Wearable Microwave Imaging Sensor for Deep Tissue Real-Time Monitoring Using a New Loss-Compensated Backpropagation Technique

Abstract

We present a new imaging method for continuous real-time monitoring of deep human tissues using a body-worn radio-frequency sensor. The sensor consists of a set of transmitting and receiving dipole-like probes placed around the human body. The Green’s function of the medium is derived by modeling the scattering of the cross-sectional pixels using metallic cylinders of equivalent sections. A new method of overcoming the signal path loss in the biological medium is then introduced, which increases the image resolution. For the first time, the proposed wearable sensor is shown to overcome a number of challenges with body-worn imaging including the shape uncertainty of the torso. To analyze this, a new boundary mismatch parameter (BMP) is introduced and its effect on image recovery is examined quantitatively. Demonstration of the concept is carried out by retrieving differential permittivity image of the underlying cross-section, employing data obtained from several imaging scenarios using both full-wave simulations involving deep human tissues with multiple outer layers (i.e. skin, fat, muscle, bone) and experiments. Overall, the proposed sensor can be the basis for a portable and low-cost complement to conventional imaging techniques (such as, X-Ray CT, MRI etc.) suitable for a wide range of applications, such as, prevention of pulmonary diseases, brain functionality imaging and monitoring of gastric emptying etc.