAAI FSS20 AI in Government and Public Sector
- KDD epiDAMIK Workshop 2020
- Speaking at the Institute of Medicine, National Academy of Sciences
- Giving a tutorial at University of Virginia = DataPalooza
- Presenting at JupyterCon
- Presenting at StrataAI
- Teaching data science for the National Library of Medicine
- Presenting at the first, inaugural Jean Bartik Computing Symposium at West Point Military Academy
- Leading a panel for Health AI at NVIDIA GTC 2018
- Presenting on Diversity in AI for NVIDIA GTC 2019
- Being honored as the first DCFemTech awardee at my company