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 Experiment Videos

Performance-based classifier combination in atlas-based image segmentation using expectation-maximization parameter

Torsten Rohlfing1, Daniel B Russakoff, Calvin R Maurer

  • 1Image Guidance Laboratories, Department of Neurosurgery, Stanford University, Stanford, CA 94305-5327, USA. rohlfing@stanford.edu

IEEE Transactions on Medical Imaging
|September 2, 2004
PubMed
Summary
This summary is machine-generated.

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

Deep Learning-Based Segmentation of Geographic Atrophy: A Multi-Center, Multi-Device Validation in a Real-World Clinical Cohort.

Diagnostics (Basel, Switzerland)·2025
Same author

Deep Learning Model for Automated Classification of Macular Neovascularization Subtypes in AMD.

Investigative ophthalmology & visual science·2025
Same author

Deep learning model for automatic detection of different types of microaneurysms in diabetic retinopathy.

Eye (London, England)·2025
Same author

Machine Learning Quantification of Fluid Volume in Eyes With Retinal Vein Occlusion Treated With Aflibercept: The REVOLT Study.

Journal of vitreoretinal diseases·2025
Same author

Topographical Quantification of Retinal Fluid in Type 3 MNV and Associations With Short-Term Visual Outcomes.

American journal of ophthalmology·2024
Same author

Deep-learning based automated quantification of critical optical coherence tomography features in neovascular age-related macular degeneration.

Eye (London, England)·2023
Same journal

LLM-enhanced Neuron Segmentation and Reconstruction in Complex Mouse Brain Images.

IEEE transactions on medical imaging·2026
Same journal

Matrixed-Spectrum Decomposition Accelerated Linear Boltzmann Transport Equation Solver for Fast Scatter Correction in Multi-Spectral CT.

IEEE transactions on medical imaging·2026
Same journal

The Ritz Adjoint Method for MRI Pulse Design.

IEEE transactions on medical imaging·2026
Same journal

Physiology-guided Self-supervised Learning for Simultaneous Dual-Tracer PET Separation.

IEEE transactions on medical imaging·2026
Same journal

Informed-Exploration Reinforcement Learning for Automated Virtual Coronary Intervention Planning.

IEEE transactions on medical imaging·2026
Same journal

4D Reconstruction of Fetal Left Ventricle from Echocardiography via 2.5D Radial Segmentation and Graph-Fourier Reconstruction.

IEEE transactions on medical imaging·2026
See all related articles

Improved accuracy in biomedical image segmentation is achieved by weighting individual classifiers based on their estimated performance, outperforming equal weighting methods. This approach enhances pattern recognition for complex datasets.

Area of Science:

  • Pattern Recognition
  • Biomedical Image Analysis
  • Computational Neuroscience

Background:

  • Combining decisions from multiple classifiers can significantly improve classification accuracy over individual classifiers.
  • This principle has been previously demonstrated for atlas-based segmentation in biomedical imaging.
  • Conventional fusion methods often weight classifiers equally, such as with sum rule fusion.

Purpose of the Study:

  • To propose and evaluate novel performance-based classifier fusion methods for atlas-based biomedical image segmentation.
  • To address limitations of equal-weighting fusion techniques by incorporating classifier performance estimates.
  • To enhance the accuracy of segmenting complex 3D microscopy images, specifically bee brains.

Main Methods:

Related Experiment Videos

  • Developed two multiclass extensions of the expectation-maximization (EM) algorithm for estimating individual classifier performance.
  • Method 1: Independent parameter estimation per class with integration.
  • Method 2: Simultaneous consideration of all classes for parameter estimation.

    • Main Results:

    • Proposed EM-based performance-weighted fusion methods demonstrated superior accuracy compared to equal-weighting (sum rule) fusion.
    • Validation studies using transformed atlases and manual segmentation comparisons confirmed the efficacy of the proposed methods.
    • Applied methods to atlas-based segmentation of 3D confocal microscopy images of bee brains.

    Conclusions:

    • Performance-weighted classifier fusion significantly improves atlas-based segmentation accuracy in biomedical imaging.
    • The proposed EM-based methods offer a more effective approach than traditional equal-weighting fusion techniques.
    • These findings have implications for enhancing the analysis of complex biological structures from imaging data.