Development of virtual fluoroscopy using mobile C-arm

Abstract

C-arm is used in many medical operations. C-arm can provide intra-operative fluoroscopic image but causes cumulative radiation exposure to the surgeon and operating personnel. Virtual fluoroscopy is introduced to solve this problem. Virtual fluoroscopy superimposes the surgical instrument onto the fluoroscopy in real time without continuous imaging. The location and orientation of surgical instrument are tracked and virtually projected on the image by software. Virtual fluoroscopy reduces the radiation exposure to the surgeon and informs the surgeon of the 3D location of tool by multi-planar guidance. To realize the virtual fluoroscopy several steps are needed. The distortion of C-arm must be removed and the geometric parameters must be estimated. In real time the location and orientation of tool should be tracked. In this study the new distortion correction method is proposed. The new distortion correction method requires less effort and time to calibrate the distortion of C-arm image. Through experiment it is turned out that this method provides acceptable accuracy to the virtual fluoroscopy. The geometric calibration finds the geometric parameters using the calibration phantom. The parameters are calculated by direct linear transformation. This method finds geometric parameters in real time without expensive opto-electronic device such as $Optotrak^？$. In this study the $Microscribe^？$ is used to measure the position and orientation of tool in real time. To evaluate the developed system the position error and the angle error are estimated at 10 cases. The evaluation results show that the mean position error is 1.59mm and the mean angle error is 2.48˚