Marker-to-Object Calibration Using Landmark Touch.

Area of Science:

• Medical Imaging
• Robotics
• Biomedical Engineering

Background:

• Fiducial markers are crucial for medical instrument tracking in image-guided interventions.
• Precise marker placement and accurate marker-to-object transformation are challenging, especially for non-standard geometries.
• Existing methods struggle with accuracy and robustness in complex scenarios.

Purpose of the Study:

• To introduce a novel calibration method for accurate marker-to-object transformation in image-guided interventions.
• To improve the robustness and translational accuracy of fiducial marker calibration.
• To validate the method's effectiveness across different tracking modalities and clinical applications.

Main Methods:

• A novel calibration method involving sequential touching of object landmarks with a fixed tip.
• Implementation of an inverse sample consensus filter to mitigate measurement outliers.
• Validation through simulations and experiments using electromagnetic tracking and magnetic resonance imaging systems.

Main Results:

• Demonstrated superior translational and rotational accuracy compared to conventional methods.
• Achieved translational errors of 0.61 ± 0.11 mm (electromagnetic tracking) and 0.60 mm (MRI).
• Showcased sub-millimeter calibration accuracy in diverse scenarios, including a robotic surgery use case.

Conclusions:

• The proposed calibration method offers high accuracy and robustness for fiducial marker placement.
• It is effective across different tracking modalities and suitable for integration into image-guided workflows.
• The method holds significant potential for advancing research and clinical applications in image-guided interventions.