Current breast cancer screening, using X-ray mammography has various drawbacks. These include the use of ionising radiation, the need for breast compression, high cost, and the difficulty in implementing this technology in rural communities. The prototype of a portable screening device is presented; its design is aimed at addressing the aforementioned problems.

Eight machine learning classifiers have been trained to predict the presence or absence of tumour tissue from numerical breast models. Scattered electromagnetic field values have been used as inputs for the classifiers. The performances of the algorithms are presented and discussed, with two of the classifiers achieving metrics comparable to those obtained by current X-Ray mammography modalities.

Further improvements on classification features might be necessary to adapt the proposed approach to clinical practice. To this extent, the use of time-domain information arising from both breasts in the classification process will be considered in the future.