Abstract
Automatic polyp detection in endoscopy (or colonoscopy) images is challenging because the types of polyp and their appearances are diverse, and the colors and textures of polyps are quite similar to those of normal tissues in many cases. It is thus often very difficult to distinguish polyps from normal tissues using conventional methodology. To effectively resolve these challenges, we propose a framework based on multi-classifier learning and a contour intensity difference (CID) measure. To detect polyps of diverse appearances, we first classify polyps into K types according to their shape via unsupervised learning. We then train K classifiers to detect the K types of polyp. This multi-classifier learning improves the polyp detection rate. However, false positives also increase because colon structures look similar to polyps. To reduce false positives while preserving the high detection rate, we propose a CID measure. Experimental results using public and our own datasets show that the proposed methods are promising for detecting polyps with diverse appearances.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
We aim to train multiple classifiers that can detect specific shapes of polyps against the background. Therefore, in this paper, we do not cluster the negative samples.
In order to find the CID probability densities, we performed the following process. First, we manually detected all polyps and the FP factors in the datasets. Second, we extracted all CID measures of polyps and FP factors in the test dataset and found the CID probability densities. For example, when we conducted fivefold cross-validation, we have five test datasets, and obtain five CID probability densities of polyps and FP factors, respectively. In practice, we set the CID distributions from the first fold test dataset as the representatives since the five distributions from each fold are very similar to each other.
Test scenario 3 is challenging because ODB contains more diverse polyps and complex scenes than those in CVC and the numbers of training and test samples are not balanced (training: 300 ≪ test: 1432).
References
World cancer report. (2014). IARC Nonserial Publication.
Baxter, N. N., Goldwasser, M. A., Paszat, L. F., Saskin, R., Urbach, D. R., & Rabeneck, L. (2009). Association of colonoscopy and death from colorectal cancer. Annals of Internal Medicine, 150, 1–8.
van Rijn, J. C., Reitsma, J. B., Stoker, J., Bossuyt, P. M., van Deventer, S. J., & Dekker, E. (2006). Polyp miss rate determined by tandem colonoscopy: A systematic review. The American Journal of Gastroenterology, 101, 343–350.
Ameling, S., Wirth, S., Paulus, D., Lacey, G., & Vilarino, F. (2009). Texture-based polyp detection in colonoscopy. Bildverarbeitung für die Medizin, 346–350. doi:10.1007/978-3-540-93860-6_70.
Iakovidis, D. K., Maroulis, D. E., & Karkanis, S. A. (2006). An intelligent system for automatic detection of gastrointestinal adenomas in video endoscopy. Computers in Biology and Medicine, 36, 1084–1103.
Karkanis, S. A., Iakovidis, D. K., Maroulis, D. E., Karras, D. A., & Tzivras, M. (2003). Computer-aided tumor detection in endoscopic video using color wavelet features. IEEE Transactions on Information Technology in Biomedicine, 7, 141–152.
Li, B.-P., & Meng, M. Q.-H. (2012). Comparison of several texture features for tumor detection in CE images. Journal of Medical Systems, 36, 2463–2469.
Alexandre, L. A., Casteleiro, J., & Nobreinst, N. (2007). Polyp detection in endoscopic video using SVMs. Knowledge Discovery in Databases, 2007, 358–365.
Tjoa, M. P., & Krishnan, S. M. (2003). Feature extraction for the analysis of colon status from the endoscopic images. BioMedical Engineering OnLine, 2, 1–17.
Dietterich, T. G. (2000). Ensemble methods in machine learning. Multiple Classifier Systems, 1857, 1–15.
Dalal, N., & Triggs, B. (2005). Histograms of oriented gradients for human detection. Computer Vision and Pattern Recognition, 1, 693–886.
Zhang, C., Platt, J. C., & Viola, P. A. (2005). Multiple instance boosting for object detection. Advances in Neural Information Processing Systems, 7, 1417–1424.
Bernal, J., Sánchez, J., & Vilarino, F. (2012). Towards automatic polyp detection with a polyp appearance model. Pattern Recognition, 45, 3166–3182.
Iakovidis, D. K., Maroulis, D. E., Karkanis, S. A., & Brokos, A. (2005). A comparative study of texture features for the discrimination of gastric polyps in endoscopic video, Computer-Based Medical Systems, 575–580.
Li, B., & Meng, M. Q.-H. (2012). Automatic polyp detection for wireless capsule endoscopy images. Expert Systems with Applications, 39, 10952–10958.
Bieniek, A., & Moga, A. (2000). An efficient watershed algorithm based on connected components. Pattern Recognition, 6, 907–916.
Hwang, S., and Oh, J-H., Tavanapong, W., Wong, J., & de Groen, P. C. (2007). Polyp detection in colonoscopy video using elliptical shape feature, International Conference on Image Processing, 2, II-465.
Li, P., Chan, K. L., & Krishnan, S. M. (2005). Learning a multi-size patch-based hybrid kernel machine ensemble for abnormal region detection in colonoscopic images. Computer Vision and Pattern Recognition, 2, 670–675.
Phong, B. T. (1975). Illumination for computer generated pictures. Communications of the ACM, 18, 311–317.
Canny, J. (1986). A computational approach to edge detection. Pattern Analysis and Machine Intelligence, 6, 679–698.
Dollár, P., & Zitnick, C. L. (2013). Structured Forests for Fast edge Detection, International Conference on Computer Vision, 1841–1848.
Coope, I. D. (1993). Circle fitting by linear and nonlinear least squares. Journal of Optimization Theory and Applications, 76, 381–388.
Everingham, M., Zisserman, A., Williams, C. K., Van Gool, L., Allan, M., Bishop, C. M., et al., (2006). The 2005 pascal visual object classes challenge. Machine Learning Challenges. Evaluating Predictive Uncertainty, Visual Object Classification, and Recognising Tectual Entailment, 117–176. doi:10.1007/11736790_8.
Dollár, P., and Wojek, C., Schiele, B., & Perona, P. (2009). Pedestrian detection: A benchmark, Computer Vision and Pattern Recognition, 304–311.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Cho, YJ., Bae, SH. & Yoon, KJ. Multi-Classifier-Based Automatic Polyp Detection in Endoscopic Images. J. Med. Biol. Eng. 36, 871–882 (2016). https://doi.org/10.1007/s40846-016-0190-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40846-016-0190-4