A Study of Portal Image Enhancement for Image Guided Radiation Therapy

• Main Authors: Yuan-Po Chen, 陳元柏
• Other Authors: Hueisch-Jy Din
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/13617112871666109652

碩士 === 義守大學 === 資訊工程學系碩士班 === 98 === Image guided radiation therapy (IGRT) is one of the advanced radiation therapy techniques. Verification images are used to align patient''s position with simulation images to achieve high precision therapy. Portal images are acquired by electronic portal imaging devices (EPID) with megavoltage (MV) x-ray, and they are inherently poor in terms of contrast due to Compton Effect . To compare with diagnostic x-ray images, portal image are usually lack of detail information for human vision. Thus, effective enhancement techniques should be developed to address this problem. In this thesis, we proposed a new approach which combines GLG (Gray Level Grouping) and Retinex theory by regional enhancement and pseudo-data filling, to effectively improve the visual quality of portal image. Results of this study demonstrate that the proposed method could sketch more details of bony structures than traditional image enhancement techniques and beneficial for IGRT.