Verification of the tumor volume delineation method using a fixed threshold of peak standardized uptake value.

Abstract

We aimed to determine the difference in tumor volume associated with the reconstruction model in positron-emission tomography (PET). To reduce the influence of the reconstruction model, we suggested a method to measure the tumor volume using the relative threshold method with a fixed threshold based on peak standardized uptake value (SUVpeak). The efficacy of our method was verified using 18F-2-fluoro-2-deoxy-D-glucose PET/computed tomography images of 20 patients with lung cancer. The tumor volume was determined using the relative threshold method with a fixed threshold based on the SUVpeak. The PET data were reconstructed using the ordered-subset expectation maximization (OSEM) model, the OSEM+time-of-flight (TOF) model, and the OSEM+TOF+point-spread function (PSF) model. The volume differences associated with the reconstruction algorithm (%VD) were compared. For comparison, the tumor volume was measured using the relative threshold method based on the maximum SUV (SUVmax). For the OSEM and TOF models, the mean %VD values were -0.06±8.07 and -2.04±4.23% for the fixed 40% threshold according to the SUVmax and the SUVpeak, respectively. The effect of our method in this case seemed to be minor. For the OSEM and PSF models, the mean %VD values were -20.41±14.47 and -13.87±6.59% for the fixed 40% threshold according to the SUVmax and SUVpeak, respectively. Our new method enabled the measurement of tumor volume with a fixed threshold and reduced the influence of the changes in tumor volume associated with the reconstruction model.