Image processing algorithms for digital mammography: A pictorial essay

Digital mammography systems allow manipulation of fine differences in image contrast by means of image processing algorithms. Different display algori thms have advantages and disadvantages for the specific tasks required in b reast imaging-diagnosis and screening. Manual intensity windowing can produ ce digital mammograms very similar to standard screen-film mammograms but i s limited by its operator dependence. Histogram-based intensity windowing i mproves the conspicuity of the lesion edge, but there is loss of detail out side the dense parts of the image. Mixture-model intensity windowing enhanc es the visibility of lesion borders against the fatty background, but the m ixed parenchymal densities abutting the lesion may be lost. Contrast-limite d adaptive histogram equalization can also provide subtle edge information but might degrade performance in the screening setting by enhancing the vis ibility of nuisance information. Unsharp masking enhances the sharpness of the borders of mass lesions, but this algorithm may make even an indistinct mass appear more circumscribed. Peripheral equalization displays lesion de tails well and preserves the peripheral information in the surrounding brea st, but there may be flattening of image contrast in the nonperipheral port ions of the image. Trex processing allows visualization of both lesion deta il and breast edge information but reduces image contrast.