N. Brieu, B. Glocker, N. Navab, M. Groher MAP-MRF Optimal Partitioning for Dynamic Texture Segmentation of Thrombus in Time-Series Microscopic Images MICCAI 2010 Workshop on Spatio Temporal Image Analysis for Longitudinal and Time-Series Image Data (STIA'10), 24 September 2010, Beijing, China (bib)
	An automatic segmentation of forming thrombus in dynamic microscopic scenes is proposed in this work. We perform a prerequisite temporal partitioning of the image time series and subsequently optimally segment the thrombus region in each frame of the sequence. The partition criterion, which encourages cuts at time points where the areas of the dynamic objects significantly change, is optimized in a MAP-MRF framework. We employ the resulting image intervals to derive maps of dynamic textures, which reveal regions with different temporal behavior. These maps are finally integrated into a level set segmentation, which fuses regional and temporal cues to achieve an accurate result for each frame. Tests on synthetic and real data of zebrafish specimen reveal the high potential of the presented method compared to other approaches.
	This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each authors copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.