In this paper, an application of multiple-input multiple-output (MIMO) radar techniques to medical imaging is investigated. MIMO radar, which can be viewed as a form of microwave tomography for medical imaging geometries, offers an alternative to current X-ray technology. Of particular interest is the detection of cancerous tumors in breast tissue. Previously, ultrawideband radar techniques have been considered. Here, an intermediate bandwidth is assumed, which enables the digitization of the entire baseband signal. However, this comes with the cost of self-interference because of transmitting and receiving simultaneously. Because the scattering response of tumors is weak compared to self-interference, mitigation of the direct transmitter-to-receiver signal is important to detection. In this application, three interference suppression techniques are considered: temporal mitigation, receive-array mitigation, and transmit-waveform optimization. Mitigation residuals, caused by calibration errors and signal nonlinearities, are included in the analysis. A system concept is introduced, and results from simulations using a simplified physical model are presented.