Related Experiment Video
Updated: Oct 10, 2025
08:20
Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images
Published on: October 27, 2023
1.8K
MRFGRO: a hybrid meta-heuristic feature selection method for screening COVID-19 using deep features.
Arijit Dey1, Soham Chattopadhyay2, Pawan Kumar Singh3
1Department of Computer Science and Engineering, Maulana Abul Kalam Azad University of Technology, Kolkata, West Bengal, 700064, India.
Scientific Reports
|December 16, 2021
Summary
A new hybrid model using deep learning and meta-heuristic algorithms accurately detects COVID-19 from CT scans. This approach offers a faster and more efficient alternative to traditional RT-PCR testing for diagnosing the respiratory disease.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Computational Biology
Background:
- COVID-19, a global pandemic, primarily affects the respiratory system.
- Traditional diagnostic methods like RT-PCR are time-consuming.
- Medical image analysis, particularly CT scans, shows promise for rapid COVID-19 detection.
Purpose of the Study:
- To develop a hybrid model for efficient and accurate COVID-19 detection using chest CT scans.
- To improve diagnostic speed and accuracy compared to existing methods.
Main Methods:
- A hybrid model combining fine-tuned Convolutional Neural Networks (CNNs) (GoogleNet, ResNet18) for feature extraction.
- A novel Manta Ray Foraging based Golden Ratio Optimizer (MRFGRO) for feature selection.
- Implementation and validation on three public datasets: COVID-CT, SARS-COV-2, and MOSMED.
Main Results:
- Achieved state-of-the-art classification accuracies: 99.15% (COVID-CT), 99.42% (SARS-COV-2), and 95.57% (MOSMED).
- Demonstrated superior efficiency compared to local texture descriptors for COVID-19 detection.
- Validated the model's effectiveness on diverse datasets.
Conclusions:
- The proposed hybrid model provides a highly efficient and accurate method for COVID-19 detection from CT scans.
- This AI-driven approach offers a significant advancement in rapid diagnosis of the respiratory disease.
- The model's performance suggests its potential for clinical application in pandemic scenarios.
