A Deep Learning and Handcrafted Based Computationally Intelligent Technique for Effective COVID-19 Detection from...

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Computational Biology

Background:

• The COVID-19 pandemic has caused significant global health and economic disruption.
• Automated medical image analysis aids in disease diagnosis and management.
• Existing diagnostic methods require enhancement for efficiency and accuracy.

Purpose of the Study:

• To propose a novel, efficient framework for COVID-19 classification using medical imaging.
• To develop a hybrid feature extraction approach for enhanced diagnostic performance.
• To improve the accuracy and speed of COVID-19 detection systems.

Main Methods:

• A hybrid feature extraction method combining deep learning (ResNet101, DenseNet201) and handcrafted features (Weber Local Descriptor with DCT).
• Image data preprocessing followed by feature extraction and fusion.
• Feature selection using entropy and performance evaluation against established methods.

Main Results:

• The proposed framework demonstrated superior performance compared to existing methods.
• Achieved high accuracy in COVID-19 classification.
• Showcased improved efficiency in terms of processing time.

Conclusions:

• The developed hybrid feature extraction framework is effective for efficient COVID-19 classification.
• This approach offers a promising tool for automated detection and diagnosis of COVID-19.
• The system provides a valuable contribution to combating the pandemic through advanced medical imaging analysis.