Detection of small bowel tumor in wireless capsule endoscopy images using an adaptive neuro-fuzzy inference system

Abstract

Automatic diagnosis tool helps physicians to evaluate capsule endoscopic examinations faster and more accurate. The purpose of this study was to evaluate the validity and reliability of an automatic post-processing method for identifying and classifying wireless capsule endoscopic images, and investigate statistical measures to differentiate normal and abnormal images. The proposed technique consists of two main stages, namely, feature extraction and classification. Primarily, 32 features incorporating four statistical measures (contrast, correlation, homogeneity and energy) calculated from co-occurrence metrics were computed. Then, mutual information was used to select features with maximal dependence on the target class and with minimal redundancy between features. Finally, a trained classifier, adaptive neuro-fuzzy interface system was implemented to classify endoscopic images into tumor, healthy and unhealthy classes. Classification accuracy of 94.2% was obtained using the proposed pipeline. Such techniques are valuable for accurate detection characterization and interpretation of endoscopic images.

Keywords

adaptive neuro-fuzzy inference system / co-occurrence matrix / wireless capsule endoscopy / texture feature

References

[1] Alizadeh.M, Zadeh HS, MaghsoudiOH . Segmentation of small bowel tumors in wireless capsule endoscopy using level set method: 27th International Symposium on Computer-Based Medical Systems (CBMS), New York City, USA, 2014[C]. Los Alamitos: IEEE, 2014.

[2] EskandariH, Talebpour A, AlizadehM , Polyp detection in wireless capsule endoscopy images by using region-based active contour model: 19th Iranian Conference on Biomedical Engineering (ICBME), Tehran, Iran, 2013[C]. Tehran: IEEE, 2013.

[3] AlizadehM, Talebpour A, ZadehHS , Effects of improved adaptive gamma correction method on wireless capsule endoscopy images: illumination compansation and edge detection: 20th Iranian Conference on Electrical Engineering (ICEE), Tehran, Iran, 2012[C].Tehran: IEEE, 2012.

[4] Maghsoudi OH, Talebpour  A, Zadeh HS , Informative and uninformative regions detection in WCE frames, columbia international publishing[J]. J Adv Comput, 2014, 3(1): 12–34.

[5] AlizadehM, Eskandari H, SharzehiK . Lymphangiectasia detection in wireless capsule endoscopy images using fisher transform method: 41st Annual Northeast Bioengineering Conference (NEBEC), Troy, USA, 2015[C].Syracuse: IEEE, 2015.

[6] AlizadehM, Sharzehi K, TalebpourA , Detection of uninformative regions in wireless capsule endoscopy images: 41st Annual Northeast Bioengineering Conference (NEBEC), Troy, USA, 2015[C].Syracuse: IEEE, 2015.

[7] Kodogiannis VS, Boulougoura  M, Lygouras JN , A neuro-fuzzy-based system for detecting abnormal patterns in wireless-capsule endoscopic images[J]. Neurocomput, 2007, 70(4-6): 704–717.

[8] HemmatiHR, Kamali-asl AR, TalebpourAR , Segmentation of carotid arteries in computed tomography angiography images using fast marching and graph cut methods: 21st Iranian Conference on Electrical Engineering (ICEE), Mashhad, Iran, 2013[C].Mashhad: IEEE, 2013.

[9] LeeJ, Oh JH, ShahSK , Automatic classification of digestive organs in wireless capsule endoscopy videos: SAC 07 Proceedings of the ACM Symposium, Seoul, South Korea, 2007[C].  New York: ACM, 2007.

[10] Li B, Meng  MQH. Tumor recognition in wireless capsule endoscopy images using textural features and SVM-based feature selection[J]. IEEE Trans Inf Technol Biomed, 2012, 16(3): 323–329.

[11] BarbosaDJC, Ramos J, CorreiaJH , Automatic detection of small bowel tumors in Capsule Endoscopy based in Color Curvelet Covariance statistical texture descriptors: Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Minneapolis, USA, 2009[C].Piscataway: IEEE, 2009.

[12] MartinsMM, Barbosa DJ, RamosJ , Small bowel tumors detection in capsule endoscopy by gaussian modeling of color curvelet covariance coefficients: Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Buenos Aires, Argentina, 2010[C].Piscataway: IEEE, 2010.

[13] CharisisV, Hadjileontiadis LJ, LiatsosCN , Abnormal pattern detection in wireless capsule endoscopy images using nonlinear analysis in RGB color space: 32nd Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Buenos Aires, Argentina, 2010[C]. Piscataway: IEEE, 2010.

[14] Kodogiannis VS, Boulougoura  M. An adaptive neurofuzzy approach for the diagnosis in wireless capsule endoscopy imaging[J]. Int J Inf Tech, 2007, 13(1): 46–59.

[15] BarbosaDJC, Ramos J, LimaCS . Detection of small bowel tumors in capsule endoscopy frames using texture analysis based on the discrete wavelet transform: Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Vancouver, Canada, 2008[C]. Piscataway: IEEE, 2008.

[16] Marcin Blachnik, Jorma Laaksonen. Image classification by histogram features created with learning vector quantization: Lecture Notes in Computer Science International Artificial Neural Networks (ICANN), Prague, Czech, 2008[C]. Berlin: Springer-Verlag, 2008.

[17] Oveisi F, Oveisi  S, Erfanian A , Tree-structured feature extraction using mutual information[J]. IEEE Trans Neural Netw Learn Syst, 2012, 23(1): 127–137.

[18] Tourassi GD, Frederick  ED, Markey MK , Application of the mutual information criterion for feature selection in computer-aided diagnosis[J]. Med Phys, 2001, 28(12): 2394–2402

[19] Alizadeh M. Image guided radiation therapy: applications in radiology and endoscopy[J]. Am J Bioengine Biotech, 2016, 2(1): 15–23.

[20] Güler I, Ubeyli  ED. Adaptive neuro-fuzzy inference system for classification of EEG signals using wavelet coefficients[J]. J Neurosci Methods, 2005, 148(2): 113–121.

[21] Asli.  C, I.  Burhan Turksen. Modeling Uncertanity with Fuzzy Logic[M]. Germany: Springer-Verlag Berlin Heidelberg, 1st Edition, 2009: 1–400.

[22] Li B, Meng  MQH, Lau JYW . Computer-aided small bowel tumor detection for capsule endoscopy[J]. Artif Intell Med, 2011, 52(1): 11–16.

[23] Kodogiannis VS, Lygouras  JN. Neuro-fuzzy classification system for wireless-capsule endoscopic images, world academy of science[J]. Eng Tech, 2008, 21: 620–628.

[24] Vaidhehi V. The role of dataset in training anfis system for course advisor[J]. Int J Innov Res Adv Eng, 2014, 1(6): 249–253.

Acknowledgements

The authors thank all the staff of the Endoscopy Unit of Internal Medicine at Shariati Hospital for their helpful contribution.

RIGHTS & PERMISSIONS

2017 2017 by the Journal of Biomedical Research. All rights reserved