Theoretical-practical studies for implementation and optimization of the Dual Energy Mammography (DEM) for assisting on early diagnosis of breast cancer

Abstract

The Dual Energy Mammography (DEM) diagnostic technique is accomplished through images generated by two X ray beams with different energies, which are combined to provide images with complementary information from conventional mammography. Beyond microcalcification thickness and glandular fraction, the DEM can also map the density and effective atomic number distributions of the tissue. Those characteristics proportionate the DEM the potential of supplying images with high radiologicals sensibility and specificity, mainly for cases of denser breasts, what reduce the number of false-positives. Nowadays, the dual energy mammography technique still presents some open issues as the pre-processing algorithm study (specially for scatter corrections and noise reduction), the best way to combine the corrected images generated by the radiation beams (post-processing algorithm), the adjustment of the used energies, the use of mono or polyenergetic beams, the absorbed dose and others. Since the DEM can be easily and promptly implemented in digital mammographs, the study and optimization of this imaging technique has the capability to significantly aid neoplastic breast diagnosis. In this way the general objective of this work is to address the referred questions through theoretical/practical studies to implement new pre and post-processing algorithms which allow: (i) the determination of several breast tissue parameters, (ii) the morphological breast characteristics mapping, (iii) as well as the choice of the technical parameters (kVp, anode/filter combination, mAs) which provide the best informative content with the lowest absorbed dose