Detection of Early Gastric Cancer from Endoscopic Images Using Wavelet Transform Modulus Maxima

Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 738)

Abstract

It is said that the overlooking rate of early gastric cancer in endoscopic examination reaches 20–25% in Japan, and it is desirable to develop a detection method for early gastric cancer from endoscopic images to reduce the overlooking rate. We propose a new method for detecting early gastric cancer from endoscopic images using the wavelet transform modulus maxima (WTMM). First, our method converts the original image into the CIE L*a*b* color space. Next, we apply the dyadic wavelet transform (DYWT) to the a* component image and compute the WTMM of the high frequency component. It is shown that the WTMM of the abnormal parts tends to become smaller than the WTMM of the normal parts. We describe the method detecting the abnormal parts based on these features in detail, we show experimental results demonstrating that the proposed method are able to detect the regions suspected of being early gastric cancer from endoscopic images.

Keywords

Image analysis Frequency analysis Early gastric cancer Dyadic wavelet transform Wavelet transform modulus maxima 

References

  1. 1.
    IARC, Stomach cancer estimated incidence, mortality and prevalence worldwide in 2012. http://globocan.iarc.fr/old/FactSheets/cancers/stomach-new.asp
  2. 2.
    T. Aoyama, T. Yoshikawa, T. Watanabe, T. Hayashi, T. Ogata, H. Cho, A. Tsuburaya, Macroscopic tumor size as an independent prognostic factor for stage II/III gastric cancer patients who underwent D2 gastrectomy followed by adjuvant chemotherapy with S-1. Gastric Cancer 14, 274–278 (2011)CrossRefGoogle Scholar
  3. 3.
    M. Serrano, I. Kikuste, M. Dinis-Ribeiro, Advanced endoscopic imaging for gastric cancer assessment: new insights with new optics? Best Pract. Res. Clin. Gastroenterol. 28, 1079–1091 (2014)CrossRefGoogle Scholar
  4. 4.
    M. Kaise, Advanced endoscopic imaging for early gastric cancer. Best Pract. Res. Clin. Gastroenterol. 29, 575–587 (2015)CrossRefGoogle Scholar
  5. 5.
    M. Song, T.L. Ang, Early detection of early gastric cancer using image-enhanced endoscopy: current trends. Gastrointest. Interv. 3, 1–7 (2014)CrossRefGoogle Scholar
  6. 6.
    Y. Morimoto, M. Kubo, M. Kuramoto, H. Yamaguchi, T. Kaku, Development of a New Generation Endoscope System with Lasers “LASEREO”. Fujifilm Res. Dev. 58, 6 pp. (2013)Google Scholar
  7. 7.
    T. Hu, Y.H. Lu, C.G. Cheng, X.C. Sun, Study on the early detection of gastric cancer based on discrete wavelet transformation feature extraction of FT-IR spectra combined with probability neural network. Spectroscopy 26(3), 155–165 (2011)CrossRefGoogle Scholar
  8. 8.
    H. Matsunaga, H. Omura, R. Ohura, T. Minamoto, Daubechies wavelet-based method for early esophageal cancer detection from flexible spectral imaging color enhancement image. Adv. Intell. Syst. Comput. 448, 939–948 (2016)Google Scholar
  9. 9.
    T. Minamoto, K. Tsuruta, S. Fujii, Edge-preserving image denoising method based on dyadic lifting schemes. IPSJ Trans. Comput. Vis. Appl. 2, 48–58 (2010)CrossRefGoogle Scholar
  10. 10.
    T. Minamoto, R. Ohura, A blind digital image watermarking method based on the dyadic wavelet transform and interval arithmetic. Appl. Math. Comput. 226, 306–319 (2014)MATHGoogle Scholar
  11. 11.
    R. Ohura, H. Omura, Y. Sakata, T. Minamoto, Computer-aided diagnosis method for detecting early esophageal cancer from endoscopic image by using dyadic wavelet transform and fractal dimension. Adv. Intell. Syst. Comput. 448, 929–938 (2016)Google Scholar
  12. 12.
    S. Mallat, A Wavelet Tour of Signal Processing: The Sparse Way, 3rd edn. (Academic, Burlington, MA, 2008)MATHGoogle Scholar
  13. 13.
    S. Mallat, S. Zhong, Characterization of signals from multiscale edges. IEEE Trans. Pattern Anal. Mach. Intell. 14(7), 710–732 (1992)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Department of Information ScienceSaga UniversitySagaJapan

Personalised recommendations