Multi-COVID-Net: Multi-objective optimized network for COVID-19 diagnosis from chest X-ray images

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Radiology

Background:

• The global spread of Coronavirus Disease 2019 (COVID-19) has strained healthcare systems worldwide.
• Efficient screening of hospitalized patients using chest radiography (CXR) is crucial for managing the pandemic.
• Limited availability of COVID-19 testing kits necessitates rapid, automated diagnostic alternatives.

Purpose of the Study:

• To propose a fully automated deep learning (DL) architecture for diagnosing COVID-19 using chest X-ray images.
• To develop an efficient method for identifying COVID-19, pneumonia, and non-COVID-19 cases from CXR.
• To enhance diagnostic capabilities in resource-limited settings during the COVID-19 pandemic.

Main Methods:

• A two-step DL architecture was developed, comprising feature extraction and classification stages.
• The Multi-Objective Grasshopper Optimization Algorithm (MOGOA) was employed to optimize the DL network layers, creating the Multi-COVID-Net.
• The model was trained and validated using publicly available chest X-ray datasets.

Main Results:

• The Multi-COVID-Net demonstrated high performance in automatically classifying chest X-ray images.
• The model accurately distinguished between COVID-19, pneumonia, and non-COVID-19 cases.
• Comparative analysis showed superior performance of Multi-COVID-Net over existing state-of-the-art methods.

Conclusions:

• The proposed Multi-COVID-Net offers a promising automated solution for COVID-19 diagnosis via CXR.
• This deep learning approach can serve as a valuable tool for rapid screening and early detection.
• The study highlights the potential of AI in augmenting diagnostic capacity during public health crises.