Summary: Automatic Region-of-Interest Segmentation and
Pathology Detection in Magnetically Guided
Capsule Endoscopy
Philip W. Mewes1,2
, Dominik Neumann1
, Oleg Licegevic1
, Johannes Simon1
,
Aleksandar Lj. Juloski1
, and Elli Angelopoulou2
1
Siemens AG, Healthcare Sector, Erlangen, Germany
2
Pattern Recognition Lab, University of Erlangen-N¨urnberg, Erlangen, Germany
philip.mewes@informatik.uni-erlangen.de
Abstract. Magnetically-guided capsule endoscopy (MGCE) was intro-
duced in 2010 as a procedure where a capsule in the stomach is navigated
via an external magnetic field. The quality of the examination depends on
the operator's ability to detect aspects of interest in real time. We present
a novel two step computer-assisted diagnostic-procedure (CADP) algo-

Source: Angelopoulou, Elli - Department of Computer Science, Friedrich Alexander University Erlangen Nürnberg

Collections: Computer Technologies and Information Sciences