Automated Detection of COVID-19 from Multimodal Imaging Data Using Optimized Convolutional Neural Network Model.

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Infectious Diseases

Background:

• COVID-19 infections rapidly increased during the pandemic, necessitating accurate and timely diagnosis.
• Early identification of COVID-19 is crucial for disease control and treatment planning.
• Advanced diagnostic methods are required to meet the growing need for widespread COVID-19 detection.

Purpose of the Study:

• To develop and validate an optimized deep learning system for accurate COVID-19 detection.
• To enhance convolutional neural network (CNN) models using multi-modal imaging data fusion.
• To improve COVID-19 diagnostic accuracy through feature fusion and optimization techniques.

Main Methods:

• Comparative analysis of various convolutional neural network (CNN) models to select an optimal architecture.
• Enhancement of the selected CNN model via feature fusion from multi-modal imaging datasets.
• Application of the Jaya optimization technique for optimal feature vector merging.
• Classification of samples using a Support Vector Machine (SVM) classifier.

Main Results:

• The proposed fine-tuned system achieved a high accuracy rate of 98.7% on a dataset of 10,000 samples.
• The optimized multi-modal approach significantly outperformed existing state-of-the-art network models.
• The system demonstrated superior performance in distinguishing between COVID-19 positive and negative cases.

Conclusions:

• The developed system offers a highly accurate and efficient method for COVID-19 diagnosis.
• Multi-modal data fusion combined with optimization techniques enhances deep learning model performance for disease detection.
• This approach holds promise for improving early detection and management of COVID-19.