Related Experiment Video
Updated: Jul 25, 2025
08:05
Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia
Published on: December 19, 2020
14.2K
Optimal feature selection for COVID-19 detection with CT images enabled by metaheuristic optimization and artificial
Dattaprasad A Torse1, Rajashri Khanai2, Krishna Pai1
1Department of ECE, KLE Dr. M.S. Sheshgiri College of Engineering and Technology, Udyambag, Belagavi, KA 590008 India.
Multimedia Tools and Applications
|June 26, 2023
Summary
This study introduces a new artificial intelligence (AI) method for detecting COVID-19 pneumonia using chest CT scans. The AI model achieved 87.2% accuracy, improving diagnostic efficiency in healthcare.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Computational Biology
Background:
- Coronaviruses (CoV) encompass diverse viruses, including those causing common colds and severe lung infections.
- The COVID-19 pandemic, originating in Wuhan, China, has mutated and spread globally, straining healthcare systems, particularly in densely populated nations like India.
- Accurate and efficient diagnostic tools are crucial for managing the widespread impact of COVID-19, especially in resource-limited settings.
Purpose of the Study:
- To develop and evaluate a novel metaheuristic algorithm for the automatic detection of COVID-19 pneumonia.
- To utilize artificial intelligence (AI) on high-resolution computed tomography (HRCT) images for improved diagnostic accuracy.
- To classify chest HRCT images into three categories: normal, COVID-19, and pneumonia.
Main Methods:
- A novel metaheuristic algorithm, the modified quantum-based marine predators algorithm (Mq-MPA), was employed for feature selection.
- A least square support vector machine (LSSVM) classifier was used for the three-class classification task.
- The performance of the proposed AI model was evaluated using classification accuracy and F1-score metrics.
Main Results:
- The proposed AI model achieved a classification accuracy of 87.2% and an F1-score of 86.3% for multiclass classification.
- The Mq-MPA feature selection algorithm demonstrated a 23% reduction in LSSVM classification time compared to deep learning models.
- Analysis of the information transfer rate (ITR) indicated the efficiency of the Mq-MPA in feature selection.
Conclusions:
- The developed AI-driven approach shows promise for accurate and efficient COVID-19 pneumonia detection using chest HRCT images.
- The Mq-MPA algorithm effectively reduces classification time, enhancing the practicality of AI in medical diagnostics.
- This AI technique can aid healthcare professionals, especially in remote areas lacking specialized expertise, by improving diagnostic capabilities.
