A texture-based direct 3D segmentation system for confocal scanning fluorescence microscopic images

The authors present a texture-based segmentation system for the identification of 3D structures in biomedical images. The segmentation is guided by an iterative octree expansion and (leaf mode) linking control algorithm and is augmented with knowledge-based classification processes. It performs segmentation directly in 3D space which is contrary to that done in common approaches where 3D structures are reconstructed from results of 2D segmentation of a sequence of consecutive, cross-sectional images. Test results of this system on real data confocal scanning fluorescence microscopic images of a developing chick embryo heart are reported