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: Apr 18, 2026

Quantitation of Protein Expression and Co-localization Using Multiplexed Immuno-histochemical Staining and Multispectral Imaging
08:40

Quantitation of Protein Expression and Co-localization Using Multiplexed Immuno-histochemical Staining and Multispectral Imaging

Published on: April 8, 2016

13.6K

Evaluation of performance metrics for histopathological image classifier optimization.

Nishant Zachariah, Sonal Kothari, Senthil Ramamurthy

    Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference
    |January 9, 2015
    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

    MRI-targeted versus systematic needle core biopsies in prostate cancer: a patient-based analysis of potential diagnostic and biologic underestimation.

    Journal of clinical pathology·2026
    Same author

    Monocular Markerless Motion Capture Enables Quantitative Assessment of Upper Extremity Reachable Workspace.

    Sensors (Basel, Switzerland)·2026
    Same author

    Polygenic Stroke Risk and Autoimmune Disease: Associations and Clinical Modifiers.

    IEEE transactions on bio-medical engineering·2026
    Same author

    Metastatic malignant non-cervical gynecological tumors involving the genitourinary system: A contemporary clinicopathologic analysis of 27 patients.

    Human pathology·2026
    Same author

    Deciphering the Genomic Landscape of Renal Cell Carcinoma Brain Metastases.

    bioRxiv : the preprint server for biology·2026
    Same author

    Single-Cell Atlas of Renal Cell Carcinoma Brain Metastasis Uncovers Mechanisms of Immune Dysfunction and Resistance.

    bioRxiv : the preprint server for biology·2026

    This study found that the Lift metric best measures the generalization of machine learning cancer prediction models in histopathological images. It outperforms common metrics like accuracy, ensuring more reliable predictions on new data.

    Area of Science:

    • Computational pathology
    • Medical image analysis
    • Machine learning in oncology

    Background:

    • Clinical decision support systems (CDSS) leverage machine learning (ML) for cancer prediction in histopathology.
    • Model generalization, the ability to perform on unseen data, is crucial for ML classifier development.
    • Existing performance metrics may not accurately reflect model robustness in this domain.

    Purpose of the Study:

    • To quantitatively compare various performance metrics for evaluating ML classifiers in cancer prediction.
    • To identify the metric that best reflects a model's generalization capability on external data.
    • To establish a reliable method for selecting robust ML models for histopathological cancer diagnosis.

    Main Methods:

    • Evaluation of 23 different performance metrics using a nested cross-validation pipeline.

    More Related Videos

    Author Spotlight: Investigating Immune Cell Dynamics in the Tumor Microenvironment — Challenges and Innovations in Cancer Prognosis
    07:32

    Author Spotlight: Investigating Immune Cell Dynamics in the Tumor Microenvironment — Challenges and Innovations in Cancer Prognosis

    Published on: April 12, 2024

    2.2K

    Related Experiment Videos

    Last Updated: Apr 18, 2026

    Quantitation of Protein Expression and Co-localization Using Multiplexed Immuno-histochemical Staining and Multispectral Imaging
    08:40

    Quantitation of Protein Expression and Co-localization Using Multiplexed Immuno-histochemical Staining and Multispectral Imaging

    Published on: April 8, 2016

    13.6K
    Author Spotlight: Investigating Immune Cell Dynamics in the Tumor Microenvironment — Challenges and Innovations in Cancer Prognosis
    07:32

    Author Spotlight: Investigating Immune Cell Dynamics in the Tumor Microenvironment — Challenges and Innovations in Cancer Prognosis

    Published on: April 12, 2024

    2.2K
  • Quantitative comparison of metric correlation between internal and external validation sets.
  • Development of a 4-class classifier using the identified optimal metric for external validation.

    • Main Results:

    • Commonly used metrics (e.g., accuracy, ROC curve) showed limitations in reflecting model robustness.
    • The Lift metric demonstrated the highest correlation (R² = 0.57) between internal and external validation sets.
    • The Lift metric was identified as the best indicator of classifier generalization performance.

    Conclusions:

    • The Lift metric provides a more accurate assessment of ML model generalization in histopathological cancer prediction than traditional metrics.
    • Utilizing the Lift metric enables the development of more robust and reliable cancer prediction models for clinical decision support.
    • This research establishes a benchmark for performance metric evaluation in computational pathology.