Automated Detection of Synapses in Serial Section Transmission Electron Microscopy Image Stacks. Kreshuk, A., Koethe, U., Pax, E., Bock, D. D., & Hamprecht, F. A. Plos One, 9(2):e87351, February, 2014. WOS:000330830700024doi abstract bibtex We describe a method for fully automated detection of chemical synapses in serial electron microscopy images with highly anisotropic axial and lateral resolution, such as images taken on transmission electron microscopes. Our pipeline starts from classification of the pixels based on 3D pixel features, which is followed by segmentation with an Ising model MRF and another classification step, based on object-level features. Classifiers are learned on sparse user labels; a fully annotated data subvolume is not required for training. The algorithm was validated on a set of 238 synapses in 20 serial 719767351 pixel images (4.5x4.5x45 nm resolution) of mouse visual cortex, manually labeled by three independent human annotators and additionally re-verified by an expert neuroscientist. The error rate of the algorithm (12% false negative, 7% false positive detections) is better than state-of-the-art, even though, unlike the state-of-the-art method, our algorithm does not require a prior segmentation of the image volume into cells. The software is based on the ilastik learning and segmentation toolkit and the vigra image processing library and is freely available on our website, along with the test data and gold standard annotations (http://www.ilastik.org/synapse-detection/sstem).
@article{kreshuk_automated_2014,
title = {Automated {Detection} of {Synapses} in {Serial} {Section} {Transmission} {Electron} {Microscopy} {Image} {Stacks}},
volume = {9},
issn = {1932-6203},
doi = {10.1371/journal.pone.0087351},
abstract = {We describe a method for fully automated detection of chemical synapses in serial electron microscopy images with highly anisotropic axial and lateral resolution, such as images taken on transmission electron microscopes. Our pipeline starts from classification of the pixels based on 3D pixel features, which is followed by segmentation with an Ising model MRF and another classification step, based on object-level features. Classifiers are learned on sparse user labels; a fully annotated data subvolume is not required for training. The algorithm was validated on a set of 238 synapses in 20 serial 719767351 pixel images (4.5x4.5x45 nm resolution) of mouse visual cortex, manually labeled by three independent human annotators and additionally re-verified by an expert neuroscientist. The error rate of the algorithm (12\% false negative, 7\% false positive detections) is better than state-of-the-art, even though, unlike the state-of-the-art method, our algorithm does not require a prior segmentation of the image volume into cells. The software is based on the ilastik learning and segmentation toolkit and the vigra image processing library and is freely available on our website, along with the test data and gold standard annotations (http://www.ilastik.org/synapse-detection/sstem).},
language = {English},
number = {2},
journal = {Plos One},
author = {Kreshuk, Anna and Koethe, Ullrich and Pax, Elizabeth and Bock, Davi D. and Hamprecht, Fred A.},
month = feb,
year = {2014},
note = {WOS:000330830700024},
keywords = {challenges, circuit reconstruction, connectomics, neuropil, retina, segmentation},
pages = {e87351},
}
Downloads: 0
{"_id":"Q4MoJBPgSmmdY6dBF","bibbaseid":"kreshuk-koethe-pax-bock-hamprecht-automateddetectionofsynapsesinserialsectiontransmissionelectronmicroscopyimagestacks-2014","author_short":["Kreshuk, A.","Koethe, U.","Pax, E.","Bock, D. D.","Hamprecht, F. A."],"bibdata":{"bibtype":"article","type":"article","title":"Automated Detection of Synapses in Serial Section Transmission Electron Microscopy Image Stacks","volume":"9","issn":"1932-6203","doi":"10.1371/journal.pone.0087351","abstract":"We describe a method for fully automated detection of chemical synapses in serial electron microscopy images with highly anisotropic axial and lateral resolution, such as images taken on transmission electron microscopes. Our pipeline starts from classification of the pixels based on 3D pixel features, which is followed by segmentation with an Ising model MRF and another classification step, based on object-level features. Classifiers are learned on sparse user labels; a fully annotated data subvolume is not required for training. The algorithm was validated on a set of 238 synapses in 20 serial 719767351 pixel images (4.5x4.5x45 nm resolution) of mouse visual cortex, manually labeled by three independent human annotators and additionally re-verified by an expert neuroscientist. The error rate of the algorithm (12% false negative, 7% false positive detections) is better than state-of-the-art, even though, unlike the state-of-the-art method, our algorithm does not require a prior segmentation of the image volume into cells. The software is based on the ilastik learning and segmentation toolkit and the vigra image processing library and is freely available on our website, along with the test data and gold standard annotations (http://www.ilastik.org/synapse-detection/sstem).","language":"English","number":"2","journal":"Plos One","author":[{"propositions":[],"lastnames":["Kreshuk"],"firstnames":["Anna"],"suffixes":[]},{"propositions":[],"lastnames":["Koethe"],"firstnames":["Ullrich"],"suffixes":[]},{"propositions":[],"lastnames":["Pax"],"firstnames":["Elizabeth"],"suffixes":[]},{"propositions":[],"lastnames":["Bock"],"firstnames":["Davi","D."],"suffixes":[]},{"propositions":[],"lastnames":["Hamprecht"],"firstnames":["Fred","A."],"suffixes":[]}],"month":"February","year":"2014","note":"WOS:000330830700024","keywords":"challenges, circuit reconstruction, connectomics, neuropil, retina, segmentation","pages":"e87351","bibtex":"@article{kreshuk_automated_2014,\n\ttitle = {Automated {Detection} of {Synapses} in {Serial} {Section} {Transmission} {Electron} {Microscopy} {Image} {Stacks}},\n\tvolume = {9},\n\tissn = {1932-6203},\n\tdoi = {10.1371/journal.pone.0087351},\n\tabstract = {We describe a method for fully automated detection of chemical synapses in serial electron microscopy images with highly anisotropic axial and lateral resolution, such as images taken on transmission electron microscopes. Our pipeline starts from classification of the pixels based on 3D pixel features, which is followed by segmentation with an Ising model MRF and another classification step, based on object-level features. Classifiers are learned on sparse user labels; a fully annotated data subvolume is not required for training. The algorithm was validated on a set of 238 synapses in 20 serial 719767351 pixel images (4.5x4.5x45 nm resolution) of mouse visual cortex, manually labeled by three independent human annotators and additionally re-verified by an expert neuroscientist. The error rate of the algorithm (12\\% false negative, 7\\% false positive detections) is better than state-of-the-art, even though, unlike the state-of-the-art method, our algorithm does not require a prior segmentation of the image volume into cells. The software is based on the ilastik learning and segmentation toolkit and the vigra image processing library and is freely available on our website, along with the test data and gold standard annotations (http://www.ilastik.org/synapse-detection/sstem).},\n\tlanguage = {English},\n\tnumber = {2},\n\tjournal = {Plos One},\n\tauthor = {Kreshuk, Anna and Koethe, Ullrich and Pax, Elizabeth and Bock, Davi D. and Hamprecht, Fred A.},\n\tmonth = feb,\n\tyear = {2014},\n\tnote = {WOS:000330830700024},\n\tkeywords = {challenges, circuit reconstruction, connectomics, neuropil, retina, segmentation},\n\tpages = {e87351},\n}\n\n","author_short":["Kreshuk, A.","Koethe, U.","Pax, E.","Bock, D. D.","Hamprecht, F. A."],"key":"kreshuk_automated_2014","id":"kreshuk_automated_2014","bibbaseid":"kreshuk-koethe-pax-bock-hamprecht-automateddetectionofsynapsesinserialsectiontransmissionelectronmicroscopyimagestacks-2014","role":"author","urls":{},"keyword":["challenges","circuit reconstruction","connectomics","neuropil","retina","segmentation"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"article","biburl":"https://bibbase.org/zotero/spintextures","dataSources":["rXHvWQJHcL8ctHS4s"],"keywords":["challenges","circuit reconstruction","connectomics","neuropil","retina","segmentation"],"search_terms":["automated","detection","synapses","serial","section","transmission","electron","microscopy","image","stacks","kreshuk","koethe","pax","bock","hamprecht"],"title":"Automated Detection of Synapses in Serial Section Transmission Electron Microscopy Image Stacks","year":2014}