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

A tree-based decision rule for identifying profile groups of cases without predefined classes: application in diffuse

Elias Zintzaras1, Maria Bai, Christos Douligeris

  • 1Department of Biomathematics, University of Thessaly School of Medicine, Larissa, Greece. zintza@med.uth.gr

Computers in Biology and Medicine
|August 10, 2006
PubMed
Summary

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

Inferring accumulation times of mitochondrial DNA deletion mutants from cross-sectional single-cell data: methodological framework and validation.

npj aging·2026
Same author

Glucagon-like Peptide-1 Receptor Agonists and Platelet Function: Potential Benefits Beyond Glycemic Control.

Pharmaceuticals (Basel, Switzerland)·2026
Same author

Assessment of the Reporting Quality of Randomized, Controlled Trials of Percutaneous Patent Foramen Ovale Closure for the Treatment of Migraine Based on the CONSORT Statement.

Innovations in clinical neuroscience·2025
Same author

Multicohort study testing the generalisability of the SASKit-ML stroke and PDAC prognostic model pipeline to other chronic diseases.

BMJ open·2024
Same author

A Risk-Difference Meta-Analysis for the Prophylactic Treatments of Chronic Migraine.

Cureus·2024
Same author

Healthy Aging in Times of Extreme Temperatures: Biomedical Approaches.

Aging and disease·2023

This study introduces a forward growing classification tree to identify low and high proliferation profiles in diffuse large B-cell lymphomas. The method successfully derived decision rules for these profile groups using k-means clustering.

Area of Science:

  • Biostatistics
  • Oncology
  • Computational Biology

Background:

  • Diffuse large B-cell lymphomas (DLBCL) exhibit variable proliferation rates impacting prognosis.
  • Identifying distinct proliferation profiles is crucial for understanding DLBCL heterogeneity.
  • Existing classification methods may require predefined classes, limiting their application to novel datasets.

Purpose of the Study:

  • To evaluate the utility of a forward growing classification tree as a supplementary method to cluster analysis.
  • To derive a decision rule for identifying distinct proliferation profile groups in DLBCL cases without predefined classes.
  • To explore the data structure and identify low and high proliferation profiles based on immunohistochemical markers.

Main Methods:

  • Applied k-means clustering to derive initial classes from immunohistochemical expression levels of proliferation proteins in 79 DLBCL cases.

Related Experiment Videos

  • Utilized a forward growing classification tree with a 2% threshold for misclassification rate improvement to establish a decision rule.
  • Defined decision rules based on splitting points of the classification tree variables.

    • Main Results:

    • Successfully identified ten distinct case clusters using k-means clustering.
    • The classification tree effectively separated cases into low and high proliferation profile groups.
    • The derived decision rule provided clear criteria for classifying DLBCL cases based on proliferation markers.

    Conclusions:

    • Forward growing classification trees serve as a valuable supplement to cluster analysis for unsupervised group identification.
    • This methodology provides robust decision rules for classifying biological profiles, such as proliferation in DLBCL.
    • The approach facilitates data exploration and enhances the understanding of complex biological datasets.