Needle deflection estimation using fusion of electromagnetic trackers

Area of Science:

• Medical Robotics
• Surgical Navigation
• Biomedical Engineering

Background:

• Accurate needle placement is crucial for minimally invasive procedures.
• Needle deflection during insertion into soft tissues introduces significant targeting errors.
• Existing methods for estimating needle bending have limitations in real-time accuracy.

Purpose of the Study:

• To develop and validate a robust method for real-time needle deflection estimation.
• To improve the accuracy of needle tip positioning during insertion into deformable tissues.
• To compensate for needle bending in image-guided interventions like prostate brachytherapy.

Main Methods:

• Fusion of data from two electromagnetic (EM) trackers with a kinematic needle deflection model.
• Utilizing a Kalman filter (KF) to integrate sensor information from base and tip EM trackers.
• Real-time estimation of needle tip position by compensating for deflection.

Main Results:

• Reduced needle tip mean estimation error from 2.39 mm to 0.31 mm at a 15 cm insertion depth.
• Demonstrated significant improvement in positioning accuracy compared to methods relying on a single EM tracker.
• Validated the method's effectiveness in simulated prostate cancer brachytherapy needle insertion.

Conclusions:

• The proposed method effectively compensates for needle bending in deformable tissues.
• Fusion of multiple EM tracker data with a KF provides a reliable and accurate needle tip position estimate.
• This technique offers a clinically practical solution for enhancing precision in needle-guided interventions.