Combining the advantages of AlexNet convolutional deep neural network optimized with anopheles search algorithm based...

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Computational Biology

Background:

• Accurate and efficient COVID-19 detection is crucial for public health.
• Existing diagnostic methods face challenges in speed and accuracy.
• Deep learning offers potential for automated medical image analysis.

Purpose of the Study:

• To develop and evaluate a novel COVID-19 detection and classification framework.
• To enhance image preprocessing techniques for improved feature extraction.
• To optimize a deep neural network model for COVID-19 diagnosis.

Main Methods:

• Implementation of an AlexNet convolutional deep neural network optimized by the Anopheles Search Algorithm (ASA).
• Preprocessing of COVID-19 dataset images using Fuzzy Gray Level Difference Histogram Equalization (FGLHE) and fuzzy stacking.
• Classification of extracted features using a Random Forest (RF) classifier (ADCNN-ASA-RFC).

Main Results:

• The proposed ADCNN-ASA-RFC framework demonstrated significant improvements in accuracy, specificity, and sensitivity compared to existing algorithms.
• Achieved higher accuracy (up to 91.66%), specificity (up to 79.13%), and sensitivity (up to 77.13%) in COVID-19 diagnosis.
• The system efficiently identified optimal solutions for accurate COVID-19 diagnosis.

Conclusions:

• The developed framework provides an efficient and accurate method for COVID-19 diagnosis.
• The integration of advanced image preprocessing and optimized deep learning models enhances diagnostic performance.
• This approach shows promise for real-world clinical application in COVID-19 detection.