XCOVNet: Chest X-ray Image Classification for COVID-19 Early Detection Using Convolutional Neural Networks

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Infectious Diseases

Background:

• The COVID-19 pandemic, caused by SARS-COV-2, has led to a global health crisis with millions of infections and fatalities.
• Current COVID-19 detection methods like RT-PCR can have delays exceeding 48 hours, hindering timely intervention and disease control.
• The rapid spread and severity of COVID-19 necessitate faster and more accurate diagnostic tools.

Purpose of the Study:

• To propose a novel deep learning model, XCOVNet, for the early detection of COVID-19 using chest X-ray images.
• To develop a two-phase classification system to accurately identify COVID-19 infections from radiographic data.
• To address the limitations of current diagnostic methods by offering a potentially quicker detection alternative.

Main Methods:

• A two-phase convolutional neural network (CNN) model named XCOVNet was developed for image classification.
• The first phase involved pre-processing a dataset of 392 chest X-ray images (50% COVID-19 positive, 50% negative).
• The second phase focused on training and optimizing the neural network to classify patients based on their X-ray images.

Main Results:

• The XCOVNet model demonstrated a high classification accuracy of 98.44% in identifying COVID-19 positive cases from chest X-rays.
• The two-phase approach effectively processed and analyzed the X-ray image dataset.
• The model's performance indicates its potential as an effective tool for early COVID-19 detection.

Conclusions:

• XCOVNet offers a promising AI-driven solution for the early detection of COVID-19 using chest X-ray imaging.
• The high accuracy achieved by XCOVNet suggests its utility in complementing or potentially accelerating diagnostic processes.
• Further research and validation are warranted to integrate XCOVNet into clinical practice for improved pandemic response.