Evaluation of digital PET/CT system for myocardial perfusion imaging

Myocardial perfusion imaging (MPI) with Positron Emission Tomography (PET) allows quantitative measurements of absolute myocardial blood flow (MBF). PET system count-rate capabilities, reconstruction techniques, and other technical factors may influence MBF quantification reproducibility and accuracy. In this thesis the aims were to evaluate the effect of different reconstruction parameters on [15O]H2O MPI using a flow phantom and clinical retrospective data from patients who had undergone [15O]H2O MPI for suspected obstructive coronary artery disease. Also, the digital and analog PET system count-rate capabilities were assessed in high count-rate studies. Finally, the aim was to establish the contribution of technical factors on quantitative reproducibility and accuracy on two digital PET systems. The different reconstruction parameters resulted in a 7 % relative error with the image-derived flow values compared to the reference flow values in phantom studies. Similar differences were measured in MBF values in patients. Also, different reconstruction algorithms resulted in similar classification of myocardial ischemia in 99 % of the subjects. The digital PET resulted in 12.8 Mcps total prompts and 0.47 Mcps trues, and the analog PET in 6.85 Mcps total prompts and 1.15 Mcps trues with the highest injected activities. The modelled flow values were reproducible on digital PET systems but future studies need to be conducted to develop a standardized and repeatable bolus injection protocol. The results of these studies showed that the digital PET system can be reliably used in MPI in terms of system count-rate capabilities and novel reconstruction techniques with small contribution from technical factors. The findings offer a basis for assessing reproducibility in MPI in multi-center studies