Minimum redundancy maximum relevance (mRMR) based feature selection from endoscopic images for automatic gastrointestinal polyp detection

Abstract

In this paper, a computer based system has been proposed as a support to gastrointestinal polyp detection. It can detect and classify gastrointestinal polyps from endoscopic video. Color wavelet (CW) features and convolutional neural network (CNN) features of endoscopic video frames are extracted. Mutual information based feature selection technique-Minimum redundancy maximum relevance (mRMR) is used to scale down feature vector. Instead of using a single classifier, Bootstrap Aggregrating (Bagging)- an ensemble classifier is used. Proposed system has been assessed against different public databases and our own datasets. Evaluation shows that, the system outperforms the existing methods.

This is a preview of subscription content, log in to check access.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Data Availability

The Endoscopic Video data used to support the findings of this study are available from following websites:

01. http://www.depeca.uah.es/colonoscopy_dataset/ [18].

02. https://polyp.grand-challenge.org/databases/ [3]

Endoscopic videos collected from Lab Aid Hospital, Dhaka can be available upon request to the corresponding author.

References

  1. 1.

    Abouelenien M, Yuan X, Giritharan B, Liu J, Tang S (2013) Cluster-based sampling and ensemble for bleeding detection in capsule endoscopy videos. American Journal of Science and Engineering 2(1):24–32

    Google Scholar 

  2. 2.

    Alexandre LA, Nobre N, Casteleiro J (2008) Color and position versus texture features for endoscopic polyp detection. In: International conference on biomedical engineering and informatics, 2008. BMEI 2008, vol 2. IEEE, pp 38–42

  3. 3.

    Bernal J, Tajkbaksh N, Sánchez FJ, Matuszewski BJ, Chen H, Yu L, Angermann Q, Romain O, Rustad B, Balasingham I, et al. (2017) Comparative validation of polyp detection methods in video colonoscopy: results from the miccai 2015 endoscopic vision challenge. IEEE Trans Med Imaging 36(6):1231–1249

    Article  Google Scholar 

  4. 4.

    Berrendero JR, Cuevas A, Torrecilla JL (2016) The mrmr variable selection method: a comparative study for functional data. J Stat Comput Simul 86(5):891–907

    Article  Google Scholar 

  5. 5.

    Billah M, Waheed S (2018) Gastrointestinal polyp detection in endoscopic images using an improved feature extraction method. Biomed Eng Lett 8(1):69–75

    Article  Google Scholar 

  6. 6.

    Billah M, Waheed S, Rahman MM (2017) An automatic gastrointestinal polyp detection system in video endoscopy using fusion of color wavelet and convolutional neural network features. Int J Biomed Imaging 2017

  7. 7.

    Ding C, Peng H (2005) Minimum redundancy feature selection from microarray gene expression data. Journal of Bioinformatics and Computational Biology 3(02):185–205

    Article  Google Scholar 

  8. 8.

    Farah FMBD, Wahid KA (2020) Computer-aided polyp detection based on image enhancement and saliency-based selection. Biomedical Signal Processing and Control 55:101530

    Article  Google Scholar 

  9. 9.

    Girshick TDR, Donahue J, Malik J (2014) Rich feature hierarchies for accurate object detection and semantic segmentation CVPR

  10. 10.

    Iakovidis DK, Maroulis DE, Karkanis SA (2006) An intelligent system for automatic detection of gastrointestinal adenomas in video endoscopy. Comput Biol Med 36(10):1084–1103

    Article  Google Scholar 

  11. 11.

    Karkanis SA, Iakovidis DK, Maroulis DE, Karras DA, Tzivras M (2003) Computer-aided tumor detection in endoscopic video using color wavelet features. IEEE Trans Inf Technol Biomedicine 7(3):141–152

    Article  Google Scholar 

  12. 12.

    Kodogiannis V, Boulougoura M (2007) An adaptive neurofuzzy approach for the diagnosis in wireless capsule endoscopy imaging. Int J Inf Technol 13(1):46–56

    Google Scholar 

  13. 13.

    Kopelman OGHJPSY, Cohen A (2019) Automated polyp detection system in colonoscopy using deep learning and image processing techniques. Journal of Gastroenterology and its Complications 3(1):101

    Google Scholar 

  14. 14.

    Li B, Fan Y, Meng MQ-H, Qi L (2009) Intestinal polyp recognition in capsule endoscopy images using color and shape features. In: 2009 IEEE international conference on robotics and biomimetics (ROBIO). IEEE, pp 1490–1494

  15. 15.

    Li B, Qi L, Meng MQ-H, Fan Y (2009) Using ensemble classifier for small bowel ulcer detection in wireless capsule endoscopy images. In: 2009 IEEE international conference on robotics and biomimetics (ROBIO). IEEE, pp 2326–2331

  16. 16.

    Liu DECSSRCYFW, Anguelov D, Berg AC (2016) Ssd: single shot multibox detector ECCV

  17. 17.

    Mehmood R, Shahzad W, Ahmed E (2017) Maximum relevancy minimum redundancy based feature subset selection using ant colony optimization. J Appl Environ Biol Sci 7(4):118–130

    Google Scholar 

  18. 18.

    Mesejo P, Pizarro D, Abergel A, Rouquette O, Beorchia S, Poincloux L, Bartoli A (2016) Computer-aided classification of gastrointestinal lesions in regular colonoscopy. IEEE Trans Medical Imaging 35(9):2051–2063

    Article  Google Scholar 

  19. 19.

    Mingjian XZGQMZHDLMS, Qu Y (2019) Automatic polyp detection in colonoscopy images: convolutional neural network, dataset and transfer learning. Journal of Medical Imaging and Health Informatics 9(1):126–133

    Article  Google Scholar 

  20. 20.

    Nagito ATJHS, Shibata T (2019) Automated detection of fundic gland polyps from endoscopic images using ssd. International Society for Optics and Photonics 11049:110492O

    Google Scholar 

  21. 21.

    Peng H, Long F, Ding C (2005) Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy. IEEE Trans Pattern Anal Mach Intell 27(8):1226–1238

    Article  Google Scholar 

  22. 22.

    Radovic M, Ghalwash M, Filipovic N, Obradovic Z (2017) Minimum redundancy maximum relevance feature selection approach for temporal gene expression data. BMC Bioinformatics 18(1):9

    Article  Google Scholar 

  23. 23.

    Redmon J, Farhadi A (2016) Yolo9000: better, faster, stronger. arXiv:1612.08242

  24. 24.

    Ribeiro E, Uhl A, Häfner M (2016) Colonic polyp classification with convolutional neural networks. In: 2016 IEEE 29th international symposium on computer-based medical systems (CBMS). IEEE, pp 253–258

  25. 25.

    Shen H-B, Chou K-C (2006) Ensemble classifier for protein fold pattern recognition. Bioinformatics 22(14):1717–1722

    Article  Google Scholar 

  26. 26.

    Sudhir FMS, Yi S (2019) Region-based automated localization of colonoscopy and wireless capsule endoscopy polyps. Appl Sci 9(12):2404

    Article  Google Scholar 

  27. 27.

    Unler A, Murat A, Chinnam RB (2011) Mr2pso: a maximum relevance minimum redundancy feature selection method based on swarm intelligence for support vector machine classification. Inf Sci 181(20):4625–4641

    Article  Google Scholar 

  28. 28.

    Vergara JR, Estévez PA (2014) A review of feature selection methods based on mutual information. Neural computing and applications 24(1):175–186

    Article  Google Scholar 

  29. 29.

    Zhu R, Zhang R, Xue D (2015) Lesion detection of endoscopy images based on convolutional neural network features. In: 2015 8th international congress on image and signal processing (CISP). IEEE, pp 372–376

  30. 30.

    Zou Y, Li L, Wang Y, Yu J, Li Y, Deng WJ (2015) Classifying digestive organs in wireless capsule endoscopy images based on deep convolutional neural network. In: 2015 IEEE international conference on digital signal processing (DSP). IEEE, pp 1274–1278

Download references

Acknowledgments

The authors are very grateful to Dr. Md. Rabiul Hossain, Gastroenterologist, Liver & Internal Medicine Specialist, Labaid Hospital, Dhaka for his valuable support, suggestions, and consultancy.

Funding

All the research work in this paper has been conducted by self funding.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Mustain Billah.

Ethics declarations

All procedures performed in this study involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Conflict of interests

The authors declare that, there is no conflict of interest regarding this paper.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Billah, M., Waheed, S. Minimum redundancy maximum relevance (mRMR) based feature selection from endoscopic images for automatic gastrointestinal polyp detection. Multimed Tools Appl (2020). https://doi.org/10.1007/s11042-020-09151-7

Download citation

Keywords

  • Minimum redundancy maximum relevance (mRMR)
  • Video endoscopy
  • Ensemble classifier
  • Feature selection
  • Convolutional Neural Network (CNN)
  • Color Wavelet (CW)
  • Feature extraction