AI helps you reading Science

AI generates interpretation videos

AI extracts and analyses the key points of the paper to generate videos automatically


pub
Go Generating

AI Traceability

AI parses the academic lineage of this thesis


Master Reading Tree
Generate MRT

AI Insight

AI extracts a summary of this paper


Weibo:
We address a central problem of neuroanatomy, namely, the automatic segmentation of neuronal structures depicted in stacks of electron microscopy images

Deep Neural Networks Segment Neuronal Membranes in Electron Microscopy Images.

NIPS, pp.2852-2860, (2012)

Cited by: 1542|Views416
EI
Full Text
Bibtex
Weibo

Abstract

We address a central problem of neuroanatomy, namely, the automatic segmentation of neuronal structures depicted in stacks of electron microscopy (EM) images. This is necessary to efficiently map 3D brain structure and connectivity. To segment biological neuron membranes, we use a special type of deep artificial neural network as a pixel ...More

Code:

Data:

0
Introduction
  • How is the brain structured? The recent field of connectomics [2] is developing high-throughput techniques for mapping connections in nervous systems, one of the most important and ambitious goals of neuroanatomy.
  • A sample of neural tissue is typically sectioned into 50-nanometer slices; each slice is recorded as a 2D grayscale image with a pixel size of about 4 × 4 nanometers.
  • Reliable automated segmentation of neuronal structures in ssTEM stacks so far has been infeasible.
  • A solution of this problem, is essential for any automated pipeline reconstructing and mapping neural connections in 3D.
  • Recent advances in automated sample preparation and imaging make this increas-
Highlights
  • How is the brain structured? The recent field of connectomics [2] is developing high-throughput techniques for mapping connections in nervous systems, one of the most important and ambitious goals of neuroanatomy
  • Binary membrane segmentation is obtained by mild postprocessing techniques discussed in Section 2.3, followed by thresholding
  • One is used as ground truth, the other to evaluate the performance of a second human observer and provide a meaningful comparison for the algorithms’ performance
  • The main strength of our approach to neuronal membrane segmentation in electron microscopy (EM) images lies in a deep and wide neural network trained by online back-propagation to become a very powerful pixel classifier with superhuman pixel-error rate, made possible by an optimized GPU implementation more than 50 times faster than equivalent code on standard microprocessors
  • Our approach outperforms all other approaches in the competition, despite not even being tailored to this particular segmentation task
  • The Deep Neural Network (DNN) acts as a generic image classifier, using raw pixel intensities as inputs, without ad-hoc post-processing
Methods
  • The DNN classifier (Section 2.1) computes the probability of a pixel p being of the former class, using as input the raw intensity values of a square window centered on p with an edge of w pixels—w being an odd number to enforce symmetry.
  • To segment a test image, the classifier is applied to all of its pixels, generating a map of membrane probabilities—i.e., a new real-valued image the size of the input image.
  • Binary membrane segmentation is obtained by mild postprocessing techniques discussed in Section 2.3, followed by thresholding
Results
  • All experiments are performed on a computer with a Core i7 950 3.06GHz processor, 24GB of RAM, and four GTX 580 graphics cards.
  • The dataset is composed by two 512 × 512 × 30 stacks, one used for training, one for testing.
  • A manually annotated ground truth segmentation is provided.
  • The organizers obtained two manual segmentations by different expert neuroanatomists.
  • One is used as ground truth, the other to evaluate the performance of a second human observer and provide a meaningful comparison for the algorithms’ performance
Conclusion
  • Discussion and conclusions

    The main strength of the approach to neuronal membrane segmentation in EM images lies in a deep and wide neural network trained by online back-propagation to become a very powerful pixel classifier with superhuman pixel-error rate, made possible by an optimized GPU implementation more than 50 times faster than equivalent code on standard microprocessors.

    The authors' approach outperforms all other approaches in the competition, despite not even being tailored to this particular segmentation task.
  • The main strength of the approach to neuronal membrane segmentation in EM images lies in a deep and wide neural network trained by online back-propagation to become a very powerful pixel classifier with superhuman pixel-error rate, made possible by an optimized GPU implementation more than 50 times faster than equivalent code on standard microprocessors.
  • The DNN acts as a generic image classifier, using raw pixel intensities as inputs, without ad-hoc post-processing
  • This opens interesting perspectives on applying similar techniques to other biomedical image segmentation tasks
Tables
  • Table1: Table 1
  • Table2: Results of our approach and competing algorithms. For comparison, the first two rows report the performance of the second human observer and of a simple thresholding approach
Download tables as Excel
Funding
  • This work was partially supported by the Supervised Deep / Recurrent Nets SNF grant, Project Code 140399
Reference
  • Segmentation of neuronal structures in EM stacks challenge - ISBI 2012. http://tinyurl.com/d2fgh7g.
    Findings
  • The Open Connectome Project. http://openconnectomeproject.org.
    Findings
  • R. Achanta, A. Shaji, K. Smith, A. Lucchi, P. Fua, and S. Susstrunk. Slic superpixels. Technical Report 149300 EPFL, (June), 2010.
    Google ScholarFindings
  • Erhan Bas, Mustafa G. Uzunbas, Dimitris Metaxas, and Eugene Myers. Contextual grouping in a concept: a multistage decision strategy for EM segmentation. In Proc. of ISBI 2012 EM Segmentation Challenge.
    Google ScholarLocate open access versionFindings
  • Sven Behnke. Hierarchical Neural Networks for Image Interpretation, volume 2766 of Lecture Notes in Computer Science. Springer, 2003.
    Google ScholarLocate open access versionFindings
  • Davi D. Bock, Wei-Chung A. Lee, Aaron M. Kerlin, Mark L. Andermann, Greg Hood, Arthur W. Wetzel, Sergey Yurgenson, Edward R. Soucy, Hyon S. Kim, and R. Clay Reid. Network anatomy and in vivo physiology of visual cortical neurons. Nature, 471(7337):177–182, 2011.
    Google ScholarLocate open access versionFindings
  • Y. Boykov, O. Veksler, and R. Zabih. Fast approximate energy minimization via graph cuts. Pattern Analysis and Machine Intelligence, IEEE Transactions on, 23(11):1222–1239, 2001.
    Google ScholarLocate open access versionFindings
  • Radim Burget, Vaclav Uher, and Jan Masek. Trainable Segmentation Based on Local-level and Segmentlevel Feature Extraction. In Proc. of ISBI 2012 EM Segmentation Challenge.
    Google ScholarLocate open access versionFindings
  • Albert Cardona, Stephan Saalfeld, Stephan Preibisch, Benjamin Schmid, Anchi Cheng, Jim Pulokas, Pavel Tomancak, and Volker Hartenstein. An integrated micro- and macroarchitectural analysis of the drosophila brain by computer-assisted serial section electron microscopy. PLoS Biol, 8(10):e1000502, 10 2010.
    Google ScholarLocate open access versionFindings
  • Dan Claudiu Ciresan, Ueli Meier, Luca Maria Gambardella, and Jurgen Schmidhuber. Deep, big, simple neural nets for handwritten digit recognition. Neural Computation, 22(12):3207–3220, 2010.
    Google ScholarLocate open access versionFindings
  • Dan Claudiu Ciresan, Ueli Meier, Luca Maria Gambardella, and Jurgen Schmidhuber. Convolutional neural network committees for handwritten character classification. In International Conference on Document Analysis and Recognition, pages 1250–1254, 2011.
    Google ScholarLocate open access versionFindings
  • Dan Claudiu Ciresan, Ueli Meier, Jonathan Masci, Luca Maria Gambardella, and Jurgen Schmidhuber. Flexible, high performance convolutional neural networks for image classification. In International Joint Conference on Artificial Intelligence, pages 1237–1242, 2011.
    Google ScholarLocate open access versionFindings
  • Dan Claudiu Ciresan, Ueli Meier, and Jurgen Schmidhuber. Multi-column deep neural networks for image classification. In Computer Vision and Pattern Recognition, pages 3642–3649, 2012.
    Google ScholarLocate open access versionFindings
  • C.A. Curcio, K.R. Sloan, R.E. Kalina, and A.E. Hendrickson. Human photoreceptor topography. The Journal of comparative neurology, 292(4):497–523, 1990.
    Google ScholarLocate open access versionFindings
  • A. Foi and G. Boracchi. Foveated self-similarity in nonlocal image filtering. In Proceedings of SPIE, volume 8291, page 829110, 2012.
    Google ScholarLocate open access versionFindings
  • Kunihiko Fukushima. Neocognitron: A self-organizing neural network for a mechanism of pattern recognition unaffected by shift in position. Biological Cybernetics, 36(4):193–202, 1980.
    Google ScholarLocate open access versionFindings
  • G. Gonzalez, F. Fleurety, and P. Fua. Learning rotational features for filament detection. In Computer Vision and Pattern Recognition, 2009. CVPR 2009. IEEE Conference on, pages 1582–1589. IEEE, 2009.
    Google ScholarLocate open access versionFindings
  • Saadia Iftikhar and Afzal Godil. The Detection of Neuronal Structures using a Patch-based Multi-features and Support Vector Machines Learning Algorithm. In Proc. of ISBI 2012 EM Segmentation Challenge.
    Google ScholarLocate open access versionFindings
  • Viren Jain, Benjamin Bollmann, Mark Richardson, Daniel R. Berger, Moritz Helmstaedter, Kevin L. Briggman, Winfried Denk, Jared B. Bowden, John M. Mendenhall, Wickliffe C. Abraham, Kristen M. Harris, N. Kasthuri, Ken J. Hayworth, Richard Schalek, Juan Carlos Tapia, Jeff W. Lichtman, and H. Sebastian Seung. Boundary Learning by Optimization with Topological Constraints. In CVPR, pages 2488–2495. IEEE, 2010.
    Google ScholarLocate open access versionFindings
  • Lee Kamentsky. Segmentation of EM images of neuronal structures using CellProfiler. In Proc. of ISBI 2012 EM Segmentation Challenge.
    Google ScholarLocate open access versionFindings
  • Bobby Kasthuri. Mouse Visual Cortex Dataset in the Open Connectome Project. http://openconnectomeproject.org/Kasthuri11/.
    Findings
  • V. Kaynig, T. Fuchs, and J. Buhmann. Geometrical consistent 3D tracing of neuronal processes in ssTEM data. Medical Image Computing and Computer-Assisted Intervention–MICCAI 2010, pages 209–216, 2010.
    Google ScholarLocate open access versionFindings
  • V. Kaynig, T. Fuchs, and J.M. Buhmann. Neuron geometry extraction by perceptual grouping in sstem images. In Computer Vision and Pattern Recognition (CVPR), 2010 IEEE Conference on, pages 2902– 2909. IEEE, 2010.
    Google ScholarLocate open access versionFindings
  • Dmitry Laptev, Alexander Vezhnevets, Sarvesh Dwivedi, and Joachim Buhmann. Segmentation of Neuronal Structures in EM stacks. In Proc. of ISBI 2012 EM Segmentation Challenge.
    Google ScholarLocate open access versionFindings
  • Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner. Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11):2278–2324, November 1998.
    Google ScholarLocate open access versionFindings
  • Ting Liu, Mojtaba Seyedhosseini, Elizabeth Jurrus, and Tolga Tasdizen. Neuron Segmentation in EM Images using Series of Classifiers and Watershed Tree. In Proc. of ISBI 2012 EM Segmentation Challenge.
    Google ScholarLocate open access versionFindings
  • A. Lucchi, K. Smith, R. Achanta, G. Knott, and P. Fua. Supervoxel-Based Segmentation of Mitochondria in EM Image Stacks With Learned Shape Features. Medical Imaging, IEEE Transactions on, (99):1–1, 2012.
    Google ScholarFindings
  • A. Lucchi, K. Smith, R. Achanta, V. Lepetit, and P. Fua. A fully automated approach to segmentation of irregularly shaped cellular structures in EM images. Medical Image Computing and Computer-Assisted Intervention–MICCAI 2010, pages 463–471, 2010.
    Google ScholarLocate open access versionFindings
  • W.M. Rand. Objective criteria for the evaluation of clustering methods. Journal of the American Statistical association, 66(336):846–850, 1971.
    Google ScholarLocate open access versionFindings
  • Maximiliam Riesenhuber and Tomaso Poggio. Hierarchical models of object recognition in cortex. Nat. Neurosci., 2(11):1019–1025, 1999.
    Google ScholarLocate open access versionFindings
  • Dominik Scherer, Adreas Muller, and Sven Behnke. Evaluation of pooling operations in convolutional architectures for object recognition. In International Conference on Artificial Neural Networks, 2010.
    Google ScholarLocate open access versionFindings
  • Thomas Serre, Lior Wolf, and Tomaso Poggio. Object recognition with features inspired by visual cortex. In Proc. of Computer Vision and Pattern Recognition Conference, 2005.
    Google ScholarLocate open access versionFindings
  • Patrice Y. Simard, Dave. Steinkraus, and John C. Platt. Best practices for convolutional neural networks applied to visual document analysis. In Seventh International Conference on Document Analysis and Recognition, pages 958–963, 2003.
    Google ScholarLocate open access versionFindings
  • K. Smith, A. Carleton, and V. Lepetit. Fast ray features for learning irregular shapes. In Computer Vision, 2009 IEEE 12th International Conference on, pages 397–404. IEEE, 2009.
    Google ScholarLocate open access versionFindings
  • Daniel Strigl, Klaus Kofler, and Stefan Podlipnig. Performance and scalability of GPU-based convolutional neural networks. In 18th Euromicro Conference on Parallel, Distributed, and Network-Based Processing, 2010.
    Google ScholarLocate open access versionFindings
  • Xiao Tan and Changming Sun. Membrane extraction using two-step classification and post-processing. In Proc. of ISBI 2012 EM Segmentation Challenge.
    Google ScholarLocate open access versionFindings
Your rating :
0

 

Tags
Comments
数据免责声明
页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果,我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问,可以通过电子邮件方式联系我们:report@aminer.cn
小科