Morphological Error Detection in 3D Segmentations

Deep learning algorithms for connectomics rely upon localized classification, rather than overall morphology. This leads to a high incidence of erroneously merged objects. Humans, by contrast, can easily detect such errors by acquiring intuition for the correct morphology of objects. Biological neurons have complicated and variable shapes, which are challenging to learn, and merge errors take a multitude of different forms. We present an algorithm, MergeNet, that shows 3D ConvNets can, in fact, detect merge errors from high-level neuronal morphology. MergeNet follows unsupervised training and operates across datasets. We demonstrate the performance of MergeNet both on a variety of connectomics data and on a dataset created from merged MNIST images.

[1]  Nir Shavit,et al.  The big data challenges of connectomics , 2014, Nature Neuroscience.

[2]  John P. Cunningham,et al.  Automated scalable segmentation of neurons from multispectral images , 2016, NIPS.

[3]  Joan Bruna,et al.  Training Convolutional Networks with Noisy Labels , 2014, ICLR 2014.

[4]  Won-Ki Jeong,et al.  Whole-brain serial-section electron microscopy in larval zebrafish , 2017, Nature.

[5]  Max Welling,et al.  Semi-Supervised Classification with Graph Convolutional Networks , 2016, ICLR.

[6]  Srinivas C. Turaga,et al.  Connectomic reconstruction of the inner plexiform layer in the mouse retina , 2013, Nature.

[7]  Gregory D. Hager,et al.  VESICLE: Volumetric Evaluation of Synaptic Inferfaces using Computer Vision at Large Scale , 2014, BMVC.

[8]  H. Sebastian Seung,et al.  Automatic Neuron Detection in Calcium Imaging Data Using Convolutional Networks , 2016, NIPS.

[9]  H. Sebastian Seung,et al.  Scalable training of 3D convolutional networks on multi- and many-cores , 2017, J. Parallel Distributed Comput..

[10]  Hayk Saribekyan,et al.  A Multicore Path to Connectomics-on-Demand , 2017, PPoPP 2017.

[11]  J. Chang,et al.  Expansion microscopy , 2018, Journal of microscopy.

[12]  William R. Gray Roncal,et al.  Saturated Reconstruction of a Volume of Neocortex , 2015, Cell.

[13]  Yoshua Bengio,et al.  Towards Biologically Plausible Deep Learning , 2015, ArXiv.

[14]  Konrad P. Körding,et al.  Toward an Integration of Deep Learning and Neuroscience , 2016, bioRxiv.

[15]  Daniel R. Berger,et al.  The Fuzzy Logic of Network Connectivity in Mouse Visual Thalamus , 2016, Cell.

[16]  Jianxiong Xiao,et al.  3D ShapeNets: A deep representation for volumetric shapes , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[17]  Hayk Saribekyan,et al.  A Multi-Pass Approach to Large-Scale Connectomics , 2016, ArXiv.

[18]  Stephen M. Plaza,et al.  Large-Scale Electron Microscopy Image Segmentation in Spark , 2016, ArXiv.

[19]  Martín Abadi,et al.  TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems , 2016, ArXiv.

[20]  Alessandro Giusti,et al.  Efficient Classifier Training to Minimize False Merges in Electron Microscopy Segmentation , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[21]  Anthony M. Zador,et al.  Using high-throughput barcode sequencing to efficiently map connectomes , 2017, bioRxiv.

[22]  Pieter Abbeel,et al.  Combinatorial Energy Learning for Image Segmentation , 2015, NIPS.

[23]  H. Sebastian Seung,et al.  Recursive Training of 2D-3D Convolutional Networks for Neuronal Boundary Prediction , 2015, NIPS.

[24]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[25]  Nir Shavit,et al.  Deep Tensor Convolution on Multicores , 2016, ICML.

[26]  Louis K. Scheffer,et al.  A visual motion detection circuit suggested by Drosophila connectomics , 2013, Nature.

[27]  Ting Zhao,et al.  Automatic Neuron Type Identification by Neurite Localization in the Drosophila Medulla , 2014, ArXiv.

[28]  Subhransu Maji,et al.  Multi-view Convolutional Neural Networks for 3D Shape Recognition , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[29]  G. Knott,et al.  Serial Section Scanning Electron Microscopy of Adult Brain Tissue Using Focused Ion Beam Milling , 2008, The Journal of Neuroscience.

[30]  H. Sebastian Seung,et al.  Learning to Agglomerate Superpixel Hierarchies , 2011, NIPS.

[31]  Peter Li,et al.  Flood-Filling Networks , 2016, ArXiv.

[32]  Mathias Niepert,et al.  Learning Convolutional Neural Networks for Graphs , 2016, ICML.

[33]  Sebastian Scherer,et al.  VoxNet: A 3D Convolutional Neural Network for real-time object recognition , 2015, 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[34]  Lin Yang,et al.  Combining Fully Convolutional and Recurrent Neural Networks for 3D Biomedical Image Segmentation , 2016, NIPS.

[35]  Anirban Chakraborty,et al.  A Context-Aware Delayed Agglomeration Framework for Electron Microscopy Segmentation , 2014, PloS one.

[36]  Hayk Saribekyan,et al.  Toward Streaming Synapse Detection with Compositional ConvNets , 2017, ArXiv.