Medical Physics Seminar – Monday, May 20, 2013
Comprehensive Characterization of [18F]NaF PET Imaging Uncertainties in a Multicenter Trial
Ngoneh Jallow (student of Dr. Robert Jeraj)
Research Assistant, Department of Medical Physics, UW-School of Medicine & Public Health, Madison, WI - USA –
Quantitative PET imaging is increasingly being used as a tool for treatment response assessment of various types of cancers replacing the conventional qualitative anatomical imaging. The overall aim of this work was to comprehensively characterize the uncertainties and errors in quantitative PET imaging. We developed and tested a methodology on a multi-center clinical trial using [18F]NaF PET/CT to assess treatment efficacy of metastatic bone lesions. In order to assess repeatability, the trial included pre-treatment test/retest of [18F]NaF PET/CT.
Quantitative harmonization of the data from the PET/CT scanners used in the trial was performed to increase quantitative accuracy of [18F]NaF PET/CT. The repeatability was determined for each patient, each center as well as between the centers. Uncertainties and errors in [18F]NaF PET imaging of multiple lesions throughout the imaging process were characterized for reconstruction method, partial volume effects, etc. Finally, the impact of these uncertainties and errors on repeatability and the response assessment was evaluated.