Lekan Adeyeri
I'm a computer science and physics undergraduate interested in computer vision, deep learning, medical imaging, 3D graphics, and tools that make technical creation more accessible.
About
Lekan Adeyeri is a computer science and physics undergraduate on leave from Pitzer College within the Claremont Colleges. His work sits between computer vision, deep learning, 3D modelling, neural surrogates for physical simulation, and tools that make technical creation more accessible and sustainable. He has contributed to Stanford-led research on deep-learning enhancement of ultra-low-dose PET/MRI, worked on a GPT-3-powered chatbot and AI tooling during a Microsoft internship, and built production-oriented software systems across research, health technology, and developer tools. He is especially interested in learned representations that make physical and geometric processes smaller, more inspectable, and more efficient to run.
Education
- Undergraduate studies in computer science and physics
- Pitzer College / the Claremont Colleges
- Currently on leave
Research
I'm interested in computer vision, deep learning, medical imaging, 3D graphics, and the connection between physics, simulation, and learning-based systems. My research record is early-career and currently centers on coauthored medical imaging work that uses deep learning to enhance ultra-low-dose PET/MRI.
Selected Papers
-
Investigating Simultaneity for Deep Learning-enhanced Actual Ultra-low-dose Amyloid PET/MRI Imaging
American Journal of Neuroradiology, 2022
Deep-learning enhancement of actual ultra-low-dose amyloid PET/MR imaging, evaluating whether simultaneous structural MRI is required for diagnostic-quality reconstruction.
-
Generalizing Ultra-low-dose PET/MRI Networks Across Radiotracers: From Amyloid to Tau
ISMRM, 2021
Study of whether an ultra-low-dose amyloid PET/MRI enhancement network generalizes across radiotracers from amyloid to tau imaging.
-
Does Simultaneous Morphological Inputs Matter for Deep Learning Enhancement of Ultra-low Amyloid PET/MRI?
ISMRM, 2021
Study of whether simultaneous morphological MRI inputs are necessary for deep-learning enhancement of ultra-low-dose amyloid PET/MRI.