Detecting Synapse Location and Connectivity by Signed Proximity Estimation and Pruning with Deep Nets

Synaptic connectivity detection is a critical task for neural reconstruction from Electron Microscopy (EM) data. Most of the existing algorithms for synapse detection do not identify the cleft location and direction of connectivity simultaneously. The few methods that computes direction along with contact location have only been demonstrated to work on either dyadic (most common in vertebrate brain) or polyadic (found in fruit fly brain) synapses, but not on both types. In this paper, we present an algorithm to automatically predict the location as well as the direction of both dyadic and polyadic synapses. The proposed algorithm first generates candidate synaptic connections from voxelwise predictions of signed proximity generated by a 3D U-net. A second 3D CNN then prunes the set of candidates to produce the final detection of cleft and connectivity orientation. Experimental results demonstrate that the proposed method outperforms the existing methods for determining synapses in both rodent and fruit fly brain. (Code at: https://github.com/paragt/EMSynConn).

[1]  Louis K. Scheffer,et al.  Fully-Automatic Synapse Prediction and Validation on a Large Data Set , 2016, Front. Neural Circuits.

[2]  Joel H. Saltz,et al.  ConvNets with Smooth Adaptive Activation Functions for Regression , 2017, AISTATS.

[3]  Gary B. Huang,et al.  Identifying Synapses Using Deep and Wide Multiscale Recursive Networks , 2014, ArXiv.

[4]  Fred A. Hamprecht,et al.  Who Is Talking to Whom: Synaptic Partner Detection in Anisotropic Volumes of Insect Brain , 2015, MICCAI.

[5]  Hossein Mobahi,et al.  Training Recurrent Neural Networks by Diffusion , 2016, ArXiv.

[6]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[7]  Fred A. Hamprecht,et al.  Automated Detection and Segmentation of Synaptic Contacts in Nearly Isotropic Serial Electron Microscopy Images , 2011, PloS one.

[8]  Louis K. Scheffer,et al.  Small Sample Learning of Superpixel Classifiers for EM Segmentation , 2014, MICCAI.

[9]  G. Urban,et al.  Automated synaptic connectivity inference for volume electron microscopy , 2017, Nature Methods.

[10]  Ting Liu,et al.  SSHMT: Semi-supervised Hierarchical Merge Tree for Electron Microscopy Image Segmentation , 2016, ECCV.

[11]  B. S. Manjunath,et al.  Synapse classification and localization in Electron Micrographs , 2014, Pattern Recognit. Lett..

[12]  Ross B. Girshick,et al.  Mask R-CNN , 2017, 1703.06870.

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

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

[15]  Moritz Helmstaedter,et al.  The Mutual Inspirations of Machine Learning and Neuroscience , 2015, Neuron.

[16]  Chandan Singh,et al.  A Deep Structured Learning Approach Towards Automating Connectome Reconstruction from 3D Electron Micrographs , 2017, ArXiv.

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

[18]  Kevin L. Briggman,et al.  Structural neurobiology: missing link to a mechanistic understanding of neural computation , 2012, Nature Reviews Neuroscience.

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

[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]  Jeff W Lichtman,et al.  Why not connectomics? , 2013, Nature Methods.

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

[23]  Pascal Fua,et al.  Learning Context Cues for Synapse Segmentation , 2013, IEEE Transactions on Medical Imaging.

[24]  Srinivas C. Turaga,et al.  Machines that learn to segment images: a crucial technology for connectomics , 2010, Current Opinion in Neurobiology.

[25]  Fred A. Hamprecht,et al.  Automated Detection of Synapses in Serial Section Transmission Electron Microscopy Image Stacks , 2014, PloS one.

[26]  Fred A Hamprecht,et al.  Multicut brings automated neurite segmentation closer to human performance , 2017, Nature Methods.

[27]  Patrick van der Smagt,et al.  SynEM, automated synapse detection for connectomics , 2017, eLife.

[28]  H. Sebastian Seung,et al.  Superhuman Accuracy on the SNEMI3D Connectomics Challenge , 2017, ArXiv.

[29]  Toufiq Parag,et al.  Annotating Synapses in Large EM Datasets , 2014, ArXiv.

[30]  Hanspeter Pfister,et al.  Anisotropic EM Segmentation by 3D Affinity Learning and Agglomeration , 2017, ArXiv.