Automatic organs' detection in WCE

Wireless capsule endoscopy (WCE) views the entire gastrointestinal (GI) tract. A main problem associated with this novel device is that too many frames must be reviewed by physicians. Thus it is essential to find an automatic and intelligent method to help physicians. One of the problems in WCE is i...

Full description

Saved in:
Bibliographic Details
Published in2012 16th CSI International Symposium on Artificial Intelligence and Signal Processing pp. 116 - 121
Main Authors Haji-Maghsoudi, Omid, Talebpour, A., Soltanian-Zadeh, H., Haji-maghsoodi, N.
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.05.2012
Subjects
Active contour
Active contours
Color spaces
Esophagus
Fuzzy Clustering
Gabor Filter
Gabor filters
Haralick Features
Image Segmentation
Level set
Medical services
Neural Network
Optical wavelength conversion
WCE
Wireless communication
Online AccessGet full text
ISBN9781467314787
1467314781
DOI10.1109/AISP.2012.6313729

Cover

More Information
Summary:Wireless capsule endoscopy (WCE) views the entire gastrointestinal (GI) tract. A main problem associated with this novel device is that too many frames must be reviewed by physicians. Thus it is essential to find an automatic and intelligent method to help physicians. One of the problems in WCE is its difficulty to distinguish among different organ's tissues. So, we introduce two novel algorithms which are able to classify main organs (among esophagus, stomach, small bowel and colon) in WCE's frames. In order to obtain our aim, we use statistic features (Haralick features) and non-statistic features (different diagrams and Gabor filter banks), colored features and non-colored features. Our experimental studies indicate good results that are shown in conclusion.
ISBN:9781467314787
1467314781
DOI:10.1109/AISP.2012.6313729