The Effect of Lesion Size and Activity Concentration on Image Quality in PET/CT Imaging: A Phantom Study

Abstract

In this study, the National Electrical Manufacturers Association (NEMA) image quality tests were conducted using the General Electric Discovery 690 model — a hybrid system of positron emission tomography/computed tomography. Quantitative evaluation of image quality within this model primarily involves the use of parameters such as contrast (C), contrast-to-noise ratio (CNR), and contrast recovery coefficient (CRC). These metrics are considered essential performance indicators, particularly in standardized assessment protocols such as those proposed by NEMA. The objective of this study was to evaluate image quality in a NEMA positron emission tomography phantom using ¹⁸F-FDG at different signal-to-background (S/Bg) ratios and varying activity concentrations, and to determine the minimum detectable lesion size. Positron emission tomography images were acquired using a NEMA 2012/IEC 2008 phantom with ¹⁸F-FDG at S/Bg ratios of 4:1, 6:1, and 8:1, across decreasing activity concentrations ranging from 9.98 to 1.93 MBq/kg. At an S/Bg ratio of 4:1, activity concentrations below 3.90 MBq/kg (as well as at 6:1 below 2.76 MBq/kg) resulted in CNR values less than 5, falling below the detectability threshold. Regarding the CRC [%] values, for the 4:1 ratio, lesions with diameters of 13 and 10 mm fell outside the acceptable limits at concentrations below 2.89 MBq/kg. Similarly, for the 6:1 ratio, lesions with a diameter of 10 mm exceeded these limits at concentrations below 5.29 MBq/kg. Multiple acquisitions were performed at different ratios and activity levels, and the lesion detectability for each condition was determined based on the Rose criterion (CNR ≥ 5).