Unsupervised summarisation of capsule endoscopy video

Capsule endoscopy is a non-invasive imaging technique commonly used for screening of the entire small intestine. It is performed by a wireless swallowable endoscopic capsule capable of transmitting thousands of video frames per examination. The visual inspection of the vast amount of images acquired during such an examination is a subjective and highly time consuming task even for experienced gastroenterologists. In this paper we propose a novel approach to the reduction of the number of the video frames to be inspected so as to enable faster inspection of the endoscopic video. It is based on symmetric non-negative matrix factorisation initialised by the fuzzy c-means algorithm and it is supported by non-negative Lagrangian relaxation to extract a subset of video frames containing the most representative scenes from a whole endoscopic examination. The experimental evaluation of the proposed approach was tested on annotated endoscopic videos with frames displaying ulcers, bleedings and normal tissues from various sites in the small intestine. The results demonstrate that the video summary produced consists of representative frames from all the abnormal findings and the normal tissues of the input video.

[1]  Dimitrios K. Iakovidis,et al.  An intelligent system for automatic detection of gastrointestinal adenomas in video endoscopy , 2006, Comput. Biol. Medicine.

[2]  Michal Mackiewicz,et al.  Colour and Texture Based Gastrointestinal Tissue Discrimination , 2006, 2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings.

[3]  Chris H. Q. Ding,et al.  Nonnegative Lagrangian Relaxation of K-Means and Spectral Clustering , 2005, ECML.

[4]  C. Ding,et al.  On the Equivalence of Nonnegative Matrix Factorization and K-means - Spectral Clustering , 2005 .

[5]  H. Sebastian Seung,et al.  Algorithms for Non-negative Matrix Factorization , 2000, NIPS.

[6]  Fernando Vilariño,et al.  A Machine Learning Framework Using SOMs: Applications in the Intestinal Motility Assessment , 2006, CIARP.

[7]  Jung-Hwan Oh,et al.  Automatic classification of digestive organs in wireless capsule endoscopy videos , 2007, SAC '07.

[8]  H. Sebastian Seung,et al.  Learning the parts of objects by non-negative matrix factorization , 1999, Nature.

[9]  R. Schoefl,et al.  Multicenter Retrospective Evaluation of Capsule Endoscopy in Clinical Routine , 2004, Endoscopy.

[10]  Baopu Li,et al.  Analysis of the gastrointestinal status from wireless capsule endoscopy images using local color feature , 2007, 2007 International Conference on Information Acquisition.

[11]  H. Sebastian Seung,et al.  Unsupervised Learning by Convex and Conic Coding , 1996, NIPS.

[12]  P. Swain,et al.  Wireless capsule endoscopy. , 2002, The Israel Medical Association journal : IMAJ.

[13]  Duncan Bell,et al.  Stomach, intestine, and colon tissue discriminators for wireless capsule endoscopy images , 2005, SPIE Medical Imaging.

[14]  Baopu Li,et al.  Wireless Capsule Endoscopy Images Enhancement by Tensor Based Diffusion , 2006, 2006 International Conference of the IEEE Engineering in Medicine and Biology Society.

[15]  John N. Lygouras,et al.  A neuro-fuzzy-based system for detecting abnormal patterns in wireless-capsule endoscopic images , 2007, Neurocomputing.

[16]  Miguel Tavares Coimbra,et al.  MPEG-7 Visual Descriptors—Contributions for Automated Feature Extraction in Capsule Endoscopy , 2006, IEEE Transactions on Circuits and Systems for Video Technology.

[17]  Daniel D. Lee,et al.  Multiplicative Updates for Classification by Mixture Models , 2001, NIPS.

[18]  Oleg Okun,et al.  Unsupervised data reduction , 2007, Signal Process..

[19]  Fernando Vilariño,et al.  Automatic Detection of Intestinal Juices in Wireless Capsule Video Endoscopy , 2006, 18th International Conference on Pattern Recognition (ICPR'06).

[20]  M. Boulougoura,et al.  Computer-assisted diagnosis of wireless-capsule endoscopic images using neural network based techniques , 2005, CIMSA. 2005 IEEE International Conference on Computational Intelligence for Measurement Systems and Applications, 2005..

[21]  V.S. Kodogiannis,et al.  Neural network-based approach for the classification of wireless-capsule endoscopic images , 2005, Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005..

[22]  Nikolaos G. Bourbakis,et al.  Detecting abnormal patterns in WCE images , 2005, Fifth IEEE Symposium on Bioinformatics and Bioengineering (BIBE'05).

[23]  Fernando Vilariño,et al.  ROC curves and video analysis optimization in intestinal capsule endoscopy , 2006, Pattern Recognit. Lett..

[24]  Jung-Hwan Oh,et al.  Blood detection in wireless capsule endoscopy using expectation maximization clustering , 2006, SPIE Medical Imaging.