COVI³D: Automatic COVID-19 CT Image-Based Classification and Visualization Platform Utilizing Virtual and Augmented...

Area of Science:

• Biomedical image analysis
• Medical imaging informatics
• Virtual and Augmented Reality applications in healthcare

Background:

• Augmented reality (AR) and virtual reality (VR) show promise in biomedical image analysis but lack automation for COVID-19 classification.
• Computed tomography (CT) scans are crucial for COVID-19 research and clinical use, yet accessible datasets for Algerian patients are limited.
• Existing methods struggle with automated segmentation and classification of infection regions in lung CT images.

Purpose of the Study:

• To design an automated AR and VR platform for analyzing, classifying, and visualizing severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) pandemic data.
• To develop a novel system for automatic CT image segmentation, localization, and volume measurement of infected lung regions.
• To create a user-friendly 3D interface for AR/VR, incorporating patient and medical staff feedback for improved engagement and scalability.

Main Methods:

• Implementation of an automatic CT image segmentation and localization system for infected lung regions.
• Elaboration of volume measurements and a lung voxel-based classification procedure.
• Development of an interactive AR and VR 3D interface, integrating user feedback.

Main Results:

• The developed AR/VR platform demonstrated superior efficiency in CT image classification compared to state-of-the-art methods.
• The system was validated using a dataset of 500 COVID-19 positive CT scans from Algerian patients.
• Computer simulations confirmed the effectiveness of the automated segmentation, classification, and visualization techniques.

Conclusions:

• The proposed AR and VR platform offers an automated solution for COVID-19 data analysis, addressing limitations in current approaches.
• The system facilitates more accurate and rapid diagnosis and treatment planning for COVID-19 patients.
• Integration of user feedback enhanced the platform's scalability, engagement, and overall evaluation.