SPECT imaging: Experimental determination of the effect of camera head tilt and appropriate quality control protocol

Although it is recognized in SPECT imaging that the camera face shall remain parallel to the axis of rotation there is no evidence of the limits that must be imposed on this parameter for acceptable reconstruction of the data. This study was made with the purpose of establishing limits of camera head tilt based on experiments designed to measure degradation of the image with increasing camera head tilt. The effect of camera head tilt is to place off-axis data in the wrong plane for reconstruction. Points off axis exhibit a pseudomotion along a line parallel to the axis of rotation so that data from one plane are ''seen'' by the reconstruction algorithm to be in different planes, depending on the angle of rotation. The changes in system resolution due to camera head tilt were assessed using the cold spheres in a Jaszczak phantom. The contrast of the spheres was measured for increasing angles of head tilt. Significant degradation occurred at tilt angles as low as one degree. Based on these observations a quality control procedure for camera head tilt was developed. A point source is imaged off-axis and a computer program has been written that is able to give the angle of head tilt based on the pseudomotion of the point source throughout the rotation of the camera head. This method can be used to calibrate the head tilt angle. It is evident that the camera head tilt must be checked for every patient study. Unlike center of rotation offsets, it is not possible to correct the raw data for camera head tilt during reconstruction