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

Stable and Accurate Orbital-Free Density Functional Theory Powered by Machine Learning.

Journal of the American Chemical Society·2025
Same author

From spots to cells: Cell segmentation in spatial transcriptomics with BOMS.

PloS one·2025
Same author

Deep learning image analysis for continuous single-cell imaging of dynamic processes in Plasmodium falciparum-infected erythrocytes.

Communications biology·2025
Same author

Structural dynamics of human ribosomes in situ reconstructed by exhaustive high-resolution template matching.

Molecular cell·2024
Same author

A deep learning-based toolkit for 3D nuclei segmentation and quantitative analysis in cellular and tissue context.

Development (Cambridge, England)·2024
Same author

Visualizing single-cell data with the neighbor embedding spectrum.

bioRxiv : the preprint server for biology·2024

Related Experiment Video

Updated: May 2, 2026

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations
12:27

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations

Published on: February 15, 2017

6.3K

Learning to segment neurons with non-local quality measures.

Thorben Kroeger1, Shawn Mikula2, Winfried Denk2

  • 1HCI, University of Heidelberg.

Medical Image Computing and Computer-Assisted Intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
|March 1, 2014
PubMed
Summary

This study introduces structured learning for image segmentation, optimizing boundary detection to prevent catastrophic errors. The new hierarchical scheme improves segmentation quality in large-scale neural circuit reconstruction.

More Related Videos

Author Spotlight: Unveiling the Molecular Basis of Pain Perception and Neuropathic Pain
05:28

Author Spotlight: Unveiling the Molecular Basis of Pain Perception and Neuropathic Pain

Published on: August 9, 2024

1.9K
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

3.6K

Related Experiment Videos

Last Updated: May 2, 2026

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations
12:27

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations

Published on: February 15, 2017

6.3K
Author Spotlight: Unveiling the Molecular Basis of Pain Perception and Neuropathic Pain
05:28

Author Spotlight: Unveiling the Molecular Basis of Pain Perception and Neuropathic Pain

Published on: August 9, 2024

1.9K
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

3.6K

Area of Science:

  • Computer Vision
  • Machine Learning
  • Neuroscience Imaging

Background:

  • Segmentation methods like hierarchical region merging and correlation clustering depend on accurate edge weights between supervoxels.
  • Current unstructured learning methods focus on individual edge classification, overlooking how local errors can lead to global segmentation failures.
  • Optimizing boundary evidence learning requires structured quality criteria, such as Rand Error or Variation of Information, to ensure robust segmentation.

Purpose of the Study:

  • To develop a novel structured learning scheme for image segmentation that directly optimizes global quality criteria.
  • To introduce a new hierarchical scheme capable of approximating solutions to NP-hard prediction problems in large-volume images.
  • To enhance the quality of segmentation in challenging neural circuit reconstruction tasks using electron microscopic images.

Main Methods:

  • Implementation of the first structured learning scheme incorporating a structured loss function for segmentation.
  • Development of a new hierarchical scheme designed to handle NP-hard prediction problems in massive image datasets.
  • Application and validation of the proposed methods on serial sectioning electron microscopic images for neural circuit reconstruction.

Main Results:

  • The proposed structured learning scheme significantly improves segmentation quality compared to existing methods.
  • The new hierarchical scheme effectively addresses the computational complexity of segmenting large-volume images.
  • Demonstrated superior partitioning quality in two challenging neural circuit reconstruction problems.

Conclusions:

  • Structured learning with a structured loss function is crucial for accurate image segmentation, especially in complex biological datasets.
  • The novel hierarchical scheme provides an efficient approach for large-scale segmentation tasks, overcoming limitations of previous methods.
  • This work advances the state-of-the-art in neural circuit reconstruction by improving segmentation accuracy and reliability.