Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Mixture of segmenters with discriminative spatial regularization and sparse weight selection.

Ting Chen¹, Baba C Vemuri, Anand Rangarajan

¹Department of CISE, University of Florida, Gainesville, FL, USA. tichen@cise.ufl.edu

Medical Image Computing and Computer-Assisted Intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention

|October 19, 2011

Summary

This study introduces a new image segmentation algorithm that learns to combine multiple weak segmenters into a strong one. This novel approach significantly improves segmentation performance, particularly in atlas-based applications.

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

Kernel Stein Discrepancy on Lie Groups: Theory and Applications.

IEEE transactions on information theory·2024

Same author

HIGHER ORDER GAUGE EQUIVARIANT CONVOLUTIONS FOR NEURODEGENERATIVE DISORDER CLASSIFICATION.

Proceedings. IEEE International Symposium on Biomedical Imaging·2024

Same author

Horospherical Decision Boundaries for Large Margin Classification in Hyperbolic Space.

Advances in neural information processing systems·2024

Same author

An Empirical Bayes Approach to Shrinkage Estimation on the Manifold of Symmetric Positive-Definite Matrices.

Journal of the American Statistical Association·2024

Same author

Toward Improving the Generation Quality of Autoregressive Slot VAEs.

Neural computation·2024

Same author

Geometric Deep Learning for Unsupervised Registration of Diffusion Magnetic Resonance Images.

Information processing in medical imaging : proceedings of the ... conference·2024

Area of Science:

Medical image analysis
Computer vision
Machine learning

Background:

Image segmentation is crucial for medical image analysis.
Combining multiple weak segmentation algorithms can improve overall accuracy.
Existing methods for combining segmenters have limitations.

Purpose of the Study:

To present a novel algorithm for automatic learning of weak segmenter combinations.
To develop a strong segmentation model by adaptively weighting weak segmenters.
To improve segmentation accuracy in medical imaging applications.

Main Methods:

A discriminative spatial regularization approach is used to learn weighted combinations during training.
A closed-form solution is derived for the cost function.

Related Experiment Videos

Sparse regularization is applied during testing to prevent overfitting.

Main Results:

The proposed algorithm significantly outperforms individual weak segmenters.
Empirical results demonstrate improved performance in atlas-based segmentation.
Comparisons show advantages over existing weak segmenter combination strategies on hippocampal datasets.

Conclusions:

The novel algorithm effectively learns adaptive combinations of weak segmenters.
This approach offers a significant advancement in image segmentation techniques.
The method shows promise for various medical imaging segmentation tasks.