Spotlight: automated confidence-based user guidance for increasing efficiency in interactive 3D image segmentation

作者: Andrew Top , Ghassan Hamarneh , Rafeef Abugharbieh

DOI: 10.1007/978-3-642-18421-5_20

关键词:

摘要: We present Spotlight, an automated user guidance technique for improving quality and efficiency of interactive segmentation tasks. Spotlight augments algorithms by automatically highlighting areas in need attention to the during interaction phase. employ a 3D Livewire algorithm as our base method where quickly provides minimal initial contour seeding. The is then evaluated based on three different metrics that probe edge strength, stability object connectivity. result this evaluation fed into novel autonomously suggests regions require intervention. Essentially, flags potentially problematic image prioritized fashion optimization process final segmentation. variety qualitative quantitative examples demonstrating Spotlight's intuitive use proven utility reducing input increasing automation.

参考文章(48)
D. M. Greig, B. T. Porteous, A. H. Seheult, Exact Maximum A Posteriori Estimation for Binary Images Journal of the royal statistical society series b-methodological. ,vol. 51, pp. 271- 279 ,(1989) , 10.1111/J.2517-6161.1989.TB01764.X
Burkhard C. Wünsche, Joachim Denzler, Stefan Schenke, GPU-Based Volume Segmentation ,(2005)
Shawn Andrews, Ghassan Hamarneh, Ahmed Saad, Fast Random Walker with Priors Using Precomputation for Interactive Medical Image Segmentation Medical Image Computing and Computer-Assisted Intervention – MICCAI 2010. ,vol. 13, pp. 9- 16 ,(2010) , 10.1007/978-3-642-15711-0_2
Pushmeet Kohli, Philip H. S. Torr, Measuring uncertainty in graph cut solutions – efficiently computing min-marginal energies using dynamic graph cuts european conference on computer vision. pp. 30- 43 ,(2006) , 10.1007/11744047_3
Leo Grady, Thomas Schiwietz, Shmuel Aharon, Rüdiger Westermann, Random Walks for Interactive Organ Segmentation in Two and Three Dimensions: Implementation and Validation Lecture Notes in Computer Science. ,vol. 8, pp. 773- 780 ,(2005) , 10.1007/11566489_95
Andrew Top, Ghassan Hamarneh, Rafeef Abugharbieh, Active Learning for Interactive 3D Image Segmentation Lecture Notes in Computer Science. ,vol. 14, pp. 603- 610 ,(2011) , 10.1007/978-3-642-23626-6_74
Simon F. Eskildsen, Lasse R. Østergaard, Active Surface Approach for Extraction of the Human Cerebral Cortex from MRI Medical Image Computing and Computer-Assisted Intervention – MICCAI 2006. ,vol. 9, pp. 823- 830 ,(2006) , 10.1007/11866763_101
Aaron E. Lefohn, Joshua E. Cates, Ross T. Whitaker, Interactive, GPU-based level sets for 3D segmentation medical image computing and computer assisted intervention. ,vol. 2878, pp. 564- 572 ,(2003) , 10.1007/978-3-540-39899-8_70
Eric N. Mortensen, William A. Barrett, Interactive segmentation with Intelligent Scissors Graphical Models and Image Processing. ,vol. 60, pp. 349- 384 ,(1998) , 10.1006/GMIP.1998.0480
M. Sadeghi, G. Tien, G. Hamarneh, M. S. Atkins, Hands-free interactive image segmentation using eyegaze Medical Imaging 2009: Computer-Aided Diagnosis. ,vol. 7260, ,(2009) , 10.1117/12.813452