Estimation of Error in Maximal Intensity Projection-Based Internal Target Volume of Lung Tumors: A Simulation and Comparison Study Using Dynamic Magnetic Resonance Imaging

Purpose: To evaluate the error in four-dimensional computed tomography (4D-CT) maximal intensity projection (MIP)-based lung tumor internal target volume determination using a simulation method based on dynamic magnetic resonance imaging (dMRI). Methods and Materials: Eight healthy volunteers and six lung tumor patients underwent a 5-min MRI scan in the sagittal plane to acquire dynamic images of lung motion. A MATLAB program was written to generate re-sorted dMRI using 4D-CT acquisition methods (RedCAM) by segmenting and rebinning the MRI scans. The maximal intensity projection images were generated from RedCAM and dMRI, and the errors in the MIP-based internal target area (ITA) from RedCAM (ε), compared with those from dMRI, were determined and correlated with the subjects' respiratory variability (ν). Results: Maximal intensity projection-based ITAs from RedCAM were comparatively smaller than those from dMRI in both phantom studies (ε = -21.64% ± 8.23%) and lung tumor patient studies (ε = -20.31% ± 11.36%). The errors in MIP-based ITA from RedCAM correlated linearly (ε = -5.13ν - 6.71, r2= 0.76) with the subjects' respiratory variability. Conclusions: Because of the low temporal resolution and retrospective re-sorting, 4D-CT might not accurately depict the excursion of a moving tumor. Using a 4D-CT MIP image to define the internal target volume might therefore cause underdosing and an increased risk of subsequent treatment failure. Patient-specific respiratory variability might also be a useful predictor of the 4D-CT-induced error in MIP-based internal target volume determination.