Hello! I am Christian Cunningham (CV), a Ph.D. candidate in Physics at Oregon State University. I graduated from Barrett Honors College at Arizona State University with two B.S. in Physics and Mathematics (Summa Cum Laude and Moeur Award Recipient). I completed my M.S. in Physics at Oregon State University in 2025. All with perfect 4.0 GPA, if you care about that sort of thing.
My dissertation is on analyzing cellular shapes from E-cadherin (a cellular adhesion biomarker) stained tissue microarrays (TMA). The motivating question I am pursuing is: what information do cellular shapes carry to understand the progression of breast cancer? To this end, I have trained both image instance-segmentation models (powered by Cellpose-SAM) and classification models. I am classifying the TMAs to the presence of triple negative cancers, the grade, and the stage. For classification, I am using both image classifiers (CNN-backed and ViT-backed) and pointcloud classifiers. With the segmentations, we obtain morphological features for the pointcloud model for both cell membrane and nuclei. Additionally, we compute Potts-like energetics for each cell for alternative pointcloud representations. Finally, we employ a Bayesian Mixture of Experts to synthesize final classifications across these different representations. Currently, I am writing up a paper summarizing the results of these efforts.
I have written this site to be viewed in a variety of ways: with javascript enabled, only CSS, no-CSS support. I also have themes for both light and dark theme preferences. This has entirely been hand coded, without use of any frameworks, just simple ol HTML + (optional) CSS + (optional) JavaScript.
When I am not working, I like tinkering with electronics - both hardware and software. Whenever I leave my proverbial cave, I like to skateboard and bike!
Why CCRL? Christian Cunningham Research Labs.
In the interest in providing my notes from classes and other engagements, My Wiki.
AI holds a special place in my heart. The first AI project that engaged me was Mar/IO, presented by SethBling. It used reinforcement learning to play a Mario Game through the NEAT algorithm. I then learned about the AI detection for XRays that could detect imperceptible differences in the shades of grey for better screenings and was intrigued by the technology. Largely, when I think of AI, I don't think of LLMs at all.
While I enjoy the topic of AI, I find writing code even more fun. As such, my work is largely unassisted by AI. I have tried out some of the tools and have used them in some places, but I find that the work that stands the test of time (and me returning to it later) is the work I write myself. One project that was basically all Antigravity, well following the ruts that I made, was the Image Segmentation Editor (link forthcoming).
At-a-glance: I currently have 3 desktops, 2 laptops, and a Raspberry Pi. I interface with two HPC clusters: one using SGE and one Slurm. I also manage a VPS (the one you are accessing this from). Across the devices, I am running:
My GPG Public Key can be obtained: here.