Abstract
We present in this paper an application of minimal surfaces and Markov random fields to the segmentation of liver tumors. The originality of the work consists in applying these models to the region adjacency graph of a watershed transform. We detail the assumptions and the approximations introduced in these models by using a region graph instead of a pixel graph. This strategy leads to an interactive method that we use to delineate tumors in 3D CT images. We detail our strategy to achieve relevant segmentations of these structures and compare our results to hand made segmentations done by experienced radiologists. This paper summarizes our participation to the MICCAI 2008 workshop called: "3D segmentation in the clinic : A Grand Challenge II".
Keywords
Source Code and Data
No source code files available for this publication.
Reviews
Xiang Deng
Friday 25 July 2008
This paper presents an interactive live tumor segmentation technique. The liver is first segmented using combined graph-cut and watershed algorithm, tumors are then segmented with a maximum a posteriori method. Manual editing is followed to refine the tumor segmentation.
Detailed comments:
1) Could the author briefly describe the average computation time of tumor segmentation and time for manual editing?
That would give readers some sense of the potential of clincial application of the proposed method.
2) For Table 4 in the manuscript, please use the same table as what we provided in the evaluation (tumor name, metric value, score, etc.) for the sake of consistency. The caption could be "Results of comparison metrics and scores for all ten test tumors"
3) As the authors state in the manuscript, "the first step of our method is manual definition of a sub volume containing one or more tumors that need to be segmented." If manually defined ROI is required to initialize the tumor segmentation, what's the point of extraction of liver boundary in Section 3.3