05/10/2023
By Jason Pagan

The Kennedy College of Sciences, Department of Physics and Applied Physics, invites you to attend a Master’s thesis defense by Jason Pagan on “Image Quality Improvement of 0.35T MRI Linac Images Using Cycle-Contrastive Unpaired Translation Networks.”

Candidate Name: Jason Pagan
Degree: MS
Defense Date: Tuesday, May, 23, 2023
Time: 11 a.m. to noon
Location: This will be a virtual defense via Zoom. Those interested in attending should contact MS candidate Jason_Pagan@student.uml.edu at least 24 hours prior to the defense to request access to the meeting.

Thesis/Dissertation Title: Image Quality Improvement of 0.35T MRI Linac Images Using Cycle-Contrastive Unpaired Translation Networks

Committee:

  • Committee Chair Atchar Sudhyadhom, Ph.D., DABR, Assistant Professor & Faculty Medical Physicist, Radiation Oncology, Harvard Medical School
  • Erno Sajo, Ph.D., M.Sc., Professor, Director of Medical Physics, Physics & Applied Physics, UMass Lowell
  • Timothy Cook, Ph.D., Associate Professor, Physics & Applied Physics, UMass Lowell

Abstract:
Motivated by the opportunity to improve patient care through enhanced onboard imaging, this study explores the potential of deep learning techniques to augment the quality and resolution of 0.35T MRI scans produced by the MRI Linac's onboard device for prostate cancer patients. Our investigation focuses on the application of the cycleCUT machine learning method to generate synthetic 3T MRI scans from 0.35T scans while preserving anatomical similarity. We aim to demonstrate the effectiveness of deep learning techniques in bridging the gap between low-field and high-field MRI scans, offering improved image quality without the need for high-field MRI equipment. The promising results of our study support the potential of deep learning methods like cycleCUT to significantly enhance low-field MRI scans, warranting further research with larger patient cohorts.