In-body localization and tracking for wireless capsule endoscopy using received signal strength of an RF-transmitter

In this master dissertation, the localization of the wireless capsule endoscopy (WCE) is studied in order to minimize the error. In particular, the influence of the different tissues of which the human body is composed and their impact on localization were analyzed. To determine the behavior of the WCE a computational model of the gastrointestinal (GI) tract and the WCE have been implemented. Moreover, a finite difference time domain solver simulation on Sim4Life has been used to study how the WCE moves through the GI tract. Four different simulations were considered, three homogeneous and one heterogeneous: one with the air model, one with fat, one with muscle, and one with a complete human body model containing all tissues. Various parameters were obtained such as the electric field along the entire GI tract, the distance between the transmitter (Tx) and the receiver (Rx), and the behavior of the Tx in the different tissues. Through the analysis of the data obtained it was possible to interpolate them according to a hyperbola and subsequently evaluate the localization error. It was clear that the position of the WCE along the GI tract influences the results obtained for the homogeneous muscle model and for the Duke heterogeneous model. For the other two models considered, the homogeneous model of fat and air, this parameter does not significantly affect the results obtained. Furthermore, the position of the receivers in respect to the skin in some cases played a key role in the accuracy of the data obtained. Moreover, the only hyperbolic relationship found between the electric field and the distance between the Tx and Rx was the one for the homogeneous air model. The other simulations did not yield this result. Finally, the localization error for the heterogeneous model was the largest one obtained (between 30 and 40 cm), although comparable with the homogeneous one for muscle. The values obtained for fat and air are smaller.