PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
Synapse
⌘+K
Synapse
April 24, 2026DiagnosticsOpen Access

A Hybrid Lung and Colon Histopathological Image Classification Framework Using MobileNetV3-Small Deep Features and Differential Evolution Optimization

View Full Paper
Ask AI
Bookmark
Share

Authors

MNMuhammad Usama NaveedSJSohail JabbarMIMuhammad Munwar Iqbal

Discussion

Loading...

Member takes

Overview

Automated classification framework demonstrates high accuracy in lung and colon cancer detection, indicating efficiency improvements.

Key Points

  • The aim is to develop an automated framework for accurately classifying lung and colon cancer using histopathological images.
  • Used MobileNetV3-Small model through transfer learning for feature extraction.
  • Optimized deep features with a differential evolution algorithm to reduce dimension.
  • Trained on an enhanced LC25000 dataset with image patches refined for better clinical application.
  • Achieved a maximum classification accuracy of 98.14% using Quadratic Support Vector Machine.
  • Demonstrated a 21.3× speed-up with bagged trees compared to traditional methods.
  • Outperformed several state-of-the-art approaches, improving baseline performance by 3.34%.

Cite This Study

Naveed et al. (2026) studied this question.

synapsesocial.com/papers/69eb092b553a5433e34b3c72https://doi.org/10.3390/diagnostics16091256
View Full Paper
Ask AI
Bookmark
Share