C-arm rotation encoding with accelerometers.

Area of Science:

• Medical Imaging
• Robotics in Medicine
• Sensor Technology

Background:

• Fluoroscopic C-arms are increasingly used in computer-assisted interventions, requiring precise knowledge of their relative poses.
• Current pose-tracking methods (optical cameras, electromagnetic trackers, radiographic fiducials) have significant limitations.
• Accurate C-arm pose estimation is crucial for the success of image-guided interventions.

Purpose of the Study:

• To develop a novel method for recovering the rotational pose of C-arms using accelerometers.
• To address the shortcomings of existing pose-tracking technologies in computer-assisted interventions.
• To demonstrate the feasibility and accuracy of accelerometer-based C-arm pose tracking.

Main Methods:

• Utilized accelerometers to measure tilt angles for C-arm pose recovery.
• Affixed an accelerometer to a C-arm analogue equipped with a webcam to mimic X-ray imaging.
• Developed an angle correction equation (ACE) by comparing accelerometer and webcam pose angle readings.
• Tested linear and polynomial ACEs for tracking primary and secondary angle rotations.

Main Results:

• The accelerometer successfully tracked the C-arm pose in various rotational scenarios.
• Angle correction equations (ACEs) were employed to refine pose tracking accuracy.
• The system achieved tracking accuracy of less than 1.0 degree.
• Demonstrated the effectiveness of accelerometers in sensing C-arm rotation.

Conclusions:

• Accelerometers provide a viable and accurate method for sensing C-arm rotation.
• The developed ACE method enables clinically adequate accuracy for C-arm pose tracking.
• This approach offers a promising solution for improving pose estimation in image-guided interventions.