摘要

Background In clinical environments, field distortion can cause significant electromagnetic tracking errors. Therefore, dynamic calibration of electromagnetic tracking systems is essential to compensate for measurement errors. Methods It is proposed to integrate the motion model of the tracked instrument with redundant EM sensor observations and to apply a simultaneous localization and mapping algorithm in order to accurately estimate the pose of the instrument and create a map of the field distortion in real-time. Experiments were conducted in the presence of ferromagnetic and electrically-conductive field distorting objects and results compared with those of a conventional sensor fusion approach. Results The proposed method reduced the tracking error from 3.94 +/- 1.61 mm to 1.82 +/- 0.62mmin the presence of steel, and from 0.31 +/- 0.22mmto 0.11 +/- 0.14mm in the presence of aluminum. Conclusions With reduced tracking error and independence from external tracking devices or pre-operative calibrations, the approach is promising for reliable EM navigation in various clinical procedures.

  • 出版日期2016-6