Encrypted login | home

Program Information

Reconstruction of Cone Beam CT Image Using Volumetric Modulated Arc Therapy Exit Beams

no image available
K Jeong

K Jeong1*, H Kuo2 , L Goddard1 , M Savacool1 , A Basavatia2 , L Hong2 , R Yaparpalvi2, D Mynampati1 , P Godoy Scripes1 , S Kalnicki2, W Tome'2 , (1) Montefiore Medical Center, Bronx, NY, (2) Montefiore Medical Center/Albert Einstein College of Medicine, Bronx, New York

Presentations

SU-E-J-99 Sunday 3:00PM - 6:00PM Room: Exhibit Hall

Purpose: To test the possibility of obtaining an image of the treated volume during volumetric modulated arc therapy (VMAT) with exit beams.

Method: Using a Varian Clinac 21EX and MVCT detector the following three sets of detector projection data were obtained for cone beam CT reconstruction with and without a Catphan 504 phantom. 1) 72 projection images from 20 x 16 cm^2 open beam with 3 MUs, 2) 72 projection images from 20 x 16 cm^2 MLC closed beam with 14 MUs. 3) 137 projection images from a test RapicArc QA plan. All projection images were obtained in 'integrated image' mode. We used OSCaR code to reconstruct the cone beam CT images. No attempts were made to reduce scatter or artifacts.

Results: With projection set 1) we obtained a good quality MV CBCT image by optimizing the reconstruction parameters. Using projection set 2) we were not able to obtain a CBCT image of the phantom, which was determined to be due to the variation of interleaf leakage with gantry angle. From projection set 3), we were able to obtain a weak but meaningful signal in the image, especially in the target area where open beam signals were dominant. This finding suggests that one might be able to acquire CBCT images with rough body shape and some details inside the irradiated target area.

Conclusion: Obtaining patient images using the VMAT exit beam is challenging but possible. We were able to determine sources of image degradation such as gantry angle dependent interleaf leakage and beams with a large scatter component. We are actively working on improving image quality.



Contact Email: