Diagnosis of COVID-19 from Multimodal Imaging Data Using Optimized Deep Learning Techniques.

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Infectious Disease Diagnostics

Background:

• The COVID-19 pandemic significantly impacted global health and healthcare systems.
• Early detection of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) is crucial for disease containment and public well-being.
• Deep learning, particularly Convolutional Neural Networks (CNNs), shows promise in analyzing medical images for disease detection.

Purpose of the Study:

• To develop and evaluate a deep learning classification method for detecting COVID-19 using chest X-ray and CT scan images.
• To compare the performance of various CNN models for COVID-19 detection.
• To determine the optimal model and imaging modality for accurate and efficient COVID-19 screening.

Main Methods:

• Utilized deep learning models, including VGG-19, ResNet-50, Inception v3, and Xception, for image classification.
• Trained and optimized CNN models using pre-processed chest X-ray and CT scan images from the Kaggle repository.
• Evaluated and compared model performance based on detection accuracy.

Main Results:

• The fine-tuned VGG-19 model achieved high accuracy in detecting COVID-19.
• VGG-19 demonstrated 94.17% accuracy for chest X-rays and 93% accuracy for CT scans.
• Chest X-rays yielded superior detection accuracy compared to CT scans when using the VGG-19 model.

Conclusions:

• The VGG-19 model is highly suitable for detecting COVID-19 from medical images.
• Chest X-rays are a more accurate and potentially cost-effective imaging modality for COVID-19 screening compared to CT scans.
• Deep learning approaches offer a reliable method for rapid COVID-19 detection, aiding healthcare professionals and reducing exposure risk.