A nonlinear model for predicting radiographic contrast.

A computerized model has been developed which permits ideal maximum radiographic contrast, associated with clinically meaningful changes in x-ray attenuation for any combination of tissues, to be predicted from the spectral energy of the primary beam and the characteristics of the film. Results suggest that these factors influence available contrast in ways which depend on the specific diagnostic task to be accomplished. Hence, it appears that no single technique is optimal for all tissue configurations. Although contrast computed in this way does not take into account other factors of diagnostic interest, such as image resolution and context-dependent cues to lesion identity, it is an independent factor which can profoundly influence the detectability of specific changes in tissue attenuation and, as such, it may be considered a reasonable basis for predicting diagnostic performance.