000170060 001__ 170060
000170060 005__ 20190401065042.0
000170060 0247_ $$2doi$$a10.1109/TMI.2011.2171705
000170060 022__ $$a0278-0062
000170060 02470 $$2ISI$$a000300197500027
000170060 037__ $$aARTICLE
000170060 245__ $$aSupervoxel-Based Segmentation of Mitochondria in EM Image Stacks with Learned Shape Features
000170060 260__ $$bInstitute of Electrical and Electronics Engineers$$c2012
000170060 269__ $$a2012
000170060 336__ $$aJournal Articles
000170060 520__ $$aIt is becoming increasingly clear that mitochondria play an important role in neural function. Recent studies show mitochondrial morphology to be crucial to cellular physiology and synaptic function and a link between mitochondrial defects and neuro-degenerative diseases is strongly suspected. EM microscopy, with its very high resolution in all three directions, is one of the key tools to look more closely into these issues but the huge amounts of data it produces make automated analysis necessary. State-of-the-art computer vision algorithms designed to operate on natural 2D images tend to perform poorly when applied to EM data for a number of reasons. First, the sheer size of a typical EM volume renders most modern segmentation schemes intractable. Furthermore, most approaches ignore important shape cues, relying only on local statistics that easily become confused when confronted with noise and textures inherent in the data. Finally, the conventional assumption that strong image gradients always correspond to object boundaries is violated by the clutter of distracting membranes. In this work, we propose an automated graph partitioning scheme that addresses these issues. It reduces the computational complexity by operating on supervoxels instead of voxels, incorporates shape features capable of describing the 3D shape of the target objects, and learns to recognize the distinctive appearance of true boundaries. Our experiments demonstrate that our approach is able to segment mitochondria at a performance level close to that of a human annotator, and outperforms a state-of-the-art 3D segmentation technique.
000170060 6531_ $$aElectron microscopy
000170060 6531_ $$asegmentation
000170060 6531_ $$asupervoxels
000170060 6531_ $$amitochondria
000170060 6531_ $$ashape features
000170060 700__ $$0242715$$aLucchi, Aurélien$$g185205
000170060 700__ $$0242712$$aSmith, Kevin$$g163328
000170060 700__ $$0242495$$aAchanta, Radhakrishna$$g172126
000170060 700__ $$aKnott, Graham
000170060 700__ $$0240252$$aFua, Pascal$$g112366
000170060 773__ $$j31$$k2$$q474-486$$tIEEE Transactions on Medical Imaging
000170060 8564_ $$s9256829$$uhttps://infoscience.epfl.ch/record/170060/files/top.pdf$$yn/a$$zn/a
000170060 909C0 $$0252087$$pCVLAB$$xU10659
000170060 909C0 $$0252320$$pIVRL$$xU10429
000170060 909CO $$ooai:infoscience.tind.io:170060$$pIC$$particle$$qGLOBAL_SET
000170060 917Z8 $$x185205
000170060 917Z8 $$x125681
000170060 937__ $$aEPFL-ARTICLE-170060
000170060 973__ $$aEPFL$$rREVIEWED$$sPUBLISHED
000170060 980__ $$aARTICLE