Fluoroscopic Navigation for a Surgical Robotic System Including a Continuum Manipulator

Area of Science:

• Medical Robotics
• Surgical Navigation
• Medical Imaging

Background:

• Continuum Manipulators (CM) offer unique advantages in minimally invasive surgery due to their flexibility.
• Accurate real-time navigation is crucial for precise CM positioning, especially in orthopedic procedures.
• Existing navigation systems may have limitations in accuracy or complexity.

Purpose of the Study:

• To develop and validate an image-based navigation solution for a surgical robotic system utilizing a Continuum Manipulator (CM).
• To estimate the CM's shape and pose relative to bone anatomy using only fluoroscopic images.
• To evaluate the system's accuracy in simulation and cadaveric studies for orthopedic applications.

Main Methods:

• Utilized image intensity-based 2D/3D registration for CM pose and shape estimation from fluoroscopic images.
• Implemented a learning-based framework for automatic CM detection and landmark identification in X-ray images.
• Employed a modified hand-eye calibration method for numerical optimization of the hand-eye matrix during registration.

Main Results:

• Achieved a mean error of 1.10±0.72 mm in simulation for CM tip positioning relative to a target entry point on the femur.
• Demonstrated a mean CM tip position error of 2.86±0.80 mm in cadaveric experiments after registration and repositioning.
• The learning-based framework effectively detected CMs and identified landmarks for registration initialization and regularization.

Conclusions:

• The proposed fluoroscopic navigation system is feasible for guiding Continuum Manipulators (CM) in orthopedic surgery.
• The image-based navigation solution provides accurate CM shape and pose estimation using readily available C-arm fluoroscopy.
• The system shows promise for enhancing precision and safety in robot-assisted orthopedic procedures.