Program Information
Interfraction Contour Comparison On 4DCBCT Vs. 4DCT for Lung SBRT Patients
E Laugeman*, P Sevak , C Liu , K Snyder , N Wen , I Chetty , Henry Ford Health System, Detroit, MI
Presentations
SU-J-CAMPUS-JT-5 (Sunday, July 30, 2017) 4:00 PM - 5:00 PM Room: Joint Imaging-Therapy Theater
Purpose: To investigate the differences between the internal target volumes (ITV) contoured on the simulation 4DCT and daily 4DCBCT images for lung cancer patients treated with stereotactic body radiotherapy (SBRT).
Methods: 4DCBCTs were acquired before each fraction for Lung SBRT patients to assess tumor motion before treatment. A 4DCBCT was acquired with a full trajectory (gantry speed of 3deg/sec), and a total of 1800 projections. The Varian RPM device was used to bin the projections into 10 phases and reconstructed with filtered backprojection. An ITV was contoured on each of the 10 phases at each fraction delivered and the ITV was the union of the contours on each phase. The ITV was also drawn on the average CBCT used for the patient localization. The dice coefficient, hausdorff distance, and center of mass were used for comparison of the ITV contoured on the maximum intensity projection generated from 4DCT, Average CBCT, and 4DCBCT.
Results: Preliminary results show a higher dice coefficient (better similarity) between average CBCT and 4DCBCT ITVs (0.91±0.07) over four fractions for a patient compared to average CBCT and 4DCBCT comparisons with the 4DCT ITV (0.83±0.02 and 0.78±0.03, respectively). Greater hausdorff distances were found for comparisons with the 4DCBCT ITV, likely due to the streak artifacts in the 10 phase 4DCBCT. The center of mass difference for the 4DCT ITV compared to average CBCT ITV increased throughout treatment (from 0.94mm in the first fraction to 1.6mm in the fourth fraction) and compared to the 4DCBCT ITV (0.65mm to 2.39mm).
Conclusion: Daily 4DCBCT has the potential to assess the tumor motion before lung SBRT patient treatment. Differences in 4DCBCT ITV and 4DCT ITV are due to a number of factors including poor image quality and tumor shrinkage. Therefore, further investigation on the fidelity of 4DCBCT for ITV delineation is necessary.
Funding Support, Disclosures, and Conflict of Interest: The study was supported by a Research Scholar Grant, RSG-15-137-01-CCE from the American Cancer Society
Contact Email: