%0 Journal Article
%T Rapid Automated Target Segmentation and Tracking on 4D Data without Initial Contours
%A Venkata V. Chebrolu
%A Daniel Saenz
%A Dinesh Tewatia
%A William A. Sethares
%A George Cannon
%A Bhudatt R. Paliwal
%J Radiology Research and Practice
%D 2014
%I Hindawi Publishing Corporation
%R 10.1155/2014/547075
%X Purpose. To achieve rapid automated delineation of gross target volume (GTV) and to quantify changes in volume/position of the target for radiotherapy planning using four-dimensional (4D) CT. Methods and Materials. Novel morphological processing and successive localization (MPSL) algorithms were designed and implemented for achieving autosegmentation. Contours automatically generated using MPSL method were compared with contours generated using state-of-the-art deformable registration methods (using and MIMVista software). Metrics such as the Dice similarity coefficient, sensitivity, and positive predictive value (PPV) were analyzed. The target motion tracked using the centroid of the GTV estimated using MPSL method was compared with motion tracked using deformable registration methods. Results. MPSL algorithm segmented the GTV in 4DCT images in seconds per phase ( resolution) as compared to seconds per phase for deformable registration based methods in 9 cases. Dice coefficients between MPSL generated GTV contours and manual contours (considered as ground-truth) were . In comparison, the Dice coefficients between ground-truth and contours generated using deformable registration based methods were 0.909 ¡À 0.051. Conclusions. The MPSL method achieved similar segmentation accuracy as compared to state-of-the-art deformable registration based segmentation methods, but with significant reduction in time required for GTV segmentation. 1. Introduction In the practice of radiation therapy better local control and survival are often associated with increased delivered dose [1]. The greatest limitation to increasing treatment dose is induced by normal lung toxicity. Due to nonperiodic breathing pattern in patients, the planned dose is very often not delivered as intended. Interfractional target motion considerably deteriorates the geometric accuracy of the delivery process. In the recent past, systems and methodologies such as TomoTherapy [2, 3] and cone-beam computer tomography (CBCT) [4, 5] were developed and used in clinical practice to improve treatment planning and delivery. A quick and accurate method of contouring structures would be useful to improve the efficacy of these systems. Manual segmentation is too time-consuming, making rapid imaging and automated target delineation very attractive for motion management in radiation therapy. A typical four-dimensional (4D) data for radiation treatment planning in lung cancer includes 10 phases (separated by 10% difference from 0 to 100% of the breathing cycle) and approximately 100 images per phase. To estimate
%U http://www.hindawi.com/journals/rrp/2014/547075/