A simulator for advanced analysis of a 5-DOF EM tracking systems in use for image-guided surgery.

Abstract

To assess the accuracy of medical electromagnetic tracking systems, reference positioning systems are generally required. Errors are unavoidable in such systems, and despite how tiny they may be, prevent the ground truth from being known. In this work, a simulator was developed and used to analyze the theoretical system performances in electromagnetic tracking. To simulate the entire tracking process, the magnetic dipole model, Faraday's law, and a mathematical optimization algorithm are applied. With the simulator, we optimized the spatial placement of the transmitter coils, analyzed the tracking accuracy by applying stochastic and optimized coil placement. Additionally, the performance of the calibration of transmitter coils' measurement error and Kalman filtering was tested. The results show that, after optimizing the spatial arrangement of the transmitter coils, the tracking accuracy is significantly improved to a much higher level compared with applying statistical arrangement. The measurement errors of the transmitter coils' positions and orientations can be totally rectified by the developed calibration algorithm when no noises are introduced. The Kalman filter reduces the sensor jitter errors caused by noise, which potentially allows the EM tracking system to reach a larger volume of interest. We proposed a simulator for advanced analysis in electromagnetic tracking without hardware requirements. Grounded on this, we performed an optimization of the spatial arrangement of the transmitter coils to improve the tracking accuracy further. The performances of the calibration algorithm and Kalman filtering were also evaluated. The developed simulator can also be applied for other analysis in electromagnetic tracking.