Voxel-wise Cross-Volume Representation Learning for 3D Neuron Reconstruction

@inproceedings{Wang2021VoxelwiseCR,
  title={Voxel-wise Cross-Volume Representation Learning for 3D Neuron Reconstruction},
  author={Heng Wang and Chaoyi Zhang and Jianhui Yu and Yang Song and Siqi Liu and Wojciech Chrzanowski and Weidong (Tom) Cai},
  booktitle={MLMI@MICCAI},
  year={2021}
}
Automatic 3D neuron reconstruction is critical for analysing the morphology and functionality of neurons in brain circuit activities. However, the performance of existing tracing algorithms is hinged by the low image quality. Recently, a series of deep learning based segmentation methods have been proposed to improve the quality of raw 3D optical image stacks by removing noises and restoring neuronal structures from low-contrast background. Due to the variety of neuron morphology and the lack… Expand

Figures and Tables from this paper

References

SHOWING 1-10 OF 29 REFERENCES
Multiscale Kernels for Enhanced U-Shaped Network to Improve 3D Neuron Tracing
TLDR
This paper proposes a new framework to overcome issues by performing neuron segmentation prior to tracing by adopting a novel 3D U-shaped convolutional neural network with multiscale kernel fusion and spatial fusion to perform the image segmentation. Expand
Single Neuron Segmentation Using Graph-Based Global Reasoning with Auxiliary Skeleton Loss from 3D Optical Microscope Images
  • Heng Wang, Yang Song, +4 authors Weidong (Tom) Cai
  • Computer Science, Engineering
  • 2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI)
  • 2021
TLDR
This work presents an end-to-end segmentation network by jointly considering the local appearance and the global geometry traits through graph reasoning and a skeleton-based auxiliary loss and demonstrates that the proposed method exceeds the counterpart algorithms in performance. Expand
3D Neuron Reconstruction in Tangled Neuronal Image With Deep Networks
  • Qiufu Li, L. Shen
  • Medicine, Computer Science
  • IEEE Transactions on Medical Imaging
  • 2020
TLDR
Experimental results show that the proposed 3D U-Net Plus network achieved very promising segmentation results, and the neurons reconstructed by the tracing algorithm using the segmentation result match significantly better with the ground truth than that using the original images. Expand
Progressive Learning for Neuronal Population Reconstruction from Optical Microscopy Images
TLDR
A progressive learning strategy to take advantages of both traditional neuron tracing methods and deep learning techniques to reconstruct more complete neuronal populations without using manual annotations is proposed. Expand
Deep Learning Segmentation of Optical Microscopy Images Improves 3-D Neuron Reconstruction
TLDR
A novel CNN architecture is designed, that takes volumetric images as the inputs and their voxel-wise segmentation maps as the outputs, that allows it to train and predict using large microscopy images in an end-to-end manner. Expand
3D Conditional Adversarial Learning for Synthesizing Microscopic Neuron Image Using Skeleton-to-Neuron Translation
TLDR
This work proposes a framework for synthesizing microscopy-realistic 3D neuron images from simulated single neuron skeletons using conditional Generative Adversarial Networks (cGAN) and builds the generator network with multi-resolution sub-modules to improve the output fidelity. Expand
Segmenting Neuronal Structure in 3D Optical Microscope Images via Knowledge Distillation with Teacher-Student Network
TLDR
This paper proposes a teacher-student learning framework for fast neuron segmentation that achieves competitive performance for segmentation accuracy while reducing the computational cost to facilitate large-scale processing. Expand
Memory and Time Efficient 3D Neuron Morphology Tracing in Large-Scale Images
TLDR
A novel Memory (and Time) Efficient Image Tracing (MEIT) framework is introduced and evaluated on the Gold dataset, the proposed method achieves better or competitive performance compared to state-of-the-art neuron tracing methods in most cases while requiring less memory and time. Expand
3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation
TLDR
The proposed network extends the previous u-net architecture from Ronneberger et al. by replacing all 2D operations with their 3D counterparts and performs on-the-fly elastic deformations for efficient data augmentation during training. Expand
Rivulet: 3D Neuron Morphology Tracing with Iterative Back-Tracking
TLDR
The proposed Rivulet algorithm, based on the multi-stencils fast-marching and iterative back-tracking, is capable of tracing discontinuous areas without being interrupted by densely distributed noises. Expand
...
1
2
3
...