FusionNet: A deep fully residual convolutional neural network for image segmentation in connectomics

Abstract

Electron microscopic connectomics is an ambitious research direction with the goal of studying comprehensive brain connectivity maps by using high-throughput, nano-scale microscopy. One of the main challenges in connectomics research is developing scalable image analysis algorithms that require minimal user intervention. Recently, deep learning has drawn much attention in computer vision because of its exceptional performance in image classification tasks. For this reason, its application to connectomic analyses holds great promise, as well. In this paper, we introduce a novel deep neural network architecture, FusionNet, for the automatic segmentation of neuronal structures in connectomics data. FusionNet leverages the latest advances in machine learning, such as semantic segmentation and residual neural networks, with the novel introduction of summation-based skip connections to allow a much deeper network architecture for a more accurate segmentation. We demonstrate the performance of the proposed method by comparing it with state-of-the-art electron microscopy (EM) segmentation methods from the ISBI EM segmentation challenge. We also show the segmentation results on two different tasks including cell membrane and cell body segmentation and a statistical analysis of cell morphology.

References in corpus (9)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Fully Convolutional Networks for Semantic Segmentation
• Going Deeper with Convolutions
• Compressed Sensing MRI Reconstruction using a Generative Adversarial Network with a Cyclic Loss
• FusionNet: A deep fully residual convolutional neural network for image segmentation in connectomics
• Parallel Multi-Dimensional LSTM, With Application to Fast Biomedical Volumetric Image Segmentation
• The Mutex Watershed and its Objective: Efficient, Parameter-Free Graph Partitioning
• Registering large volume serial-section electron microscopy image sets for neural circuit reconstruction using FFT signal whitening
• An Iterative Convolutional Neural Network Algorithm Improves Electron Microscopy Image Segmentation

Cited by in corpus (32)

• Pose Guided Person Image Generation
• FusionNet: A deep fully residual convolutional neural network for image segmentation in connectomics
• Constrained-CNN losses for weakly supervised segmentation
• Superhuman Accuracy on the SNEMI3D Connectomics Challenge
• Deep Learning Techniques for Future Intelligent Cross-Media Retrieval
• RedNet: Residual Encoder-Decoder Network for indoor RGB-D Semantic Segmentation
• Segmentation in large-scale cellular electron microscopy with deep learning: A literature survey
• Neural Stain-Style Transfer Learning using GAN for Histopathological Images
• The Mutex Watershed and its Objective: Efficient, Parameter-Free Graph Partitioning
• A Comprehensive Review of Modern Object Segmentation Approaches
• Rapid tissue oxygenation mapping from snapshot structured-light images with adversarial deep learning
• Domain Adaptive Transfer Attack (DATA)-based Segmentation Networks for Building Extraction from Aerial Images
• Quantitative Phase Imaging and Artificial Intelligence: A Review
• AUNet: Attention-guided dense-upsampling networks for breast mass segmentation in whole mammograms
• Deep-Learning Ensembles for Skin-Lesion Segmentation, Analysis, Classification: RECOD Titans at ISIC Challenge 2018
• Reconstructing the exit wave in high-resolution transmission electron microscopy using machine learning
• Learning Steerable Filters for Rotation Equivariant CNNs
• KuroNet: Pre-Modern Japanese Kuzushiji Character Recognition with Deep Learning
• Learnable Gabor modulated complex-valued networks for orientation robustness
• ACE-Net: Biomedical Image Segmentation with Augmented Contracting and Expansive Paths
• Propagated Perturbation of Adversarial Attack for well-known CNNs: Empirical Study and its Explanation
• YOLO Ensemble for UAV-based Multispectral Defect Detection in Wind Turbine Components
• Morphomics via Next-generation Electron Microscopy
• Deep Morphing: Detecting bone structures in fluoroscopic X-ray images with prior knowledge
• Global Pixel Transformers for Virtual Staining of Microscopy Images
• Reinforced Coloring for End-to-End Instance Segmentation
• PZnet: Efficient 3D ConvNet Inference on Manycore CPUs
• Multi-stage Multi-recursive-input Fully Convolutional Networks for Neuronal Boundary Detection
• Error Control and Loss Functions for the Deep Learning Inversion of Borehole Resistivity Measurements
• Convolutional nets for reconstructing neural circuits from brain images acquired by serial section electron microscopy
• A novel Deep Structure U-Net for Sea-Land Segmentation in Remote Sensing Images
• A Rigid Registration Method in TEVAR