State of art and optimization perspectives for breast imaging

Breast screening with mammography is the most effective method of detecting early-stage breast cancer and reducing related mortality. Among the intrinsic limits of mammography, in terms of clinical performance, the overlapping of normal and pathological tissues is one of the most influential. Some new techniques as Digital Breast Tomosynthesis (DBT) is expected to overcome this limitation by providing a quasi-three-dimensional (3D) image that could lead to improve the accuracy of mammography. Another way to increase accuracy and sensitivity is represented by a double exposure of the patient before and after intravenous injection of contrast media, this technique is called Contrast-enhanced digital (or spectral) mammography (CEDM, CESM). Furthermore, highly specialized software has been developed which is able to detect suspicious mammographic findings. This technology is very interesting especially in the screening field, in fact there are multiple ongoing studies evaluating the use of Artificial Intelligence (AI) as a second reader. To date, screening mammography is the only imaging modality that has proven to significantly lower breast cancer mortality. Tomosynthesis demonstrated excellent sensitivity and specificity but the technique did not meet the expectations given the risk of over diagnosis as well as the lack of reduction in the number of interval breast cancer. CESM could in some cases serve as an alternative imaging tool to MRI. AI, seems to be competing with the breast radiologist and its use as a second reader in breast screening programs is already being proposed