AIBAN-AI Bronchial Detection and Airway Navigation System

Area of Science:

• Medical Imaging
• Artificial Intelligence in Medicine
• Pulmonology

Background:

• Accurate navigation in the lungs is crucial for bronchoscopy but challenging due to complex airway anatomy.
• Artificial intelligence (AI) offers potential to improve precision and safety through real-time landmark detection and navigation guidance.

Purpose of the Study:

• To develop and evaluate an AI system for enhanced bronchoscopy navigation.
• To assess the system's accuracy in airway detection, localization, and guidance.

Main Methods:

• Developed an AI Bronchial detection and Airway Navigation (AIBAN) system integrating airway lumen detection, anatomic localization, and navigation guidance.
• Utilized a YOLOv8-based module for real-time landmark detection and a hierarchical anatomic graph for bronchoscope localization.
• Implemented a navigation module to recommend airway pathways and provide directional cues.

Main Results:

• The AIBAN system achieved 91.1% accuracy in identifying the bronchoscope's location.
• 92.6% of predictions were within one airway branch, and 94.5% within two branches of the true location.
• The navigation guidance correctly followed the anatomic pathway in 96.2% of test videos.

Conclusions:

• AIBAN enables accurate airway detection, localization, and navigation guidance in a simulated setting.
• The system provides reliable guidance and precise spatial positioning, showing promise for medical training.
• AIBAN has potential for future clinical implementation in bronchoscopy procedures.