Dense cellular segmentation for EM using 2D–3D neural network ensembles (2020-01-06T00:00:00.000000Z)

TL;DR

This work presents an algorithm using novel hybrid 2D–3D segmentation networks to produce dense cellular segmentations with accuracy levels that outperform baseline methods and approach those of human annotators, the first published approach to automating the creation of cell models with this level of structural detail.

Abstract

Biologists who use electron microscopy (EM) images to build nanoscale 3D models of whole cells and their organelles have historically been limited to small numbers of cells and cellular features due to constraints in imaging and analysis. This has been a major factor limiting insight into the complex variability of cellular environments. Modern EM can produce gigavoxel image volumes containing large numbers of cells, but accurate manual segmentation of image features is slow and limits the creation of cell models. Segmentation algorithms based on convolutional neural networks can process large volumes quickly, but achieving EM task accuracy goals often challenges current techniques. Here, we define dense cellular segmentation as a multiclass semantic segmentation task for modeling cells and large numbers of their organelles, and give an example in human blood platelets. We present an algorithm using novel hybrid 2D–3D segmentation networks to produce dense cellular segmentations with accuracy levels that outperform baseline methods and approach those of human annotators. To our knowledge, this work represents the first published approach to automating the creation of cell models with this level of structural detail.

Authors

M. Guay

1 papers

Z. Emam

1 papers

Adam Anderson

1 papers

TL;DR

Abstract

Authors

References36 items

3D ultrastructural analysis of α‐granule, dense granule, mitochondria, and canalicular system arrangement in resting human platelets

Automated Reconstruction of a Serial-Section EM Drosophila Brain with Flood-Filling Networks and Local Realignment

Med3D: Transfer Learning for 3D Medical Image Analysis

Deep learning in medical imaging and radiation therapy.

CDeep3M—Plug-and-Play cloud-based deep learning for image segmentation

Designing deep neural networks to automate segmentation for serial block-face electron microscopy

An application of cascaded 3D fully convolutional networks for medical image segmentation

Deep convolutional neural networks for segmenting 3D in vivo multiphoton images of vasculature in Alzheimer disease mouse models

Panoptic Segmentation

A 3D Coarse-to-Fine Framework for Volumetric Medical Image Segmentation

2D-3D Fully Convolutional Neural Networks for Cardiac MR Segmentation

Rethinking Atrous Convolution for Semantic Image Segmentation

VoxResNet: Deep voxelwise residual networks for brain segmentation from 3D MR images

Auto-Context Convolutional Neural Network (Auto-Net) for Brain Extraction in Magnetic Resonance Imaging

DeepVessel: Retinal Vessel Segmentation via Deep Learning and Conditional Random Field

Combining Fully Convolutional and Recurrent Neural Networks for 3D Biomedical Image Segmentation

3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation

V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation

DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs

Guest Editorial Deep Learning in Medical Imaging: Overview and Future Promise of an Exciting New Technique

STEM tomography reveals that the canalicular system and α‐granules remain separate compartments during early secretion stages in blood platelets

Advances in Neural Information Processing Systems 28

Deep Residual Learning for Image Recognition

SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation

Recursive Training of 2D-3D Convolutional Networks for Neuronal Boundary Detection

You Only Look Once: Unified, Real-Time Object Detection

U-Net: Convolutional Networks for Biomedical Image Segmentation

Fully convolutional networks for semantic segmentation

ImageNet classification with deep convolutional neural networks

Serial Block-Face Scanning Electron Microscopy to Reconstruct Three-Dimensional Tissue Nanostructure

Reconstruction of three-dimensional chemistry and geometry using focused ion beam microscopy

Pop M

Deep Learning in Medical Imaging: Overview and Future Promise of an Exciting New Technique

Recursive Training of 2D-3D Convolutional Networks for Neuronal Boundary Prediction

amira: A Highly Interactive System for Visual Data Analysis

Neural Network Ensembles, Cross Validation, and Active Learning

Field of Study

Journal Information

Name

Volume

Venue Information

Name

Type

URL

Alternate Names