Fully automated organ bud detection and segmentation for Laser Capture Microdissection applications

Laser Capture Microdissection is a technique capable of isolating and extracting specific groups of cells of interest from microscopic regions in tissue samples. The final extraction result relies completely on the operator's ability to discriminate the tissue of interest, a very difficult task in the case of an embryo organ bud which doesn't even resemble the future organ. In this work we present a fully automated approach for localization of organ buds into embryos and segmentation of the corresponding tissue using topological, textural and morphological features processed by a Machine Learning approaches, and that can be generalized for any desired bud, species, developmental stage and preparation. © 2011 IEEE.