scholarly journals Design and Development of an Automatic Blood Detection System for Capsule Endoscopy Images

Author(s):  
Pedro Pons ◽  
Reinier Noorda ◽  
Andrea Nevárez ◽  
Adrián Colomer ◽  
Vicente Pons Beltrán ◽  
...  
2010 ◽  
Vol 4 (2) ◽  
Author(s):  
Yurika Inoue ◽  
Amir Sabet Sarvestani ◽  
Jonathan Yap ◽  
Douglas L. Yeung

Video capsule endoscopy (VCE) is a method used to wirelessly scan the gastrointestinal (GI) tract lumen. Despite the effectiveness of VCE in GI endoscopic procedures, VCE is limited because it lacks the capability to exactly locate itself as well as accurately detect bleeding or clotting within the GI tract. The unreliability of localization combined with inaccurate blood detection in VCE ultimately leads to wired endoscopy for additional diagnosis. In this paper, a method to address localization of the VCE device, also known as a pill camera or capsule, in conjunction with an accurate detection of active bleeding and clotting inside the small intestine, is introduced. The Texas Instruments, ZigBee® kit, which uses a 3D trilateration method will be used for accurate location detection and image transmission throughout the VCE procedure. The system will be interfaced with software providing end users with the path and total distance traveled by the pill capsule within the small intestine. The blood detection system is enhanced by using a minilow energy wireless Raman spectrometry to scan for active bleeding or clots along the small intestinal wall. The employed spectrometry method scans for wavelengths based on blood’s optical characteristics and records any image fitting the exact spectrum. Blood detection and localization data are coupled together and then transferred to an external receiver. These two improvements together will enhance capsule endoscopy procedures and fill the gap created by the existing capsule endoscopy technologies, therefore, meeting the needs of physicians and patients.


2012 ◽  
Vol 462 ◽  
pp. 647-651
Author(s):  
Hong Tao Wang ◽  
Ke Ming Bi

This paper introduces a kind of grade detection system of underground iron ore. The system uses the technology of multi-threaded parallel processing of sensor to measure the volume and weight of the ore. Based on a certain mathematical model of grade analysis we have built, the average grade of the ore can be calculated. The trial run of the system indicates that the data of ore’s weight, volume and grade can be stored and counted automatically. What’s more, the system achieves non-stop measurement of ore’s grade.


Sign in / Sign up

Export Citation Format

Share Document