A deep learning approach for classification of COVID and pneumonia using DenseNet-201

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Deep Learning

Background:

• Chest X-rays are crucial for diagnosing respiratory illnesses like COVID-19 and pneumonia.
• Accurate and timely diagnosis is essential for effective patient management and treatment.
• Existing diagnostic methods may have limitations in speed and accuracy.

Purpose of the Study:

• To develop and evaluate a deep learning model for automated detection of COVID-19 and pneumonia using chest X-ray images.
• To create a user-friendly framework that assists radiologists in disease classification.
• To ensure compliance with Health Insurance Portability and Accountability Act (HIPAA) for protected health information security.

Main Methods:

• Utilized the DenseNet201 architecture, a deep learning approach, for image analysis.
• Employed transfer learning techniques to enhance detection capabilities.
• Developed a Graphical User Interface (GUI) for seamless image uploading and processing.
• Trained and validated the model on a dataset of chest X-ray images from Kaggle.

Main Results:

• Achieved a diagnostic accuracy of 99.1%.
• Demonstrated high sensitivity (98.5%) and specificity (98.95%) in identifying COVID-19 and pneumonia.
• The framework provides classification results to aid radiologist verification.

Conclusions:

• The proposed deep learning framework effectively detects COVID-19 and pneumonia from chest X-rays.
• The system offers a valuable tool for medical professionals, improving diagnostic efficiency and patient care.
• This bio-medical innovation shows potential for future applications in medical image analysis.