Impact and correction of the bladder uptake on ¹⁸F-FCH PET quantification: a simulation study using the XCAT2 phantom

The spill-in counts from neighbouring regions can significantly bias the quantification over small regions close to high activity extended sources. This effect can be a drawback for ¹⁸F-based radiotracers positron emission tomography (PET) when quantitatively evaluating the bladder area for diseases such as prostate cancer. In this work, we use Monte Carlo simulations to investigate the impact of the spill-in counts from the bladder on the quantitative evaluation of prostate cancer when using ¹⁸F-Fluorcholine (FCH) PET and we propose a novel reconstruction-based correction method. Monte Carlo simulations of a modified version of the XCAT2 anthropomorphic phantom with ¹⁸F-FCH biological distribution, variable bladder uptake and inserted prostatic tumours were used in order to obtain simulated realistic ¹⁸F-FCH data. We evaluated possible variations of the measured tumour Standardized Uptake Value (SUV) for different values of bladder uptake and propose a novel correction by appropriately adapting image reconstruction methodology. The correction is based on the introduction of physiological background terms on the reconstruction, removing the contribution of the bladder to the final image. The bladder is segmented from the reconstructed image and then forward-projected to the sinogram space. The resulting sinograms are used as background terms for the reconstruction. SUV_max and SUV_mean could be overestimated by 41% and 22% respectively due to the accumulation of radiotracer in the bladder, with strong dependence on bladder-to-lesion ratio. While the SUVs measured under these conditions are not reliable, images corrected using the proposed methodology provide better repeatability of SUVs, with biases below 6%. Results also showed remarkable improvements on visual detectability. The spill-in counts from the bladder can affect prostatic SUV measurements of ¹⁸F-FCH images, which can be corrected to less than 6% using the proposed methodology, providing reliable SUV values even in the presence of high radioactivity accumulation in the bladder. (paper)