Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

A framework for comparative analysis of human and mouse cortical neuron dendrites in corresponding brain regions.

Nature neuroscience·2026
Same author

Modeling 3D mesoscaled neuronal complexity through learning-based dynamic morphometric convolution.

Brain informatics·2026
Same author

AISleep: Automated and interpretable sleep staging from single-channel EEG data.

Patterns (New York, N.Y.)·2025
Same author

Bridging the dimensional gap from planar spatial transcriptomics to 3D cell atlases.

Nature methods·2025
Same author

A mouse brain atlas based on dendritic microenvironments.

Nature neuroscience·2025
Same author

Author Correction: Confocal Airy beam oblique light-sheet tomography for brain-wide cell type distribution and morphology.

Nature methods·2025

Related Experiment Video

Updated: Mar 21, 2026

Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

11.2K

Deep models for brain EM image segmentation: novel insights and improved performance.

Ahmed Fakhry1, Hanchuan Peng2, Shuiwang Ji3

  • 1Department of Computer Science, Old Dominion University, Norfolk, VA 23529, USA.

Bioinformatics (Oxford, England)
|May 7, 2016
PubMed
Summary

This study introduces a novel deep neural network (DNN) architecture optimized for segmenting brain electron microscopy (EM) images, crucial for dense circuit reconstruction. The new model achieved top performance in a challenge, advancing EM image analysis.

More Related Videos

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

3.7K
A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells
12:49

A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells

Published on: September 28, 2019

13.5K

Related Experiment Videos

Last Updated: Mar 21, 2026

Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

11.2K
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

3.7K
A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells
12:49

A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells

Published on: September 28, 2019

13.5K

Area of Science:

  • Neuroscience
  • Computer Vision
  • Biomedical Imaging

Background:

  • Accurate segmentation of brain electron microscopy (EM) images is essential for reconstructing neural circuits.
  • Existing deep neural network (DNN) models effective for image classification are not directly applicable to EM image segmentation due to task differences.

Purpose of the Study:

  • To develop an optimized DNN architecture specifically tailored for EM image segmentation.
  • To leverage the full potential of DNNs for precise segmentation of neural structures in EM data.

Main Methods:

  • Proposed a novel DNN design for EM image segmentation.
  • Trained a pixel classifier directly on raw pixel intensities without preprocessing.
  • Generated probability maps indicating membrane likelihood for each pixel.

Main Results:

  • Achieved superior performance on EM image segmentation tasks.
  • Consistently secured the best results across all three evaluation metrics in the 2D EM Image Segmentation Challenge.
  • Demonstrated the effectiveness of the novel DNN architecture for detailed neural circuit reconstruction.

Conclusions:

  • The developed DNN architecture offers a significant advancement in EM image segmentation.
  • The approach provides a robust method for accurate dense circuit reconstruction.
  • The findings highlight the potential of tailored DNNs for specialized image analysis in neuroscience.