On-demand calibration and evaluation for electromagnetically tracked laparoscope in augmented reality visualization

Area of Science:

• Medical Imaging
• Surgical Technology
• Augmented Reality

Background:

• Current laparoscopic augmented reality systems require lengthy camera calibration procedures.
• Traditional methods necessitate multiple images of a checkerboard pattern, hindering real-time operating room (OR) use.
• There is a need for a fast, on-demand calibration solution for laparoscopes in the OR.

Purpose of the Study:

• To develop a rapid calibration method for electromagnetically (EM) tracked laparoscopes.
• To enable on-demand laparoscope calibration directly within the operating room.
• To reduce the time and complexity associated with current camera calibration techniques.

Main Methods:

• A mechanical tracking mount was designed to position an EM sensor on a conventional laparoscope.
• A tool named fCalib was developed for calibrating intrinsic and extrinsic camera parameters using a single image of a special target pattern.
• A tube phantom was integrated for rapid evaluation of calibration results in the OR.

Main Results:

• The fCalib method demonstrated comparable spatial target registration error to the OpenCV method in laboratory settings.
• Re-projection accuracy was evaluated between EM and optical tracking-based fCalib in a clinical setting.
• The fCalib method's computational time averaged 14.0 seconds, with a range of 3.5 to 22.7 seconds.

Conclusions:

• A prototype for fast calibration and evaluation of EM-tracked laparoscopes was successfully developed and validated.
• The developed calibration method is accurate, fast, and suitable for on-demand use in the OR.
• The fCalib method offers a practical solution for improving augmented reality systems in laparoscopic surgery.