Abstract
Wireless capsule endoscopy is an innovative technology for visualizing anomalies in the gastrointestinal tract, useful to replace traditional endoscopic diagnosis. Its advantages are related to the capability to reach the duodenum and small intestine, while eliminating the discomfort of patients. The time spent by a physician analyzing the results of wireless capsule endoscopy video can vary between 45 and 180 minutes, limiting its widespread diffusion. Therefore, methods able to perform an automatic pre-screening of images of interest are necessary. This paper presents an innovative technique to detect bleeding regions in wireless capsule en-doscopy video. Experimental results show that the proposed algorithm exhibits a low false alarm rate, and is effective at reducing the time needed to analyze video sequences.
| Original language | English |
|---|---|
| Pages (from-to) | 1864-1868 |
| Number of pages | 5 |
| Journal | European Signal Processing Conference |
| Publication status | Published - 2009 |
| Event | 17th European Signal Processing Conference, EUSIPCO 2009 - Glasgow, United Kingdom Duration: 24 Aug 2009 → 28 Aug 2009 |