Endobronchial Ultrasound Image Simulation for Image-Guided Bronchoscopy.

Area of Science:

• Medical Imaging
• Computational Ultrasound
• Pulmonology

Background:

• Accurate lung cancer diagnosis and staging are critical for patient management.
• Current minimally invasive tools like bronchoscopy and endobronchial ultrasound (EBUS) have limitations in guiding lesion localization.
• There is a need for improved methods to assist physicians during EBUS procedures.

Purpose of the Study:

• To develop and evaluate a simulation method for endobronchial ultrasound (EBUS) to assist in the localization of extraluminal lung cancer lesions.
• To enhance the accuracy and effectiveness of EBUS-guided bronchoscopy for lung cancer diagnosis and staging.

Main Methods:

• A simulation method was developed using patient's chest computed-tomography (CT) scans to model ultrasound signal propagation.
• The method involves image preprocessing to create a 2D CT equivalent of the EBUS scan plane, EBUS scan-line computation for image generation, and image post-processing to add realistic imaging effects and artifacts.
• The simulation is compatible with both radial-probe and convex-probe EBUS devices.

Main Results:

• The simulation method produced realistic EBUS images that closely resemble those from a live device.
• The simulation performance favorably compared to an existing ultrasound simulation method.
• The method achieved a high frame rate, exceeding real-time performance (53 frames/sec), and was successfully integrated into an image-guided EBUS bronchoscopy system.

Conclusions:

• The developed EBUS simulation method is effective and practical for planning and real-time guidance during EBUS bronchoscopy.
• This simulation tool has the potential to improve the precision and safety of minimally invasive lung cancer diagnosis and staging procedures.