Comparison of noncoherent linear breast cancer detection algorithms applied to a 2-D numerical model

A comparative analysis of an imaging method based on a multifrequency Multiple Signal Classification (MUSIC) approach against two common linear detection algorithms based on noncoherent migration is made. The different techniques are tested using synthetic data generated through CST Microwave Studio and a phantom developed from MRI scans of a mostly fat breast. The multifrequency MUSIC approach shows an overall superior performance compared to the noncoherent techniques. This letter reports that this highly performing algorithm does not require any antenna calibration or phase response estimation and allows the use of efficient and complex antenna geometries without difficult algorithm redefinitions.