A simple correction method for the partial volume effect in Discovery STE PET/CT

Objectives We evaluated the partial volume effect in Discovery STE PET/CT images and developed a simple correction method to address this problem.

Methods Six spheres and the background in the phantom were filled with F-18 and we thus obtained 5 different Sphere/Background ratios (S/B ratios). Twenty-six cervical lymphnodes in 7 patients with papillary thyroid carcinoma (15 malignancies and 11 benigns) were also examined. First, we developed the recovery coefficient (RC) curves normalized to the maximum counts of the 37 mm sphere. Next, we developed a correction table to determine the true maximum counts using three parameters, including the maximum counts of both the sphere and background and the lesion diameter, by modifying the approximation formula of the RC curves.

Results The RC of the 10 mm sphere decreased to 0.58, and the gradient of the RC curves were different among each of the S/B ratios. After correction, the RC of the 10 mm sphere improved to 0.98. In clinical cases, the difference in the SUV_max between malignant and benign LNs thus became more significant after the correction. Furthermore, a positive significant correlation between the SUV_max and the diameter of the LNs disappeared after the correction.

Conclusions We developed a simple correction method for the partial volume effect and found this new method to be clinically useful for the diagnosis of LN metastasis.