Factors for clinical accuracy of a CT-based active infrared optical navigation system

LIU Ya-jun; ZHAO Jing-wei; FAN Ming-xing; LYU Yan-wei; TIAN Wei

doi:10.3969/j.issn.1002-266X.2015.48.004

Abstract：Objective To investigate the clinical factors affecting the accuracy of a CT-based active infrared optical navigation system .Methods A Sawbone model of human lumbar ( L1-L5 ) was fixed prone on bed .Point registration process was done using small titanium points at different places of L 2 of the model .Clinical accuracy was calculated by measuring coordinates of titanium points on spinal process and transverse processes using the navigation software .We inves-tigated the clinical accuracy of the navigation system under several different clinical statuses including the clinical status of different postures of the surgical bed , different distances (100, 125, 150 and 175) between the camera and the model , and shadowless lamps .Results Compared with the original position , the clinical accuracy may become worse when the bed was elevated, descended, changed to head-high or right-tilt position (all P<0.05).The clinical accuracy was not af-fected when the patients had head-down position and the left-leaning position (all P>0.05).The clinical accuracy of the registration vertebra ( L2 ) was not affected by changing the distance between the camera and the patient tracer ( all P>0.05).For the other vertebrae (L1, L3, L4 and L5), the nearer the distance between the camera and the tracer was , the better the clinical accuracy would be (all P<0.01).The clinical accuracy of L1 and L5 may become worse when we opened the shadowless lamp as compared with that of closing shadowless lamp (all P<0.05), but the clinical accuracy of L2, L3 and L4 had no obvious changes (all P>0.05).Conclusion The clinical accuracy of active infrared optical navi-gation system may be influenced by the changes of the surgical bed , long distances between the camera and the model and opening the shadowless lamps .