An experimental study of the relationship between image quality and spatial resolution for the gamma camera.

Many centres make regular measures of gamma camera performance as part of routine quality assurance programmes. Such measures may detect gradual deterioration in the camera but provide no basis on which to decide when corrective action should be taken. It is necessary to know when changes in camera performance are significant in terms of perceived image quality. In this work, one index of performance, the full width half maximum (FWHM) of the line spread function, was degraded in a controlled manner and the ability of observers to detect this change in the images produced was examined. Both simple and complex objects were investigated. A suitable decision criterion was suggested which indicated that changes of about 0.3 mm in the FWHM of the camera could be detected in the image. This figure was essentially independent of the complexity of the image, the initial FWHM of the camera and, above 300 k, the number of counts in the image. The way in which this type of experiment can help to establish a rational basis for gamma camera quality assurance programmes is discussed.