Patient-Specific Depth Error Correction in Dual-Sensor System for Image-Guided Minimally Invasive Facial Osteotomy

Area of Science:

• Medical Imaging
• Surgical Technology
• Biomedical Engineering

Background:

• Image-guided surgery requires high accuracy for procedures like facial osteotomy.
• Dual-sensor systems with Time-of-Flight (ToF) technology offer real-time tracking but face depth inaccuracies due to facial anatomy.
• Existing systems need patient-specific error correction for optimal performance.

Purpose of the Study:

• To develop and validate a patient-specific method for real-time depth error correction in dual-sensor systems for image-guided facial osteotomy.
• To improve the accuracy of registration and shape sensing in surgical navigation.
• To enhance the reliability and precision of minimally invasive facial bone surgeries.

Main Methods:

• Integration of two RGB-D sensors with Time-of-Flight (ToF) technology and an optical localizer for simultaneous real-time tracking.
• Development of a patient-specific error model using the probe's tip position tracked by the optical localizer.
• Real-time estimation of error model parameters to correct ToF sensor depth inaccuracies.

Main Results:

• Demonstrated an 87% reduction in depth errors across three patient mockups.
• Achieved an average fiducial registration accuracy of 2.33±0.23mm.
• Attained an average shape sensing accuracy of 2.25±0.38mm.

Conclusions:

• Patient-specific depth error correction significantly improves registration and shape sensing accuracy in dual-sensor systems.
• The proposed method enhances the reliability and precision of image-guided facial osteotomies.
• This approach holds potential for improved clinical outcomes and patient safety in minimally invasive facial surgeries.