TY - CHAP
T1 - Biomedical image analysis on wireless capsule endoscopy images and videos
AU - Yang, Guang
AU - Yin, Yafeng
AU - Man, Hong
N1 - Publisher Copyright:
© Springer Science+Business Media New York 2013. All rights are reserved.
PY - 2013/11/1
Y1 - 2013/11/1
N2 - Wireless capsule endoscopy (WCE) is a newly booming technology on gastrointestinal (GI) examinations. After the patient swallows the capsule, it starts taking video of the entire interior of the digestive tract, from the esophagus, stomach to small intestine, and colon. It transmits sequences of color images back to the computer, and hence allows the doctors to observe the patient's GI tract vividly and hence diagnose precisely based on the visually accessed video instead of reconstructed images such as MRI. The entire process takes hours with around 50,000 images about the inside of the GI tract. It is very time-consuming to go through those images frame by frame. Therefore, many image/video processing techniques are adopted to develop software which is expected to annotate the images in different stage of the GI tract and detect unusual events, such as bleeding, polyps, and ulcers, automatically. In another words, the problems in WCE images/videos can be categorized in two classes, video segmentations and event detections. Various methods have been discussed in the literatures and generally can be seen as two-stage approaches, describing images and classifying images or regions. In this chapter, we will review various feature descriptors and classification methods that are often used on WCE images/videos.
AB - Wireless capsule endoscopy (WCE) is a newly booming technology on gastrointestinal (GI) examinations. After the patient swallows the capsule, it starts taking video of the entire interior of the digestive tract, from the esophagus, stomach to small intestine, and colon. It transmits sequences of color images back to the computer, and hence allows the doctors to observe the patient's GI tract vividly and hence diagnose precisely based on the visually accessed video instead of reconstructed images such as MRI. The entire process takes hours with around 50,000 images about the inside of the GI tract. It is very time-consuming to go through those images frame by frame. Therefore, many image/video processing techniques are adopted to develop software which is expected to annotate the images in different stage of the GI tract and detect unusual events, such as bleeding, polyps, and ulcers, automatically. In another words, the problems in WCE images/videos can be categorized in two classes, video segmentations and event detections. Various methods have been discussed in the literatures and generally can be seen as two-stage approaches, describing images and classifying images or regions. In this chapter, we will review various feature descriptors and classification methods that are often used on WCE images/videos.
UR - http://www.scopus.com/inward/record.url?scp=84949177725&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84949177725&partnerID=8YFLogxK
U2 - 10.1007/978-1-4419-8411-1_3
DO - 10.1007/978-1-4419-8411-1_3
M3 - Chapter
AN - SCOPUS:84949177725
SN - 1441984100
SN - 9781441984104
VL - 9781441984111
SP - 23
EP - 43
BT - Selected Topics in Micro/Nano-Robotics for Biomedical Applications
ER -