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

SVM-based glioma grading: Optimization by feature reduction analysis.

Frank G Zöllner1, Kyrre E Emblem, Lothar R Schad

  • 1Computer Assisted Clinical Medicine, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany. frank.zoellner@medma.uni-heidelberg.de

Zeitschrift Fur Medizinische Physik
|April 17, 2012
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

Mapping radiosensitivity in glioblastoma using MR elastography and biomechanical modeling.

Scientific reports·2026
Same author

Decreased Tissue Sodium Concentration in Suspected Prostate Cancer Detected by Internal-Reference <sup>23</sup>Na MRI: A Prospective Exploratory Study.

Diagnostics (Basel, Switzerland)·2026
Same author

Renal Clinical Study Participants Support Data Sharing and Use of Artificial Intelligence.

Kidney international reports·2026
Same author

Adaptive Aggregation of Monte Carlo Augmented Decomposed Filters for Efficient Group-Equivariant Convolutional Neural Network.

IEEE transactions on pattern analysis and machine intelligence·2026
Same author

Automated delineation of putative non-contrast-enhancing tumor in glioblastoma: Prognostic insights.

Neuro-oncology·2026
Same author

Biomechanical mapping of tumor growth: A novel method to quantify glioma infiltration and mass effect.

Medical physics·2026

Feature reduction methods like principal component analysis (PCA) effectively classify glioma grade using support vector machines (SVM). PCA achieved 85% accuracy, simplifying analysis while maintaining high diagnostic performance.

Area of Science:

  • Neuro-oncology
  • Medical imaging analysis
  • Machine learning in medicine

Background:

  • Accurate glioma grading is crucial for treatment planning and prognosis.
  • Support vector machine (SVM) classification models show promise for automated tumor grading.
  • Feature reduction is essential for optimizing SVM model efficiency and interpretability.

Purpose of the Study:

  • To evaluate the predictive power of different feature reduction techniques for SVM-based glioma grading.
  • To compare Pearson's correlation coefficients (PCC), principal component analysis (PCA), and independent component analysis (ICA) for feature selection.
  • To determine the optimal feature reduction strategy for classifying glioma grade using cerebral blood volume (CBV) histograms and patient age.

Main Methods:

Related Experiment Videos

  • Investigated three feature reduction methods: PCC, PCA, and ICA.
  • Applied these methods to whole-tumor CBV histograms (100 bins) and patient age from 101 untreated glioma patients.
  • Utilized a previously reported SVM approach for tumor grading based on reduced feature sets.

    • Main Results:

    • Principal component analysis (PCA) yielded the highest classification accuracy at 85% (sensitivity=89%, specificity=84%) by reducing features to 3 principal components.
    • Pearson's correlation coefficients (PCC) achieved 82% accuracy (89% sensitivity, 77% specificity) with 2 dimensions.
    • Independent component analysis (ICA) resulted in 79% accuracy (87% sensitivity, 75% specificity) using 9 dimensions.
    • All three methods significantly reduced the number of features (up to 98%) while maintaining classification accuracy comparable to literature values (∼87%), improving speed by up to 30%.

    Conclusions:

    • Feature reduction analysis, particularly PCA, is a valuable approach for enhancing SVM-based glioma grading.
    • Reduced feature sets maintain high diagnostic accuracy, offering improved computational efficiency and simplicity.
    • These findings support the integration of feature reduction techniques in neuro-oncology for faster and more streamlined glioma classification.