A Novel Shape-Prior-Guided Automatic Calibration Method for Free-Hand Three-Dimensional Ultrasonography

Area of Science:

• Medical Imaging
• Robotics
• Computer-Aided Surgery

Background:

• Accurate spatial mapping is essential for ultrasound-guided surgical navigation and 3D imaging.
• Current ultrasound probe calibration methods lack robustness and efficiency due to manual processes and outlier sensitivity.

Purpose of the Study:

• To develop a fully automated and robust ultrasound probe calibration framework.
• To improve accuracy, reproducibility, and efficiency in ultrasound probe calibration.

Main Methods:

• A novel automated calibration framework utilizing geometric priors of an N-wire phantom.
• Robust feature point extraction and a hybrid outlier rejection strategy based on the Random Sample Consensus (RANSAC) algorithm.

Main Results:

• Achieved sub-millimeter accuracy (<0.6 mm) across various imaging depths.
• Calibration process completed in under two minutes with high repeatability.
• Demonstrated robustness to outliers.

Conclusions:

• The proposed framework offers a robust, accurate, and time-efficient solution for ultrasound probe calibration.
• The method shows significant potential for clinical integration in surgical navigation and 3D ultrasound applications.