Abstract
Benefiting from the rapid development of electron microscopy imaging and deep learning technologies, an increasing number of brain image datasets with segmentation and synapse detection are published. Most of the automated segmentation methods label voxels rather than producing neuron skeletons directly. A further skeletonization step is necessary for quantitative morphological analysis. Currently, several tools are published for skeletonization as well as morphological and synaptic connectivity analysis using different computer languages and environments. Recently the Julia programming language, notable for elegant syntax and high performance, has gained rapid adoption in the scientific computing community. Here, we present a Julia package, called RealNeuralNetworks.jl, for efficient sparse skeletonization, morphological analysis, and synaptic connectivity analysis. Based on a large-scale Zebrafish segmentation dataset, we illustrate the software features by performing distributed skeletonization in Google Cloud, clustering the neurons using the NBLAST algorithm, combining morphological similarity and synaptic connectivity to study their relationship. We demonstrate that RealNeuralNetworks.jl is suitable for use in terabyte-scale electron microscopy image segmentation datasets.
Highlights
We demonstrate the usage of RealNeuralNetworks.jl by analyzing a dataset with some proofread neurons
BigArrays.jl was designed for general usage and could be used to handle arrays that are too large to fit in RAM
A potential application is solving the out-of-memory issue in the simulation of quantum computing using tensor networks (Fishman et al, 2020) (Personal Communication)
Summary
Neural morphology and synaptic connectivity are closely related to brain function. With both nanometer resolution and a large field of view, advanced Electron Microscopes can produce largescale image stacks (Kornfeld and Denk, 2018; Yin et al, 2020). With additional help from proofreading (Kim et al, 2014; Zhao et al, 2018; Dorkenwald et al, 2020; Hubbard et al, 2020), reconstructed neurons with synaptic connectivity can be used for scientific discovery (Deutsch et al, 2020; Januszewski et al, 2020; Vishwanathan et al, 2020). Neurons are like trees and their skeletons can be used for morphological analysis. In contrast to manual tracing and getting a neuron skeleton directly, most existing automated segmentation methods produce voxel labeling and are skeletonized in another step
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
