Automated Polyp Detection in Colonoscopy Videos Using Shape and Context Information

Abstract

This paper presents the culmination of our research in designing a system for computer-aided detection (CAD) of polyps in colonoscopy videos. Our system is based on a hybrid context-shape approach, which utilizes context information to remove non-polyp structures and shape information to reliably localize polyps. Specifically, given a colonoscopy image, we first obtain a crude edge map. Second, we remove non-polyp edges from the edge map using our unique feature extraction and edge classification scheme. Third, we localize polyp candidates with probabilistic confidence scores in the refined edge maps using our novel voting scheme. The suggested CAD system has been tested using two public polyp databases, CVC-ColonDB, containing 300 colonoscopy images with a total of 300 polyp instances from 15 unique polyps, and ASU-Mayo database, which is our collection of colonoscopy videos containing 19,400 frames and a total of 5,200 polyp instances from 10 unique polyps. We have evaluated our system using free-response receiver operating characteristic (FROC) analysis. At 0.1 false positives per frame, our system achieves a sensitivity of 88.0% for CVC-ColonDB and a sensitivity of 48% for the ASU-Mayo database. In addition, we have evaluated our system using a new detection latency analysis where latency is defined as the time from the first appearance of a polyp in the colonoscopy video to the time of its first detection by our system. At 0.05 false positives per frame, our system yields a polyp detection latency of 0.3 seconds.

Efficient disease detection in gastrointestinal videos – global features versus neural networks

“…12 It consists of training and test sets of images and videos with corresponding ground truth showing the exact polyp location areas. This was the biggest publicly available dataset (until recently, when the owners decided to withdrawn it from the public), consisting of 20 videos from standard colonoscopies with a total of 18,781 frames and different resolutions up to full HD [44]. For this particular evaluation, we selected only frames containing polyps, which gave us 8,169 frames in total: 3,856 in the training subset and 4,313 in the test subset.…”

Computer-Aided Diagnosis in Colorectal Cancer: Current Concepts and Future Prospects

“…20 The 907 images used were collected by physicians during NBI colonoscopy examinations with similar lighting conditions and image zooming, and the maximum recognition rate was 94.1%, while the recall rate specific for C3-type polyps was only 73.6%. 21 In a study by Tajbakhsh et al (2016), another hybrid technique was proposed for the detection of colon polyps. 22 A combination of both context-and shape-based approach was considered due to the fact that a shapebased approach alone may falsely identify other polyp-like structures such as fecal content, and that a context-based approach may not be able to extract the discriminative geometric information of the polyps.…”

“…21 In a study by Tajbakhsh et al (2016), another hybrid technique was proposed for the detection of colon polyps. 22 A combination of both context-and shape-based approach was considered due to the fact that a shapebased approach alone may falsely identify other polyp-like structures such as fecal content, and that a context-based approach may not be able to extract the discriminative geometric information of the polyps. A powerful descriptor that is invariant to rotation and robust against light setting changes was used along with a two-stage classification scheme, which greatly improved the detection performance in colonoscopy videos.…”

“…The difference was due to the fact that there were fewer images in the public database, with no images that had no polyps inside. 22 …”

Polyp Detection and Segmentation from Video Capsule Endoscopy: A Review

“…Many previous studies have focused on detecting polyps on classical imaging techniques of colonoscopies (not capsule endoscopy images) [3][4][5][6][7][8][9][10]. Polyp detection approaches in colonoscopy imagery include using elliptical features [6], texture [3,11], color and position features [12,13] (see [14] for a review of polyp detection methods in colonoscopy images).…”