Home Department: undeclared
Mentor: Craig Levin (Radiology)

“Enhancing PET Image Reconstruction via Deep Learning Models”

Positron emission tomography (PET) is an imaging technique that is used to study cancer, heart disease, and neurological disorders. Two commonly used image reconstruction algorithms are used to visualize PET data: OSEM (faster but makes fuzzier images) and BSREM (slower but generates more accurate images). Ethan’s research project aims to develop a deep learning (DL) model to improve OSEM-reconstructed images to a quality comparable to that of BSREM-reconstructed PET images. If successful, this approach allows faster reconstruction of high-quality PET images.