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

Hybrid I-ResNeT-ViT and Cost-Sensitive InceptionV3 Models for Tumour Severity and Malignancy Classification Using

Jagadish N1,2, Ravi Kumar J1, Prasad A Y3

  • 1Department of Computer Science and Engineering, Atria Institute of Technology, Bengaluru, Visvesvaraya Technological University (VTU), Belagavi, Karnataka, 590018, India.

Asian Pacific Journal of Cancer Prevention : APJCP
|May 22, 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

Robust Brain Tumor Detection and Classification From Multichannel MRI Using Deep Learning.

Developmental neurobiology·2025
Same author

Effect of Different Irrigant Activation Techniques on the Penetration of Calcium Hydroxide, an Intracanal Medicament: An In Vitro Study.

Cureus·2023
See all related articles
This summary is machine-generated.

A novel deep learning framework accurately classifies early-stage tumor severity and malignancy using multiple imaging types. This AI tool enhances diagnostic accuracy and supports clinical decisions in oncology.

Area of Science:

  • Artificial Intelligence in Oncology
  • Medical Imaging Analysis
  • Deep Learning for Cancer Diagnosis

Background:

  • Accurate early tumor severity and malignancy classification are crucial for effective patient treatment and outcomes.
  • Conventional diagnostic methods face challenges in precisely differentiating early-stage tumors, leading to potential misdiagnosis or treatment delays.

Purpose of the Study:

  • To develop an advanced deep learning framework for automated, accurate, and interpretable classification of tumor severity and malignancy.
  • The framework aims to integrate multiple medical imaging modalities for comprehensive analysis.

Main Methods:

  • Utilized high-resolution mammography, MRI, and CT images from public repositories.
  • Employed Histogram Equalization with Region-Based Segmentation (HE-RBS) for data pre-processing and an iResNet with ViT Feature Fusion (iRViT-HFF) for feature extraction.
Keywords:
Deep LearningHistogram Equalization with Region-Based SegmentationMalignancyTumour Severity

Related Experiment Videos

  • Implemented an Explainable Cost-Sensitive InceptionV3 (CS-InceptionV3) model for classification and interpretability.

    • Main Results:

    • Achieved high performance metrics: 97.6% accuracy, 96.9% sensitivity, 98.3% specificity, and 97.2% F1-score.
    • Demonstrated superior performance compared to conventional and other deep learning approaches.
    • Successfully classified early-stage tumors across various imaging modalities and provided interpretable heatmaps.

    Conclusions:

    • The developed hybrid deep learning framework offers a reliable and interpretable solution for early-stage tumor classification.
    • This tool can significantly support and enhance clinical oncology workflows and decision-making.