Comparing the performance of mammographic enhancement algorithms: A preference study

OBJECTIVE. The objective of this study was to compare the performance of fo ur image enhancement algorithms on secondarily digitized (i.e., digitized f rom film) mammograms containing masses and microcalcifications of known pat hology in a clinical soft-copy display setting. MATERIALS FIND METHODS. Four different image processing algorithms (adaptiv e unsharp masking, contrast-limited adaptive histogram equalization, adapti ve neighborhood contrast enhancement, and wavelet-based enhancement) were a pplied to one image of secondarily digitized mammograms of forty cases (10 each of benign and malignant masses and 10 each of benign and malignant mic rocalcifications). The four enhanced images and the one unenhanced image we re displayed randomly across three high-resolution monitors. Four expert ma mmographics ranked the unenhanced and the four enhanced images from 1 (best ) to 5 (worst). RESULTS. For microcalcifications, the adaptive neighborhood contrast enhanc ement algorithm was the most prefered in 49% of the interpretations, the wa velet-based enhancement in 28%, and the unenhanced image in 13%. For masses , the unenhanced image was the most preferred in 58% of cases, followed by the unsharp masking: algorithm (28%). CONCLUSION. Appropriate image enhancement improves the visibility of microc alcifications. Among the different algorithms, the adaptive neighborhood co ntrast enhancement algorithm was preferred most often. For masses, no signi ficant improvement was observed with any of these image pr processing appro aches compared with the unenhanced image. Different image processing approa ches may need to be used, depending on the type of lesion. This study has i mplications for the practice of digital mammography.