Program Information
Automatic 3-D Prostate Segmentation CT Image Via Patch-Based Density Constraints Clustering
K Sheng1 , G Shuiping2*, Y Yao3 , (1) UCLA School of Medicine, Los Angeles, CA, (2) Xidian university, Xi'an, ShanXi, (3) HangZhou Vocational Technical College, HangZhou, ZheJiang.
Presentations
TU-C2-GePD-I-1 (Tuesday, August 1, 2017) 10:00 AM - 10:30 AM Room: Imaging ePoster Lounge
Purpose: To effectively manage tumor motion in prostate treatment using CT guided radiotherapy, automated segmentation needs to be performed. Among existing segmentation methods, 2D-based deformable models are the most popular. However, previous methods often assume all the slices are independent and ignore the context relations among the neighbored slices. Also, conventional deformable-based methods for segmenting prostate in CT images are time-consuming and manual intervention. In this study, an automatic 3-D prostate segmentation CT image via Patch-based density constraints clustering (PDCC) is developed.
Methods: Prostate is segmented with the following three strategies: 1) compared with only using pixel intensity information, Supervoxels-based 3D patch includes more structure contexts to deal with low contrast problem in prostate CT images. 2) Compacting and extracting discriminative information in the each patch with 3D gray-gradient co-occurrence matrix are used to distinguish tiny texture difference between prostate and non-prostate. 3) Density constraints clustering algorithm focus on a higher density than their neighbors’ points with relatively small distance to cope with two nearby organs touch together. Further, clusters are recognized regardless of their shape and of the dimensionality of the space in which they are embedded.
Results: The proposed method has been evaluated on 10 patients’ prostate CT image database where each patient includes 50 treatment images, and several state-of-the-art prostate CT segmentation algorithms with various evaluation metrics have been as comparisons.
Conclusion: Experimental results demonstrate that the proposed method achieves higher segmentation accuracy and lower average surface distance.
Funding Support, Disclosures, and Conflict of Interest: DOE DE-SC0017057 NIH R44CA183390 NIH R01CA188300 NIH R43CA183390 NIH U19AI067769 2015 ZheJiang Provincial Public Welfare Technology Application Research Project(2015C31167) 2016 HangZhou Social Development of Autonomous Application Projects(20160533B71)
Contact Email: