Paul Calle
Postdoctoral Research Fellow at University of Oklahoma
Norman, Oklahoma, United States
Summary
👤
Senior🎓
Top SchoolPaul Calle is a machine learning researcher and postdoctoral fellow with nine years of experience applying deep learning to healthcare problems, including medical image classification, segmentation, and EHR-driven model evaluation. He holds a PhD (in progress/done per profile) in Computer Science and a Master’s in Data Science, and has built and optimized CNN and nnU-Net models on HPC systems including Summit, achieving high performance (e.g., renal vessel segmentation IoU ~0.89). His work blends rigorous uncertainty quantification (nested cross-validation, hyperparameter optimization) with practical engineering—production-style Python libraries, Git-based workflows, and Linux/supercomputer operations. He has translated research to clinical settings through internships that demonstrated time savings in radiology triage and prototyped ICH segmentation on AWS. He also experiments with LLMs for behavior-change messaging, showing a knack for combining classical ML, deep learning, and emerging generative models.
8 years of coding experience
9 years of employment as a software developer
Bachelor of Science - BS Mechanical Engineering, Bachelor of Science - BS Mechanical Engineering at National University of EngineeringDoctor of Philosophy - PhD Computer Science, Doctor of Philosophy - PhD Computer Science at University of Oklahoma
English, French, Spanish
Github Skills (14)
segmentation9
r-package6
predictive6
evaluation6
recall6
datasets5
predictive-modeling5
machine-learning5
modeling4
cran4
machine-learning-models4
evaluation-framework3
data-science3
package-repository2
Programming languages (4)
Github contributions (5)
thepanlab/Supercomputing_Instructions
May 2020 - Sep 2021
Useful information to use the supercomputers
Contributions:38 commits, 1 PR, 25 pushes in 1 year 4 months
thepanlab/Endoscopic_OCT_Epidural
Jun 2021 - Aug 2023
Contributions:24 pushes in 2 years 3 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.