Room: Exhibit Hall | Forum 6
Purpose: This work evaluates the ability of the Elekta 4D cone beam computed tomography (CBCT) Symmetry protocol to quantify internal target motion for lung and liver lesions.
Methods: A commercial respiratory motion phantom was used to simulate breathing motion for lung and liver targets. Respiratory correlated simulation 4D-CT scans were acquired for each treatment site at 10 and 15bpm, with the phantom moving 5mm and 15mm in the superior-inferior direction. Additional images were collected at each breathing rate with the targets moving along a 10mm diagonal in the sagittal plane. A planning internal target volume (ITV-P) was created from each 4D-CT scan. Corresponding 3D and 4D-CBCT images were collected using the Elekta Versa HD XVI imaging system. The 3D-CBCT was used to generate an ITV called the 3D-ITV. Images from each 4D-CBCT scan were then used to create maximum intensity projection (MIP) and time averaged (TA) CBCT data sets. An ITV was generated from each of these 4D-CBCT scans, called the MIP-ITV and TA-ITV, respectively. Each cone beam ITV volume was compared to the corresponding ITV-P using the dice similarity coefficient (DSC) and Hausdorff distance.
Results: Average DSCs for the 3D-ITV, MIP-ITV, and TA-ITV were 0.762, 0.797, and 0.776 for the lung target and 0.377, 0.547, and 0.456 for the liver target, respectively. The Hausdorff distance metric showed that the 4D-CBCT was able to quantify target motion better than the 3D-CBCT for most of the cases in this work. However, for the targets with 5mm motion, we observed that the 3D-CBCT performed better when compared to 4D-CBCT.
Conclusion: The Elekta 4D-CBCT Symmetry protocol is capable of accurately quantifying target motion for lung and liver lesions. From our study, we found that the 3D-CBCT can better delineate target motion that is less than 5mm when compared to a 4D-CBCT.