On the development of an innovative mm-wave imaging system for breast cancer detection

In this Thesis, a complete study of an innovative mm-wave imaging system for breast cancer detection is presented. This work is strongly multidisciplinary and involves several partners both inside and outside the University of Pavia. The main medical partner outside the University is the European Institute of Oncology (Italian acronym, IEO), in Milan. Outside the University of Pavia there are also the Institute dElectronique and Telecomunication de Rennes and the University of Malta; while, inside the University there are the bioengineering laboratory, the civil and architecture department, and the laboratory of bioinformatics mathematical modelling and synthetic biology. In addition, this project was originally funded by the Italian Association for Cancer Research (Italian acronym, AIRC), and it is currently funded by the University of Pavia funding, Blue Sky Research project MULTIWAVE. This work can be divided in three main branches: the dielectric characterization of biological tissues up to 50 GHz (in particular, human breast ex-vivo samples); the realization of realistic breast phantoms for the test of the prototype on phantoms in a controlled environment; and the test of the mm-wave imaging prototype on realistic breast phantoms. In particular, in the first chapter of the Thesis, three experimental campaigns both on animal and human tissues (ex-vivo, in-vivo and in-loco) in the frequency range [0.5-50] GHz have been deeply investigated, as the dielectric properties describe the way in which the electromagnetic fields interact with, and propagate within, a tissue, and therefore the exact knowledge of these properties is the starting point for all possible microwave-based technologies. In the second chapter, several recipes for the preparation of realistic breast phantoms capable of mimicking the dielectric properties of ex-vivo breast tissues up to 50 GHz have been proposed. In particular, two macro-categories of phantom are presented: one based on the use of gelatin as a solidifying agent and on dishwashing liquid as a surfactant to mimic the average dielectric properties of the different categories of healthy and neoplastic breast tissues, and one on the use of waste-oil hardener as a solidifying agent and Polysorbate80 as a surfactant to mimic the dielectric properties of particularly fatty healthy tissues. The dielectric properties of the produced phantoms were compared to the ones of the human breast ex-vivo tissue up to 50 GHz. In the third and last chapter, the test of the linear prototype of the millimeter wave system for the detection of different types of targets was presented. In particular, two targets in air, metal sphere and cylinder of water and gelatin, and on two targets (cylinder of water and gelatin, and currant) included in two different phantoms (oil and waste-oil hardener, and Polysorbate80 and waste-oil hardener), deep up to 2 cm below the surface of the phantom were used to test the mm-wave imaging prototype. The synthetic array was composed of 24 antennas in 28 positions, with spacing between phase centers of adjacent radiators equal to half wavelength in air at 30 GHz, 5mm. The results showed that the proposed system is able to correctly identify the position of the target in both scenarios, in air and in phantom, with sub-cm resolutions in both cases. Then, in this year, the proof of concept of the feasibility of the imaging system with central working frequency of around 30 GHz for the early detection of breast cancer was demonstrated. Therefore, in conclusion, in this Ph.D. Thesis, the feasibility of a system of this type for applications of screening and possibly diagnosis for breast cancer in women whose breasts lend themselves to scanning with mammography is demonstrated in many respects, strongly encouraging further analysis in this direction.