Skeleton-based image registration of serial electron microscopy sections

2019 ◽  
Author(s):  
Xi Chen ◽  
Lijun Shen ◽  
Qiwei Xie ◽  
Hua Han
Keyword(s):  
Electron Microscopy ◽  
Image Registration ◽  
Serial Electron Microscopy

scholarly journals Dual Networks for High-Precision and High-Speed Registration of Brain Electron Microscopy Images

2020 ◽  
Vol 10 (2) ◽  
pp. 86
Author(s):  
Chang Shu ◽  
Tong Xin ◽  
Fangxu Zhou ◽  
Xi Chen ◽  
Hua Han
Keyword(s):  
Electron Microscopy ◽  
Image Registration ◽  
Network Architecture ◽  
High Speed ◽  
Deformation Field ◽  
Volume Reconstruction ◽  
Microscopy Image ◽  
Complex Models ◽  
Field Estimation ◽  
Microscopy Images

It remains a mystery as to how neurons are connected and thereby enable use to think, and volume reconstruction from series of microscopy sections of brains is a vital technique in determining this connectivity. Image registration is a key component; the aim of image registration is to estimate the deformation field between two images. Current methods choose to directly regress the deformation field; however, this task is very challenging. It is common to trade off computational complexity with precision when designing complex models for deformation field estimation. This approach is very inefficient, leading to a long inference time. In this paper, we suggest that complex models are not necessary and solve this dilemma by proposing a dual-network architecture. We divide the deformation field prediction problem into two relatively simple subproblems and solve each of them on one branch of the proposed dual network. The two subproblems have completely opposite properties, and we fully utilize these properties to simplify the design of the dual network. These simple architectures enable high-speed image registration. The two branches are able to work together and make up for each other’s drawbacks, and no loss of accuracy occurs even when simple architectures are involved. Furthermore, we introduce a series of loss functions to enable the joint training of the two networks in an unsupervised manner without introducing costly manual annotations. The experimental results reveal that our method outperforms state-of-the-art methods in fly brain electron microscopy image registration tasks, and further ablation studies enable us to obtain a comprehensive understanding of each component of our network.

scholarly journals Automated sub-5 nm image registration in integrated correlative fluorescence and electron microscopy using cathodoluminescence pointers

2017 ◽  
Vol 7 (1) ◽  
Author(s):  
Martijn T. Haring ◽  
Nalan Liv ◽  
A. Christiaan Zonnevylle ◽  
Angela C. Narvaez ◽  
Lenard M. Voortman ◽  
...  
Keyword(s):  
Electron Microscopy ◽  
Image Registration ◽  
Fluorescence And Electron Microscopy

scholarly journals The rich get richer: synaptic remodeling between climbing fibers and Purkinje cells in the developing cerebellum begins with positive feedback addition of synapses

2019 ◽  
Author(s):  
Alyssa Michelle Wilson ◽  
Richard Schalek ◽  
Adi Suissa-Peleg ◽  
Thouis Ray Jones ◽  
Seymour Knowles-Barley ◽  
...  
Keyword(s):  
Electron Microscopy ◽  
Purkinje Cell ◽  
Postnatal Development ◽  
Purkinje Cells ◽  
Positive Feedback ◽  
Climbing Fiber ◽  
Climbing Fibers ◽  
List Type ◽  
Developing Cerebellum ◽  
Serial Electron Microscopy

SUMMARYDuring postnatal development, cerebellar climbing fibers strongly innervate a subset of their original Purkinje cell targets and eliminate their connections from the rest. In the adult, each climbing fiber innervates a small number of Purkinje cells and each Purkinje cell is innervated by a single climbing fiber. To get insight about the processes responsible for this remapping, we reconstructed serial electron microscopy datasets from mice during the first postnatal week. In contrast to adult connectivity, individual neonatal climbing fibers innervate many nearby Purkinje cells, and multiple climbing fibers innervate each Purkinje cell. Between postnatal days 3 and 7, Purkinje cells retract long dendrites and grow many proximal dendritic processes. On this changing landscape, individual climbing fibers selectively add many synapses to a subset of Purkinje cell targets in a positive-feedback manner, without pruning synapses from other Purkinje cells. The active zone sizes of synapses associated with powerful versus weak inputs are indistinguishable. These results show that changes in synapse number rather than synapse size are the predominant form of early developmental plasticity. Finally, although multiple climbing fibers innervate each Purkinje cell in early postnatal development, the number of climbing fibers and Purkinje cells in a local cerebellar region nearly match. Thus, initial over-innervation of Purkinje cells by climbing fibers is economical, in that the number of axons entering a region is enough to assure that each axon ends up with a postsynaptic target, and that none branched there in vain.HIGHLIGHTSDeveloping climbing fibers establish synapses on many neighboring Purkinje cells unlike the sparse pattern of innervation in later lifeClimbing fibers add many synapses onto a few of their Purkinje targets before the pruning stage in a rich-get-richer type processThe synapse sizes of strengthened and weakened climbing fiber inputs are indistinguishable.Exuberant branching of climbing fiber axons in early postnatal life appears to be economical because the numbers of axons and Purkinje cells in a local region match, ensuring that each axon can establish a long-lasting connection thereBLURBHigh-resolution serial electron microscopy reconstructions reveal that climbing fiber-Purkinje cell synaptic refinement in the developing cerebellum begins with significant synapse addition. Climbing fibers focus their synapses onto a smaller number of Purkinje cells by selectively adding synapses onto some target cells. All axons that project to a region in development play a role in the final connectivity.

Sign in / Sign up

Export Citation Format

Share Document