Estimating the Thickness of a Compressed Breast
Many recent mammography systems have built-in analogue or digital thickness meters but their accuracy and precision are currently wanting . In clinical practice, most existing systems do not have such indicators. Burch and Law  have reported putting lead markers on the top compression plate and then using the magnification of the separation between these markers to estimate breast thickness, a technique replicated by Smith et al. . However, although their results appear promising, with an average accuracy of the order of 2mm or so, and with a maximum error of 4.9mm, analysis of the projective equations reveals that the method is far from stable: a change in the projected measure of just 1mm produces a change in breast thickness estimate of at least 2.4mm. Furthermore, in many cases, the lead markers were reported to be not visible on the film due to scatter and sometimes were projected onto other lead markers or the breast. Moreover, this technique does not allow for retrospective estimation of breast thicknesses which is crucial for estimating; the accumulated radiation dose.
KeywordsCalibration Data Roughness Measure Average Absolute Difference Average Absolute Error Breast Thickness
Unable to display preview. Download preview PDF.