Interactive Contour Delineation and Refinement in Treatment Planning of Image Guided Radiation Therapy
w zhou*, Y Xie, Shenzhen Institutes of Advanced TechnologyShenzhenSU-E-J-113 Sunday 3:00PM - 6:00PM Room: Exhibit Hall
The accurate contour delineation of the target and/or organs at risk (OAR) is essential in treatment planning for image guided radiation therapy (IGRT). Although lots of automatic contour delineation approaches have been proposed, few of them can fulfill the necessities of applications in terms of accuracy and efficiency. Moreover, doctors would like to analyze the characteristics of regions of interests (ROI) and adjust contours manually during IGRT. The objective of this work is to develop an efficient tool of curve fitting for interactive contour delineation in IGRT.
In this work, a novel approach of curve fitting for interactive contour delineation is proposed. Initially, a region which contains the interesting object is selected by mouse clicks and motion manually, then the program could automatically select important control points from the region, and the fitting method of Hermite cubic curves will be used to fit the control points. Hence, the optimized curve could be revised by moving its control points interactively. Finally, in order to improve the accuracy of contour delineation, the process of the curve refinement based on the maximum gradient magnitude is proposed. Points on the curve are automatically adjusted towards the positions with maximum gradient magnitude.
Hermite cubic curves exhibit superior efficiency for interactive segmentation and contour delineation of medical images, and the proposed scheme of the curve refinement greatly improves the final contour delineation to be subpixel level. Experimental results of real images in clinic applications demonstrate the efficiency, accuracy and robustness of the proposed process.
We have designed a fast tool for interactive contour delineation in the application of IGRT. Experimental results show that Hermite cubic curves and the curve refinement based on the maximum gradient magnitude possess superior performance on the proposed platform in terms of accuracy, robustness and calculation time.
Funding Support, Disclosures, and Conflict of Interest: This work is supported in part by grants from National Natural Science Foundation of China (NSFC: 81171402), NSFC Joint Research Fund for Overseas Research Chinese, Hong Kong and Macao Young Scholars (30928030), National Basic Research Program 973 (2010CB732606) from Ministry of Science and Technology of China, and Guangdong Innovative Research Team Program (No. 2011S013) of China.