Related Experiment Video
Updated: Jul 1, 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
A hybridized feature extraction for COVID-19 multi-class classification on computed tomography images
Hassana Abubakar1, Fadi Al-Turjman2,3, Zubaida S Ameen4
1Biomedical Engineering Department, Faculty of Engineering, Near East University, Mersin 10, Turkey.
Heliyon
|March 11, 2024
Summary
Early diagnosis of COVID-19 is crucial. Combining deep learning features with Histogram of Oriented Gradient (HOG) and Support Vector Machines (SVM) achieved 99.4% accuracy for detecting the virus.
Area of Science:
- Medical Imaging
- Machine Learning
- Artificial Intelligence
Background:
- COVID-19, caused by SARS-CoV-2, has led to millions of deaths globally.
- The virus continuously mutates, creating more transmissible strains.
- Traditional diagnostic methods like RT-PCR face limitations in rapid, widespread diagnosis.
Purpose of the Study:
- To develop an effective and accurate method for early COVID-19 diagnosis.
- To improve the performance of machine learning classifiers for COVID-19 detection.
- To explore the efficacy of combined feature extraction techniques.
Main Methods:
- Utilized Histogram of Oriented Gradient (HOG) for feature extraction.
- Employed eight deep learning models for feature extraction.
- Applied K-Nearest Neighbour (KNN) and Support Vector Machines (SVM) for classification.
- Proposed a combined feature set integrating HOG and deep learning features.
Main Results:
- The VGG-16 + HOG combined feature achieved 99.4% overall accuracy when using the SVM classifier.
- The concatenated feature approach significantly enhanced classifier performance.
- This hybrid method demonstrated high efficacy in COVID-19 detection.
Conclusions:
- The proposed concatenated feature of HOG and deep learning features improves SVM classifier performance for COVID-19 detection.
- This approach offers a promising avenue for accurate and early diagnosis of COVID-19.
- Medical imaging combined with advanced machine learning techniques can aid in managing respiratory disease pandemics.
