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 Video

Updated: Jun 25, 2026

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
07:35

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

Published on: October 11, 2018

Improving performance of computer-aided detection scheme by combining results from two machine learning classifiers.

Sang Cheol Park1, Jiantao Pu, Bin Zheng

  • 1Department of Radiology, University of Pittsburgh, 3362 Fifth Avenue, Pittsburgh, PA 15213, USA. parks@upmc.edu

Academic Radiology
|February 10, 2009
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

Regulatory crosstalk between KLF5, miR-29a and Fbw7/CDC4 cooperatively promotes atherosclerotic development.

Biochimica et biophysica acta. Molecular basis of disease·2017
Same author

Near-Infrared Light Triggered Upconversion Optogenetic Nanosystem for Cancer Therapy.

ACS nano·2017
Same author

Non-contact method to freely control the radiation patterns of antenna with multi-folded transformation optics.

Scientific reports·2017
Same author

Cost Effectiveness of Imatinib, Dasatinib, and Nilotinib as First-Line Treatment for Chronic-Phase Chronic Myeloid Leukemia in China.

Clinical drug investigation·2017
Same author

Improving Performance of Breast Cancer Risk Prediction by Incorporating Optical Density Image Feature Analysis: An Assessment.

Academic radiology·2017
Same author

Preliminary study on ultrasound-guided prostate biopsy specimen scores.

Experimental and therapeutic medicine·2017
Same journal

Improving Reliability of MRI Lumbar Spinal Stenosis Assessment Across Radiology and Spine Specialties: Impact of a Structured Education Intervention.

Academic radiology·2026
Same journal

Advances in CT and MRI for Yttrium-90 Radioembolization of Hepatocellular Carcinoma.

Academic radiology·2026
Same journal

Homogeneity of Liver Fat Distribution Serves as a Diagnostic Marker for Metabolic Dysfunction-Associated Steatohepatitis.

Academic radiology·2026
Same journal

MRI-based Predictors and Risk Constellations of Chronic Ankle Instability After Acute Lateral Ankle Sprain: A Multicenter Study.

Academic radiology·2026
Same journal

Early Prediction of Pathological Complete Response to Neoadjuvant Chemotherapy in Breast Cancer using a Longitudinal US-based Stack-model.

Academic radiology·2026
Same journal

Evaluating the Impact of Embolization on Outcomes in Iliopsoas Hematomas: A Multicenter Retrospective Propensity-matched Study.

Academic radiology·2026
See all related articles

Combining artificial neural networks (ANN) and k-nearest neighbor (KNN) classifiers in computer-aided detection (CAD) for breast masses improves diagnostic performance. This approach enhances accuracy and reliability in classifying true-positive and false-positive detections.

Area of Science:

  • Medical Imaging
  • Machine Learning in Healthcare
  • Breast Cancer Diagnostics

Background:

  • Computer-aided detection and diagnosis (CAD) schemes commonly use machine learning classifiers to distinguish true-positive from false-positive findings.
  • Global data-based and local instance-based classifiers are frequently employed for optimizing CAD performance.

Purpose of the Study:

  • To investigate the correlation between global data-based (artificial neural network - ANN) and local instance-based (k-nearest neighbor - KNN) classifiers.
  • To assess the potential performance improvement of a CAD scheme for breast mass detection by combining these two classifier types.

Main Methods:

  • A CAD scheme utilizing image filtering and region growth was employed.
  • An artificial neural network (ANN) and a k-nearest neighbor (KNN) algorithm were used as the global and local classifiers, respectively.

Related Experiment Videos

Last Updated: Jun 25, 2026

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
07:35

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

Published on: October 11, 2018

  • Performance was evaluated on an independent dataset of 400 mammographic examinations (200 cancer, 200 negative) using the free-response receiver-operating characteristic (FROC) method.

    • Main Results:

    • The ANN classifier (AUC=0.891) outperformed the KNN classifier (AUC=0.845).
    • Low correlation coefficients were observed between ANN and KNN detection scores (0.436 for true-positive, 0.161 for false-positive).
    • Combining classifiers increased the area under the curve (AUC) to 0.912, reduced AUC standard error by 14.4%, and improved sensitivity to 80.3% at 0.3 false positives per image.

    Conclusions:

    • Global (ANN) and local (KNN) machine-learning classifiers yield low correlated results in breast mass detection.
    • Combining the detection scores from ANN and KNN significantly enhances overall CAD performance and reliability.
    • This combined approach reduces the standard error in CAD performance assessment, indicating more robust evaluation.