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

Detecting potential labeling errors in microarrays by data perturbation.

Andrea Malossini1, Enrico Blanzieri, Raymond T Ng

  • 1Department of Information and Communication Technology, University of Trento, 38050 Povo, Italy. malossin@dit.unitn.it

Bioinformatics (Oxford, England)
|July 6, 2006
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

Smartphone-based Detection of Group A Streptococcal Pharyngitis in Ugandan Children: A Pilot Study.

The Pediatric infectious disease journal·2026
Same author

Validation of a Machine Learning Model for Predicting Postmastectomy Radiotherapy Recommendation Following Immediate Breast Reconstruction.

Plastic and reconstructive surgery. Global open·2025
Same author

CaseReportBench: An LLM Benchmark Dataset for Dense Information Extraction in Clinical Case Reports.

Proceedings of machine learning research·2025
Same author

Natural language processing for local, regional, and distant breast cancer relapse identification in pathology reports.

Breast cancer research and treatment·2025
Same author

Excess mortality in COVID-19-negative people with non-communicable disorders during the first pandemic wave.

BMC public health·2025
Same author

DNA methylation differences between cord and adult white blood cells reflect postnatal immune cell maturation.

Communications biology·2025
Same journal

3DICE: Interpretable 3D Cross-Modal Learning for Drug-Target Interaction Prediction and Large-Scale Drug Discovery.

Bioinformatics (Oxford, England)·2026
Same journal

KASSPer: Kinase Active Site Structure Prediction using Protein and Ligand Language Models and Its Application to Virtual Screening.

Bioinformatics (Oxford, England)·2026
Same journal

IDR searcher: a search engine solution for public image resources.

Bioinformatics (Oxford, England)·2026
Same journal

KCFtools: Rapid alignment-free method for introgression screening and GWAS using k-mer profiles.

Bioinformatics (Oxford, England)·2026
Same journal

Meta2DB: Curated shotgun metagenomic feature sets and metadata for health state prediction.

Bioinformatics (Oxford, England)·2026
Same journal

conMItion: an R package adjusting confounding factors for associations in multi-omics.

Bioinformatics (Oxford, England)·2026
See all related articles

Accurate medical classification relies on correctly labeled data. This study introduces algorithms to automatically detect mislabeled samples, improving classifier performance, especially with limited data.

Area of Science:

  • Medical informatics
  • Machine learning in healthcare

Background:

  • Accurate data labeling is crucial for medical classification performance.
  • Subjectivity in medical sample labeling (e.g., biopsies) can lead to mislabeled data.
  • Even a few mislabeled samples can significantly degrade classifier accuracy, particularly in small datasets.

Purpose of the Study:

  • To develop an automated method for detecting potentially mislabeled samples in medical datasets.
  • To address the challenge of data quality issues impacting machine learning model performance in healthcare.

Main Methods:

  • Proposed two algorithms: classification-stability and leave-one-out-error-sensitivity.
  • Utilized the leave-one-out perturbation matrix as a core computational component.
  • Evaluated a support vector machine-based classification-stability algorithm on real-world datasets.

Related Experiment Videos

Main Results:

  • The classification-stability algorithm, particularly its support vector machine variant, demonstrated high accuracy in identifying mislabeled samples across three datasets.
  • The generated 'suspect lists' of potentially mislabeled data were of high quality.
  • A correction heuristic proved beneficial when human review was not feasible.

Conclusions:

  • Automated detection of mislabeled samples is feasible and effective using the proposed algorithms.
  • The developed methods can enhance the reliability of medical classification models by improving data quality.
  • These techniques offer a valuable tool for data curation in medical machine learning applications.