Deep Transfer Learning-Based Framework for COVID-19 Diagnosis Using Chest CT Scans and Clinical Information

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Infectious Diseases

Background:

• COVID-19, a global pandemic, requires rapid and accurate diagnostic methods.
• Conventional tests like PCR have significant turnaround times.
• Chest CT scans show promise but have limitations when used alone.

Purpose of the Study:

• To develop a deep learning framework for fast and accurate COVID-19 diagnosis.
• To integrate clinical features with chest CT scan analysis.
• To improve diagnostic capabilities beyond conventional methods.

Main Methods:

• Utilized Artificial Neural Networks (ANN) to analyze clinical data for infection probability.
• Employed a deep learning model for classifying chest CT scan images.
• Integrated clinical information with CT scan data for a comprehensive diagnostic approach.

Main Results:

• The deep learning model achieved 99% accuracy in classifying chest CT scans.
• The framework effectively combines clinical data and imaging for diagnosis.
• Demonstrated a faster diagnostic process compared to traditional methods.

Conclusions:

• Deep learning integration of clinical data and CT scans offers a highly accurate COVID-19 diagnostic tool.
• This approach can significantly expedite the diagnosis of COVID-19.
• Highlights the potential of AI in managing infectious disease outbreaks.