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: Jan 9, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

8.6K

Performance Evaluation of Transfer Learning Based Medical Image Classification Techniques for Disease Detection.

Zeeshan Ahmad, Shudi Bao, Meng Chen

    Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference
    |December 3, 2025
    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

    Antimicrobial susceptibility among <i>Streptococcus pneumoniae</i> and <i>Haemophilus influenzae</i> collected globally between 2015 and 2017 as part of the Tigecycline Evaluation and Surveillance Trial (TEST).

    Infection and drug resistance·2019
    Same author

    Prevalence of phycotoxin contamination in shellfish from the Northern Bering Sea and the Chukchi Sea.

    Toxicon : official journal of the International Society on Toxinology·2019
    Same author

    Effects of Polypropylene Fibre and Strain Rate on Dynamic Compressive Behaviour of Concrete.

    Materials (Basel, Switzerland)·2019
    Same author

    Evaluation of the adequacy of measles laboratory diagnostic tests in the era of accelerating measles elimination in Beijing, China.

    Vaccine·2019
    Same author

    Profiling of histone 3 lysine 27 acetylation reveals its role in a chronic DSS-induced colitis mouse model.

    Molecular omics·2019
    Same author

    Seroepidemiology of measles in Beijing, China: a cross-sectional study.

    Human vaccines & immunotherapeutics·2019

    Transfer learning (TL) effectively classifies medical images using deep learning, with InceptionV3 outperforming other models. TL benefits limited data, but performance varies with model architecture and dataset.

    Area of Science:

    • Medical Imaging
    • Artificial Intelligence
    • Computer Vision

    Background:

    • Medical image classification is crucial for disease identification.
    • Deep learning models are powerful but require extensive data for training from scratch.
    • Transfer learning (TL) offers a solution by reusing pre-trained models.

    Purpose of the Study:

    • To comprehensively analyze transfer learning techniques for medical image classification using deep convolutional neural networks.
    • To evaluate the performance of six pre-trained models on a custom chest X-ray dataset.

    Main Methods:

    • Evaluation of six deep convolutional neural network models (AlexNet, VGG16, ResNet18, ResNet34, ResNet50, InceptionV3) using transfer learning.
    • Application on a custom chest X-ray dataset for disease detection.

    Related Experiment Videos

    Last Updated: Jan 9, 2026

    Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
    04:09

    Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

    Published on: October 10, 2018

    8.6K
  • Analysis of uncertainty and runtime for robustness and computational efficiency.

    • Main Results:

    • InceptionV3 demonstrated superior performance across all metrics.
    • ResNet models showed improved performance with increased depth.
    • Transfer learning proved beneficial, particularly for limited datasets, with performance influenced by model architecture, data size, and domain similarity.

    Conclusions:

    • Transfer learning is a valuable technique for medical image classification, especially with limited data.
    • Model selection for TL depends on factors like architecture, dataset size, and domain similarity.
    • A well-trained feature extractor allows efficient prediction with lightweight models.