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

Generative-discriminative basis learning for medical imaging.

Nematollah K Batmanghelich1, Ben Taskar, Christos Davatzikos

  • 1Department of Electrical and System Engineering, University of Pennsylvania, Philadelphia, PA 19104, USA. batmanghelich@gmail.com

IEEE Transactions on Medical Imaging
|July 28, 2011
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

Magnetic Resonance Imaging01:24

Magnetic Resonance Imaging

Magnetic resonance imaging (MRI) is a noninvasive medical imaging technique based on a phenomenon of nuclear physics discovered in the 1930s, in which matter exposed to magnetic fields and radio waves was found to emit radio signals. In 1970, a physician and researcher named Raymond Damadian noticed that malignant (cancerous) tissue gave off different signals than normal body tissue. He applied for a patent for the first MRI scanning device in clinical use by the early 1980s. The early MRI...

You might also read

Related Articles

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

Sort by
Same author

Genetic and Clinical Correlates of AI-Based Brain Aging Patterns in Cognitively Unimpaired Individuals.

JAMA psychiatry·2024
Same author

Dimensional Neuroimaging Endophenotypes: Neurobiological Representations of Disease Heterogeneity Through Machine Learning.

ArXiv·2024
Same author

Five dominant dimensions of brain aging are identified via deep learning: associations with clinical, lifestyle, and genetic measures.

medRxiv : the preprint server for health sciences·2024
Same author

Gene-SGAN: discovering disease subtypes with imaging and genetic signatures via multi-view weakly-supervised deep clustering.

Nature communications·2024
Same author

Genomic loci influence patterns of structural covariance in the human brain.

Proceedings of the National Academy of Sciences of the United States of America·2023
Same author

Personalized functional brain network topography is associated with individual differences in youth cognition.

Nature communications·2023

This study introduces a new method for reducing dimensions in medical images, improving classification accuracy and clinical interpretability. The approach enhances predictions for conditions like Alzheimer's disease using semi-supervised learning.

Area of Science:

  • Medical Imaging Analysis
  • Machine Learning for Healthcare
  • Computational Neuroscience

Background:

  • High-dimensional medical imaging data (millions of voxels) presents challenges for accurate classification.
  • Existing dimensionality reduction methods may not preserve clinically relevant discriminative signals.
  • Semi-supervised learning (SSL) offers potential for improving model performance with limited labeled data.

Purpose of the Study:

  • To develop a novel dimensionality reduction technique for medical image classification.
  • To create a low-dimensional, clinically interpretable feature representation from high-dimensional data.
  • To extend the method to a semi-supervised learning (SSL) framework for enhanced performance.

Main Methods:

  • Formulated dimensionality reduction as a constrained optimization problem combining generative and discriminative objectives.

Related Experiment Videos

  • Developed a novel large-scale algorithm to solve the optimization problem.
  • Extended the method to the semi-supervised learning (SSL) setting using labeled and unlabeled medical imaging data.

    • Main Results:

    • Achieved state-of-the-art or comparable accuracy rates in the fully supervised setting, with representations consistent with clinical findings.
    • Demonstrated superior or comparable performance to existing methods on benchmark datasets in the SSL setting.
    • In medical imaging, the SSL extension slightly improved generalization accuracy for Alzheimer's disease (AD) and normal control (NC) classification and predicted MCI to AD conversion.

    Conclusions:

    • The proposed dimensionality reduction method is effective for classification in medical imaging, offering improved accuracy and interpretability.
    • The semi-supervised extension enhances model generalization and shows promise for predicting disease progression, such as mild cognitive impairment (MCI) to AD conversion.
    • This approach provides a powerful tool for analyzing complex medical imaging datasets and advancing diagnostic capabilities.