Fast and Memory Efficient Segmentation of Lung Tumors Using Graph Cuts

In medical imaging, segmenting accurately lung tumors stay a quite challenging task when touching directly with healthy tissues. In this paper, we address the problem of extracting interactively these tumors with graph cuts. The originality of this work consists in (1) reducing input graphs to reduce resource consumption when segmenting large volume data and (2) introducing a novel energy formulation to inhibit the propagation of the object seeds. We detail our strategy to achieve relevant segmentations of lung tumors and compare our results to hand made segmentations provided by an expert. Comprehensive experiments show how our method can get solutions near from ground truth in a fast and memory efficient way.