A Saliency-based Unsupervised Method for Angiectasia Detection in Endoscopic Video Frames
The detection of angiectasia, the primary suspected lesion in patients with obscure gastrointestinal bleeding, presents a challenging problem for physicians. In this paper, we present a saliency based unsupervised method for automatic localization and detection of angiectasia in wired and capsule endoscopic images. To achieve comparable illumination in images from both modalities, image enhancement based on Retinex is performed on the capsule endoscopic images. A saliency detection algorithm has been proposed where the saliency map is formed from the processed images using two distinctness measures: pattern distinctness and color distinctness. The angiectasia specific saliency detection algorithm is able to highlight the lesion affected areas. An adaptive thresholding is performed based on the saliency peaks detected from the gradient images. The performance of the proposed method is evaluated on a dataset consisting of 3602 images, among which 968 images show the indication of angiectasia. The method achieves very high localization score (95.04%), localization precision, moderate specificity (>80%) and a very low detection latency (<0.2 s) for both imaging modalities. A comparison with state-of-the-art saliency detection methods exhibits the efficacy of proposed saliency detection algorithm for angiectasia localization and detection.
KeywordsAngiectasia Endoscopy Retinex Saliency detection Detection latency
- 11.Iakovidis, D. K., Chatzis, D., Chrysanthopoulos, P., & Koulaouzidis, A. (2015). Blood detection in wireless capsule endoscope images based on salient superpixels. In: 37th Annual International Conference of the IEEE, Engineering in Medicine and Biology Society (EMBC), (pp. 731–734).Google Scholar
- 15.Stockman, D. L. (2016). Diagnostic pathology: Vascular. Netherland: Elsevier.Google Scholar
- 16.Okuhata, H., Nakamura, H., Hara, S., Tsutsui, H., & Onoye, T. (2013). Application of the real-time Retinex image enhancement for endoscopic images. In: 2013 35th Annual International Conference of the IEE, Engineering in Medicine and Biology Society (EMBC), pp. 3407–3410.Google Scholar
- 19.Deeba, F., Mohammed, S. K., Bui, F. M., & Wahid, K. A. (2016). Unsupervised Abnormality Detection Using Saliency and Retinex based Color Enhancement. In: The 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, pp. 3871–3874.Google Scholar
- 20.Margolin, R., Tal, A., & Zelnik-Manor, L. (2013). What makes a patch distinct?. In: Proceedings of IEEE Computer Society Conference Computer Vision Pattern Recognition, 1139–1146.Google Scholar
- 21.Cheng, W. C., Cheng, H. C., Chen, P. J., Kang, J. W., Yang, E. H., Sheu, B.-S., et al. (2015). Higher net change of index of hemoglobin values between colon polyp and nonpolyp mucosa correlates with the presence of an advanced colon adenoma. Advances in Digestive Medicine. doi: 10.1016/j.aidm.2015.04.005.Google Scholar
- 23.CapsuleEndoscopy.org. Retrieved from http://www.capsuleendoscopy.org/Pages/default.aspx.
- 24.El Atlas Gastrointestinal. Retrieved from http://www.gastrointestinalatlas.com/.
- 29.Harel, J., Koch, C., & Perona, P. (2007). Graph-based visual saliency. In Advances in neural information processing systems, 545–552.Google Scholar
- 31.WEO Clinical Endoscopy Atlas (2014). http://www.endoatlas.org/. Accessed 25 May 2016.