Development of a novel imaging technique for orthogonal kilovoltage portal images of thoracic radiotherapy

• Main Authors: Shih-Yen Wang, 王士彥
• Other Authors: 董尚倫
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/hh28br

碩士 === 中山醫學大學 === 生物醫學科學學系碩士班 === 102 === Introduction The orthogonal kilovoltage (kV) X-ray portal imaging is widely used in image-guided radiation therapy (IGRT) for the patient setup verification. In recent years, assessments of the radiation dose and image quality of kV portal imaging have become increasingly important. In thoracic IGRT, the anterior-posterior (AP) projection kV portal imaging is frequently used. It is helpful for reducing the patient doses by using the posterior-anterior (PA) projection kV portal imaging. The purpose of this study is the development of a novel imaging technique for orthogonal kV portal images of thoracic radiotherapy. Materials and Method A linac-based (Varian 21ix) on-board image system was used. To simulate kV portal images of patient, a RANDO phantom was imaged using clinical chest exposure settings. PMMA slabs were used to simulate bone tissues. AP and PA kV portal images were acquired. The mean and standard deviation of pixel values of vertebral body and lung tissue were measured for each image and then the contrast-to-noise ratio (CNR) was calculated. The equivalent thickness of PMMA that would require the same pixel values as the RANDO phantom was estimated. Then, the CDRAD phantom was placed in the bottom of the PMMA slabs to simulate the real position of vertebra. The CDRAD images were analyzed by using the CDRAD analyzer program. The inverse image quality figure (IQFinv) of each CDRAD image was calculated. The PC-based Monte Carlo program (PCXMC) was used for calculating the organ doses and effective dose of each exposure. The doses, CNRs, and IQFinvs of AP and PA kV portal images were compared. Results The equivalent thickness of vertebra was 15 cm, and was thicker than that of chest about 25%. For image quality assessments, the CNR of AP projection (bone to lung tissue) was higher than that of PA projection about 11%. When the CDRAD was positioned in the bottom of the 14 cm PMMA slabs, the IQFinv values of AP and PA projection images were 2.19 and 2.22 (1/mm2), respectively. For dosimetry assessments, the ratio of entrance air KERMA between AP and PA projection was 0.87±0.04 (μGy/μGy). Applying PA projection kV portal imaging, the organ doses of breast, thymus and heart decreased about 92, 94, and 78%. The doses of lung and esophagus form AP and PA projection were comparable. The effective doses decreased 44±4% when PA projection were applied. Conclusion Using the PA projection kV portal imaging, the doses to RANDO phantom can be significantly reduced and the image quality was slightly decreased. The PA projection kV portal imaging may be suitable for orthogonal kV portal images of thoracic IGRT.