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

Nonlinear support vector machine visualization for risk factor analysis using nomograms and localized radial basis

Baek Hwan Cho1, Hwanjo Yu, Jongshill Lee

  • 1Department of Biomedical Engineering, Hanyang University, Seoul 133-605, Korea. uranus@bme.hanyang.ac.kr

IEEE Transactions on Information Technology in Biomedicine : a Publication of the IEEE Engineering in Medicine and Biology Society
|March 20, 2008
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

Noninvasive Detection of Acute Hyperglycemia Using Signal from Wearable ECG Sensors Considering Individual HRV Response Delays to Glucose.

Biosensors·2026
Same author

What affects natural killer cell activity: a cross-sectional study.

Frontiers in immunology·2026
Same author

Erratum: A review of clinical pharmacology considerations in antibody-drug conjugates approved by the US Food and Drug Administration between 2000 and 2025.

Translational and clinical pharmacology·2026
Same author

Association Between Complete Blood Cell Count (CBC)-Derived Inflammatory Markers and Natural Killer Cell Activity: A Cross-Sectional Study.

Journal of inflammation research·2026
Same author

Supercapacitive Carbon-Coated δ-MnO<sub>2</sub> Nanosheets as a Solid-Contact Material for Self-Testing Potassium-Ion Biosensors.

Small (Weinheim an der Bergstrasse, Germany)·2026
Same author

<i>Aster yomena</i> Alleviates Chronic Unpredictable Mild Stress (CUMS)-Induced Depressive Cognitive Dysfunction by Regulating the HPA Axis and TLR4/NF-κB Pathway.

Journal of microbiology and biotechnology·2026
Same journal

Categorization and segmentation of intestinal content frames for wireless capsule endoscopy.

IEEE transactions on information technology in biomedicine : a publication of the IEEE Engineering in Medicine and Biology Society·2013
Same journal

An intelligent scoring system and its application to cardiac arrest prediction.

IEEE transactions on information technology in biomedicine : a publication of the IEEE Engineering in Medicine and Biology Society·2013
Same journal

Guest editorial: Multimedia services and technologies for e-health (MUST-EH).

IEEE transactions on information technology in biomedicine : a publication of the IEEE Engineering in Medicine and Biology Society·2013
Same journal

Editorial: From “information technology in biomedicine” to “biomedical and health informatics”.

IEEE transactions on information technology in biomedicine : a publication of the IEEE Engineering in Medicine and Biology Society·2013
Same journal

Equipment location in hospitals using RFID-based positioning system.

IEEE transactions on information technology in biomedicine : a publication of the IEEE Engineering in Medicine and Biology Society·2013
Same journal

Distributed system for cognitive stimulation over interactive TV.

IEEE transactions on information technology in biomedicine : a publication of the IEEE Engineering in Medicine and Biology Society·2012
See all related articles

A new localized radial basis function (LRBF) kernel and visualization system improve the interpretability of nonlinear support vector machines (SVMs) for disease diagnosis. This approach maintains high accuracy while offering clearer insights for physicians.

Area of Science:

  • Machine Learning
  • Medical Informatics
  • Data Visualization

Background:

  • Nonlinear classifiers like support vector machines (SVMs) with radial basis function (RBF) kernels offer high accuracy in disease diagnosis.
  • A key limitation is the difficulty in visualizing these complex models, hindering intuitive interpretation for medical professionals.

Purpose of the Study:

  • To introduce a novel localized radial basis function (LRBF) kernel and a visualization system for risk factor analysis (VRIFA).
  • To enhance the interpretability of nonlinear SVMs in medical diagnosis without compromising prediction accuracy.
  • To enable physicians to better understand and utilize the results of complex classification models.

Main Methods:

  • Development of the localized radial basis function (LRBF) kernel, a new nonlinear kernel for SVMs.

Related Experiment Videos

  • Implementation of the visualization system for risk factor analysis (VRIFA) utilizing nomograms and the LRBF kernel.
  • Evaluation of the LRBF kernel and VRIFA system on three benchmark medical datasets: breast cancer, diabetes, and heart disease.

    • Main Results:

    • The LRBF kernel demonstrated classification performance comparable to the traditional RBF kernel.
    • The LRBF kernel, when visualized using a nomogram, significantly improved the interpretability of nonlinear SVM results.
    • The LRBF kernel showed increased robustness to noisy features compared to the RBF kernel.
    • A trade-off was observed: the LRBF kernel's accuracy degraded more than RBF's when crucial features were removed.

    Conclusions:

    • The developed LRBF kernel and VRIFA system effectively address the interpretability challenge in nonlinear SVMs for medical diagnosis.
    • The LRBF kernel offers a viable alternative to RBF kernels, providing a balance between predictive power and visual explainability.
    • The VRIFA system successfully visualizes results from both linear and nonlinear SVMs employing LRBF kernels, aiding clinical decision-making.