A novel adaptive localization technique for wireless capsule endoscopy

Hamed Farhadi1, Esmaeil S. Nadimi2, Javid Atai3, Kaveh Pahlavan4, Mikael Skoglund1, Vahid Tarokh5. 1KTH Royal Institute of Technology, 2University of Southern Denmark, 3The University of Sydney, 4Worcester Polytechnic Institute, 5Harvard University

Objective of the technology:

Wireless capsule endoscopy (WCE) is an emerging technique to enhance Gastroenterologists information about the patient’s G.I. tract. Localization of capsule inside human body in this case is an active area of research. This can be thought of as a subdomain of micro and bio-robotics fields and also that of localization in heterogeneous media. If capsule and micro-robot localization problem in human body is solved, then it may potentially lead to less invasive G.I. diseases treatments and other micro-robot assisted medical procedures.

Description of the method:

We propose a novel technique that uses RF signals radiated from the capsule to determine its location. In the proposed technique, multiple receiver elements on the body surface receive noisy versions of the RF signals radiated from the capsule, and an algorithm computes the location of the capsule using these set of received signals. For the purpose of location estimation, the signal propagation inside human body is characterized using an approximate linear model. The model parameters change slowly as the capsule navigates inside the heterogeneous medium of human body. We have designed an adaptive localization algorithm based on expectation maximization (EM) technique that tracks changes in the propagation environment and updates/improves the estimated locations.

Preliminary results:

The accuracy of the proposed technique has been verified using numerical simulations. The preliminary results show that the proposed method significantly outperforms the non-adaptive localization techniques existing in the literature.


The proposed technique will help Gastroenterologists specialists to use WCE not only to discover medical problems inside the G.I. track, but also to know where the problems are exactly located. This will enable the next generations of WCE systems in which a capsule can be used to conduct certain surgeries in specific locations; to deliver drug to a desired point; or to collect tissue samples from a location of interest inside the G.I. track.

« Return to SAGES 2015 abstract archive