A filtering algorithm based on SPECT lesion detectability potential

This paper describes an effective filtering algorithm which employs a priori information of SPECT lesion detectability potential for the filtering of degraded projection image prior to backprojection. For a given set of physical conditions, the authors determined m minimum detectable lesion sizes (MDLSs) from SPECT lesion detectability contrast-detail curves, uniformly-choosing m lesion contrasts between 0.0 and 1.0. Related to the authors´ previous work, an optimal estimate of given projection image was attempted as a weighted combination of the subimages from the m optimal filters whose design is focused on maximizing the local S/N ratios of the determined MDLS-lesions, and the weighting on each subimage was controlled by the difference between the subimage and a maximum-resolution-recovered projection image. The results of the authors´ simulation study showed the adaptive effect of this algorithm and its good performance in effectively reducing image noise and preserving/enhancing lesion structures