Deep Ensemble Model for COVID-19 Diagnosis and Classification Using Chest CT Images

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Computational Biology

Background:

• COVID-19 pandemic necessitates rapid diagnostic tools.
• Computed tomography (CT) scans offer precise and quick COVID-19 diagnosis.
• Artificial intelligence (AI), machine learning (ML), and deep learning (DL) show promise in medical image analysis.

Purpose of the Study:

• To design an AI-based ensemble model for COVID-19 detection and classification using CT scans.
• To enhance diagnostic accuracy and efficiency in the face of limited medicinal resources.
• To develop a novel approach for automated COVID-19 classification from medical images.

Main Methods:

• Developed an AI-based Ensemble Model for Detection and Classification (AIEM-DC).
• Applied Gaussian filtering (GF) for image noise reduction and quality enhancement.
• Utilized Shark Optimization Algorithm (SOA) with DL models (RNN, LSTM, GRU) for feature extraction.
• Employed an Improved Bat Algorithm with Multiclass Support Vector Machine (IBA-MSVM) for classification.

Main Results:

• The AIEM-DC technique demonstrated promising classification performance on benchmark CT datasets.
• Achieved high accuracy in detecting and classifying COVID-19 cases.
• Outperformed recent state-of-the-art approaches in COVID-19 classification accuracy.

Conclusions:

• The proposed AIEM-DC model offers an effective and accurate solution for COVID-19 diagnosis using CT scans.
• The ensemble approach with optimized parameters shows significant potential for clinical application.
• This AI-driven method can aid in managing diagnostic challenges during pandemics.