Unlocking the Power of 3D Convolutional Neural Networks for COVID-19 Detection: A Comprehensive Review

Area of Science:

• Artificial Intelligence in Medical Imaging
• Deep Learning for Disease Detection

Background:

• Three-dimensional convolutional neural networks (3D CNNs) are increasingly vital for analyzing medical images.
• Advancements in imaging technology have enabled sophisticated AI applications for COVID-19 diagnostics.

Purpose of the Study:

• To conduct a comprehensive review of 3D CNN algorithms for COVID-19 segmentation and classification.
• To evaluate the efficacy of various 3D CNN methodologies across different medical imaging modalities.

Main Methods:

• Systematic review of recent advancements in 3D CNN methodologies for COVID-19 detection.
• Screening and analysis of 60 research papers from academic repositories like Springer and Elsevier.
• Evaluation based on specific criteria, focusing on network architectures and algorithms.

Main Results:

• 3D CNNs demonstrate high accuracy and rapid detection capabilities for COVID-19.
• Identified trends show diverse network architectures for COVID-19 detection compared to other diseases.
• Key findings detail strengths, limitations, and future research directions for 3D CNN applications.

Conclusions:

• 3D CNN algorithms show significant potential for advancing medical image segmentation and classification.
• These networks can enhance COVID-19 detection and management, leading to improved healthcare outcomes.
• Further research into 3D CNNs is crucial for optimizing clinical diagnosis and treatment strategies.