Adaptive marker-free registration using a multiple point strategy for real-time and robust endoscope electromagnetic...

Area of Science:

• Medical Imaging
• Surgical Navigation
• Biomedical Engineering

Background:

• Computer-assisted endoscopic interventions require accurate registration of pre-clinical images to physical space.
• Electromagnetic navigation is crucial for real-time endoscope tracking during diagnosis and treatment.
• Current methods face challenges with assumptions about endoscope trajectory, particularly in intraluminal organs.

Purpose of the Study:

• To propose an adaptive, marker-free registration method for electromagnetic navigation in endoscopic interventions.
• To address limitations of existing methods by relaxing the assumption of endoscope operation along organ centerlines.
• To enhance the accuracy of spatial alignment between electromagnetic trackers and pre-clinical images.

Main Methods:

• Developed an adaptive strategy generating multiple registration points based on sensor measurements and endoscope tip calibration.
• Implemented a multiple point selection strategy to adaptively choose the optimal registration point.
• The optimal point is selected as the one closest to the hollow organ's centerline.

Main Results:

• Significantly reduced target registration error in static phantom validation from 5.32 mm to 2.59 mm.
• Reduced target registration error in dynamic phantom validation from at least 7.58 mm to 4.71 mm.
• Demonstrated improved accuracy compared to current available registration methods.

Conclusions:

• The proposed adaptive marker-free registration method enhances accuracy in electromagnetic navigation for endoscopic interventions.
• This strategy effectively overcomes the limitations of endoscope trajectory assumptions.
• The findings support improved precision and reliability in computer-assisted surgical procedures.