摘要
Bleeding in the digestive tract is one of the most common gastrointestinal (GI) tract diseases,as well as the complication of some fatal diseases. Wireless capsule endoscopy (WCE),which is widely applied in the clinical field,allows physicians to noninvasively examine the entire GI tract. However,it is very laborious and timeconsuming to detect the huge amount of WCE images,and limits its wider application. It is urgent and necessary to develop the automatic and intelligent computer aided bleeding detection technique. This paper improves the Euler distance with the covariance matrix of image to measure the colour similarity in CIELab colorimetric system,and proposes a novel method of bleeding detection in WCE images. The experiments demonstrate that the bleeding region in WCE images can be correctly recognized and marked out,and the sensitivity of this method is 92%,the specificity is 95%.
Bleeding in the digestive tract is one of the most common gastrointestinal （GI） tract diseases,as well as the complication of some fatal diseases. Wireless capsule endoscopy （WCE）,which is widely applied in the clinical field,allows physicians to noninvasively examine the entire GI tract. However,it is very laborious and timeconsuming to detect the huge amount of WCE images,and limits its wider application. It is urgent and necessary to develop the automatic and intelligent computer aided bleeding detection technique. This paper improves the Euler distance with the covariance matrix of image to measure the colour similarity in CIELab colorimetric system,and proposes a novel method of bleeding detection in WCE images. The experiments demonstrate that the bleeding region in WCE images can be correctly recognized and marked out,and the sensitivity of this method is 92%,the specificity is 95%.
基金
the National High Technology Research and Development Program (863) of China(No.2006AA04Z368)
the National Natural Science Foundation of China (No.30570485)
