RESUMO
PURPOSE: Respiration-induced motion artifacts in cone-beam CT (CBCT) can be corrected using a model of patient motion obtained from respiration-correlated CT (RCCT). This approach assumes that respiration-induced organ deformations at simulation, when RCCT scans are normally acquired, are still valid at treatment. The purpose of this study is to compare lung tumor image quality in motion-corrected CBCT images derived from treatment-day RCCT(tx) to simulation-day RCCT(sim) patient images. METHODS: In an IRB-approved study, lung cancer patients receive an RCCT at simulation, and an RCCT, gated CBCT and 1-minute CBCT at one treatment session. CBCT projections from the 1-minute scan are sorted according to breathing amplitude from an external monitor and reconstructed and warped to obtain a motion-corrected MC-CBCT at end expiration. Motion correction uses a model adapted from either RCCT(tx) or RCCT(sim), thus obtaining MC-CBCT(tx) and MC-CBCT(sim) images respectively. A gated CBCT, in which gantry rotation and projection acquisition occur within a gate at end expiration, serves as ground truth for comparison. Quality of MC-CBCT images is evaluated from tumor-to-background contrast ratio (TBCR) values measured by delineating the tumor and annular volume around it on the gated CBCT then transferring the contours and aligning them to each MC-CBCT. RESULTS: TBCR is found tobe lower in MC-CBCT(sim) images, relative to MC-CBCT(tx), in four out of five patients with mean 21% reduction in a range 9-39%. In the remaining case, where there was no change in TBCR, tumor motion observed in the RCCT was small (2mm). Tumor motion extent relative to diaphragm is observed to change between RCCT(tx) and RCCT(sim) scans. CONCLUSIONS: Preliminary results indicate that deformation patterns in lung do change between simulation and treatment. Such variations may reduce the validity of using simulation data for motion-corrected CBCT at treatment. The findings require confirmation with larger numbers of patients. NIH/NCI award R01 CA126993, research grant from Varian Medical Systems.
RESUMO
We have evaluated an automated registration procedure for predicting tumor and lung deformation based on CT images of the thorax obtained at different respiration phases. The method uses a viscous fluid model of tissue deformation to map voxels from one CT dataset to another. To validate the deformable matching algorithm we used a respiration-correlated CT protocol to acquire images at different phases of the respiratory cycle for six patients with nonsmall cell lung carcinoma. The position and shape of the deformable gross tumor volumes (GTV) at the end-inhale (EI) phase predicted by the algorithm was compared to those drawn by four observers. To minimize interobserver differences, all observers used the contours drawn by a single observer at end-exhale (EE) phase as a guideline to outline GTV contours at EI. The differences between model-predicted and observer-drawn GTV surfaces at EI, as well as differences between structures delineated by observers at EI (interobserver variations) were evaluated using a contour comparison algorithm written for this purpose, which determined the distance between the two surfaces along different directions. The mean and 90% confidence interval for model-predicted versus observer-drawn GTV surface differences over all patients and all directions were 2.6 and 5.1 mm, respectively, whereas the mean and 90% confidence interval for interobserver differences were 2.1 and 3.7 mm. We have also evaluated the algorithm's ability to predict normal tissue deformations by examining the three-dimensional (3-D) vector displacement of 41 landmarks placed by each observer at bronchial and vascular branch points in the lung between the EE and EI image sets (mean and 90% confidence interval displacements of 11.7 and 25.1 mm, respectively). The mean and 90% confidence interval discrepancy between model-predicted and observer-determined landmark displacements over all patients were 2.9 and 7.3 mm, whereas interobserver discrepancies were 2.8 and 6.0 mm. Paired t tests indicate no significant statistical differences between model predicted and observer drawn structures. We conclude that the accuracy of the algorithm to map lung anatomy in CT images at different respiratory phases is comparable to the variability in manual delineation. This method has therefore the potential for predicting and quantifying respiration-induced tumor motion in the lung.
