ABSTRACT
Objective To evaluate the factors affecting the measurement of lung tumor activity concentration (AC) with FDG-PET. Methods High count Monte-Carlo simulations,based on VIP-Man (marked VHP atlas) added with spherical lung tumors of different diameter and activity,were performed to generate the essentially noiseless projection data for PET. The factors related with the measurement of tumor AC were evaluated. Results Using the maximum value in the tumor region of interest (ROI) generally resulted in the over-estimation of AC,and the degree of over-estimation would also increase as the noise became severer. For the small tumors with low AC,the proper partial volume correction was needed. Drawing ROI after post-reconstruction filtering could improve the reproducibility of AC measurement compared with no filtering first. Conclusion The accuracy and reliability of AC measurement depends strongly on factors including tumor size,post-reconstruction filter,noise level,ROI definition,PVE correction,etc.