A Simple Algorithm for Increased Helical Pitch in Cone-Beam CT

A goal of cone-beam CT is to decrease the scan time needed to image a volume of the patient. Four things are required to meet this goal: (1) a detector with multiple rows of sensors, (2) higher bandwidth data acquisition system and image processor to handle the increased data rates and need for fast reconstruction, (3) high helical pitch, (4) a practical reconstruction algorithm that lends itself to fast reconstruction times. Within the last several years, CTsystems with first two, then four row detector arrays have been introduced to the marketplace [1]-[5]. The four row arrays are actually combinations of finer pitched detectors. Table I gives a comparison of the detector designs for the Toshiba Aquilion Multi, the LightSpeed QX/i from General Electric Medical Systems, the Siemens Volume Zoom and Marconi Mx8000. All the major manufacturers have announced plans for an output of 8 and/or 16 slices, which requires a new or modified detector design from Siemens and Marconi. Toshiba has presented images taken on a 256row prototype detector array [6], [7]. Thus, detector development for wider cone-angles is underway and will lead to faster scan times. The manufacturers are working on the second requirement as computer electronics improve. This presentation looks at the last two requirements. We present a scheme for high helical pitch while maintaining a practical although approximate reconstruction algorithm.