Approximation of the center and radius of C-arm images for automatic on-line camera calibration

Online C-arm calibration procedure is an important issue for the use of this image modality intraoperatively in the surgical rooms. Automatic detection of the images center and radius is one of the key step for camera calibration. In the current approaches, this is measured manually, and the geometrical measurements are supplied as parameters to the calibration algorithm. In this paper, a method fully automatic which detects the minimal circle containing the pertinent information is presented. It is based on a combination of image processing and computational geometry. The method has been tested on real patient data, acquired with different scanners, in various hospitals. It was shown to be reliable and fast. It is considered to be part of a commercial C-arm-based navigation system.