Efficient needle guidance: multi-camera augmented reality navigation without patient-specific calibration.

Area of Science:

• Medical Imaging
• Surgical Navigation
• Augmented Reality

Background:

• Augmented reality (AR) systems enhance surgical navigation but often rely on patient-specific markers.
• These markers disrupt workflows and require extensive preoperative calibration, limiting efficiency and precision.

Purpose of the Study:

• To develop a novel multi-camera AR navigation system that eliminates the need for patient-specific markers.
• To improve surgical navigation accuracy and streamline workflows in needle-based procedures.

Main Methods:

• Developed a marker-free AR system using ceiling-mounted markers mapped to fixed imaging devices.
• Integrated a hierarchical optimization framework for marker mapping and multi-camera calibration.
• Employed deep learning for marker detection and vision-based pose compensation for patient movement.

Main Results:

• Achieved an average puncture accuracy of 3.72 ± 1.21 mm in phantom and simulated clinical experiments.
• Reduced needle placement time by 20 seconds compared to marker-based methods.
• Demonstrated effective correction of patient movement errors with low positional and angular deviations.

Conclusions:

• The marker-free AR guidance system streamlines surgical workflows and enhances needle navigation accuracy.
• The system's simplicity, cost-effectiveness, and adaptability make it suitable for diverse clinical settings.
• Potential for improved precision, reduced procedural time, and better patient outcomes in needle-based interventions.