Design of an Accurate Near Infrared Optical Tracking System in Surgical Navigation

Near infrared optical tracking system (NIOTS) is widely used in biomedical application, especially in surgical navigation. It largely determines the accuracy of tracking surgical tools. So we design an NIOTS and propose the corresponding algorithms to make it work well. Two near infrared cameras (NICs) are constructed in this NIOTS and a near infrared calibration board is designed to calibrate them. The NIOTS is then calibrated using the calibration results of NICs. After the image points of markers on surgical tools are obtained, a two-step stereo matching algorithm is proposed to match the subpixel points in the left and right images. Then the three dimensional coordinates of markers are computed and the surgical tools are tracked in real time. The experimental results show the accuracy of system calibration, stereo matching, and tracking. The average tracking errors of three markers in 150 positions are 0.0090 mm, 0.0071 mm, and 0.0060 mm, which demonstrate the designed NIOTS can accurately track surgical tools.