A new simpler image quality index based on body size for FDG-PET/CT

Abstract

Noise equivalent count density (NECdensity) is often used to evaluate the image quality of whole-body fluorodeoxyglucose tomography tests. However, this index is calculated using the patient volume, which is difficult to obtain at every facility. In this study, we proposed new image quality indices that can be evaluated at all facilities. In total, 94 patients were enrolled in the study. The correlations of patients’ body weight and BMI with volume were examined. New image quality indices normalized by body weight and BMI were defined as NECbw and NECbmi, respectively. Correlations between NECbw, NECbmi, and NECdensity were examined. Further, the correlations between these two new indices and visual scores were evaluated. Good correlations were observed between volume and body weight (r = 0.861, P < 0.001) and between volume and BMI (r = 0.728, P < 0.001). NECbw and NECbmi correlated well with NECdensity (r = 0.954 for NECbw and r = 0.897 for NECbmi, P < 0.001). These correlations improved when the examined bed positions were set to the same number. Additionally, the correlations of visual scores with NECbw and NECbmi were similar to those between the visual score and NECdensity. Our investigation indicated that the newly proposed image quality metrics, NECbw and NECbmi, were easily calculated and as useful as NECdensity for evaluating image quality when subjects had similar physiques.