摘要
由于Wireless Capsule Endoscopy(WCE)在消化道中采集到的巨大数量的图像均需要医务人员靠肉眼来排查,给医生带来巨大的负担。该文提供了一种基于支持向量机(Support Vector Machine,SVM)分类器的胶囊内窥镜出血智能识别方法,创立一种新的特征参数,并对SVM参数的选择进行实验优化,最终达到94%的特异度与83%的灵敏度。
Because the huge number of images of the digestive tract by Wireless Capsule Endoscopy（WCE） are left to the medical personnels detected by their eyes, huge burden leaves to doctors. This article provides a classification of method based on SVM（Support Vector Machine） for the capsule endoscopy bleeding intelligent recognition. We created a new kind of feature parameter, and the experiment result can reach 83% specificity and 94% sensitivity.
出处
《中国医疗器械杂志》
CAS
2015年第1期9-12,共4页
Chinese Journal of Medical Instrumentation
基金
国家自然科学基金(31170968)
关键词
特征提取
SVM
出血检测
无线胶囊内窥镜
颜色空间
feature selection
SVM
bleeding detection
wireless capsule endoscopy
color space
