Automatic IVUS segmentation of atherosclerotic plaque with Stop & Go snake

Since the upturn of intravascular ultrasound (IVUS)as an imaging technique for the coronary arterysystem, much research has been done to simplify thecomplicated analysis of the resulting images. In thisstudy, an attempt to develop an automatic tissuecharacterization algorithm for IVUS images wasdone. We concentrated on the segmentation ofcalcium and soft plaque, because these structurespredict the extension and the vulnerability of theatherosclerotic disease, respectively. The first step in the procedure was the extraction of texture featureslike local binary patterns, co-occurrence matricesand Gabor filter banks. After dimensionalityreduction, the resulting feature space was used forclassification, constructing a likelihood map torepresent different coronary plaques. Theinformation in this map was organized using arecently developed geodesic snake formulation,the so-called Stop & Go snake. The novelty of ourstudy lies in this last step, as it was the first time to apply the Stop & Go snake to segment IVUSimages