A Concept for the Application of a Hierarchical Image Subdivision to the Segmentation Editing Problem

Efficient 3D segmentation editing tools are important components in the segmentation process for cases where automatic segmentation algorithms fail to provide sufficient results. In this paper, we propose a novel generic concept for 3D manual correction of a given segmentation mask. It is based on a hierarchical subdivision of the image generated by the Interactive Watershed Transform. The user adds missing parts to the given segmentation or removes parts from it by coarsely defining foreground and background regions via markers. Our method is independent of the algorithm by which the initial segmentation has been generated, without the requirement of mapping its result to the concepts and data structures of the algorithm used in the editing step. This allows modifications to be kept local. In addition, only a few assumptions on the object of interest are made. We have successfully applied our algorithm to bone segmentation in CT angiography.

[1]  Pascal Fua,et al.  SLIC Superpixels Compared to State-of-the-Art Superpixel Methods , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Gareth Funka-Lea,et al.  An Energy Minimization Approach to the Data Driven Editing of Presegmented Images/Volumes , 2006, MICCAI.

[3]  Alexandre X. Falcão,et al.  How to Complete Any Segmentation Process Interactively via Image Foresting Transform , 2010, 2010 23rd SIBGRAPI Conference on Graphics, Patterns and Images.

[4]  Heinz-Otto Peitgen,et al.  Lung lobe segmentation by anatomy-guided 3D watershed transform , 2003, SPIE Medical Imaging.

[5]  Marco Nolden,et al.  Interactive segmentation framework of the Medical Imaging Interaction Toolkit , 2009, Comput. Methods Programs Biomed..

[6]  Marie-Pierre Jolly,et al.  A splice-guided data driven interactive editing , 2013, 2013 IEEE 10th International Symposium on Biomedical Imaging.

[7]  Adam P. Harrison,et al.  IntellEditS: Intelligent Learning-Based Editor of Segmentations , 2013, MICCAI.

[8]  Takashi Ijiri,et al.  Contour‐based Interface for Refining Volume Segmentation , 2010, Comput. Graph. Forum.

[9]  C D Claussen,et al.  Automatic bone and plaque removal using dual energy CT for head and neck angiography: feasibility and initial performance evaluation. , 2010, European journal of radiology.

[10]  Yan Kang,et al.  Interactive 3D editing tools for image segmentation , 2004, Medical Image Anal..

[11]  Jan Hendrik Moltz,et al.  Sketch‐Based Editing Tools for Tumour Segmentation in 3D Medical Images , 2013, Comput. Graph. Forum.

[12]  Hiroaki Naito,et al.  Dual-energy direct bone removal CT angiography for evaluation of intracranial aneurysm or stenosis: comparison with conventional digital subtraction angiography , 2009, European Radiology.

[13]  Heinz-Otto Peitgen,et al.  The Skull Stripping Problem in MRI Solved by a Single 3D Watershed Transform , 2000, MICCAI.

[14]  Horst K. Hahn,et al.  A Generic Concept for Object-based Image Analysis , 2010, VISAPP.

[15]  Heinz-Otto Peitgen,et al.  HWT - hybrid watershed transform: optimal combination of hierarchical interactive and automated image segmentation , 2007, SPIE Medical Imaging.

[16]  Heinz-Otto Peitgen,et al.  IWT-interactive watershed transform: a hierarchical method for efficient interactive and automated segmentation of multidimensional gray-scale images , 2003, SPIE Medical Imaging.

[17]  Peter Mindek,et al.  ViviSection: Skeleton‐based Volume Editing , 2013, Comput. Graph. Forum.

[18]  Gopal B. Avinash,et al.  Semiautomatic bone removal technique from CT angiography data , 2001, SPIE Medical Imaging.

[19]  Samuel S. Silva,et al.  A 3D Tool for Left Ventricle Segmentation Editing , 2010, ICIAR.

[20]  Leonid I. Dimitrov,et al.  Bone Segmentation in CT Angiography Data Using a Probabilistic Atlas , 2003, VMV.

[21]  Horst Bischof,et al.  Liver Segmentation in CT Data: A Segmentation Refinement Approach , 2007 .