Virtual Healthcare Center for COVID-19 Patient Detection Based on Artificial Intelligence Approaches

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Infectious Diseases

Background:

• The COVID-19 pandemic, originating in late 2019, has caused global health and economic crises.
• Resource-limited regions face exacerbated risks due to financial and health disparities.
• Increasing demand for medical supplies, including blood test components, strains global supply chains.

Purpose of the Study:

• To introduce a novel COVID-19 disease diagnosis (CDD) tool leveraging deep learning.
• To provide an automated system for symptom checking and COVID-19 detection.
• To enhance pandemic response through efficient and accurate diagnostic capabilities.

Main Methods:

• The CDD tool employs a two-step deep learning approach.
• Step 1: Automatic patient symptom analysis to predict infection probability.
• Step 2: Automated diagnosis using X-ray or CT scans based on predicted probability.

Main Results:

• The proposed CDD scheme demonstrates high efficiency in diagnosing COVID-19.
• Achieved an accuracy rate exceeding 90% in diagnostic performance.
• Outperformed existing diagnostic schemes in numerical evaluations.

Conclusions:

• Deep learning techniques applied to medical images offer a promising avenue for pandemic control.
• The CDD tool provides an effective automated solution for COVID-19 diagnosis.
• The high accuracy of the CDD tool can support public health efforts in managing the pandemic.