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

Domain-Specific Transfer Learning for Gastric Cancer Tissue Classification.

Venkata Ramana Kaneti1, Santosh Reddy P2, Alpha Vijayan3

  • 1Department of Computer Science and Engineering, VNR Vignana Jyothi Institute of Engineering and Technology, Hyderabad, Telangana, 500090, India.

Journal of Imaging Informatics in Medicine
|July 7, 2026
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

Optimized cluster based routing protocol for IoT enabled healthcare data networks.

Scientific reports·2026
Same author

Enhancing urban traffic congestion prediction through efficientnet and optimized ensemble learning models.

Scientific reports·2025
See all related articles
This summary is machine-generated.

CTransPath, a Swin Transformer model, achieves high accuracy in classifying gastric cancer tissues. This computational pathology approach offers a reproducible benchmark for gastric cancer research.

Area of Science:

  • Computational pathology
  • Artificial intelligence in oncology
  • Histopathology image analysis

Background:

  • Gastric cancer is a leading cause of cancer-related deaths globally.
  • Accurate histopathological classification is crucial for cancer diagnosis and prognosis.
  • Manual tissue classification is labor-intensive and prone to inconsistencies.

Purpose of the Study:

  • To evaluate the performance of CTransPath, a Swin Transformer model, for eight-class gastric tissue classification.
  • To assess the effectiveness of a two-stage transfer learning approach with domain-specific pretraining.
  • To establish a reproducible research benchmark for computational pathology in gastric cancer.

Main Methods:

  • CTransPath model pretrained on 15 million histopathological images.
Keywords:
Ablation studyDeep learningExplainable artificial intelligenceFoundation modelsGastric cancerGrad-CAMHistopathologySHAPTissue classificationTransfer learningVision transformer

Related Experiment Videos

  • Utilized the HMU-GC-HE-30K dataset with 31,096 gastric tissue samples.
  • Employed tenfold stratified cross-validation with data augmentation, class-weighted loss, and cosine annealing.

    • Main Results:

    • Achieved a macro AUC of 96.82% and patch-level accuracy of 76.74%.
    • Demonstrated strong patch-level discriminative ability across eight tissue classes.
    • CTransPath outperformed larger models like UNI and Virchow2 in internal validation.

    Conclusions:

    • CTransPath shows significant potential for automated histopathological classification of gastric cancer.
    • Task-oriented design and domain-specific pretraining are effective for computational pathology.
    • Further external validation is required before clinical application.