Advertisement

Brain Tumor Segmentation from MRI Images Using Deep Learning Framework

  • Suchismita DasEmail author
Conference paper
  • 15 Downloads
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 1119)

Abstract

Brain tumors are the most common and aggressive diseases which lead to a very short life expectancy in their highest grade. Thus, automatic brain tumor detection is necessary at early stage to reduce death frequency due to this. Magnetic Resonance Imaging (MRI) is a widely used imaging technique to assess but it is not practically possible to perform manual segmentation on the large amount of data produced by MRI in due time. This paper focuses on tumor detection from magnetic resonance images having dataset of brain images to classify the tumor and non-tumor symptoms using deep learning fully convolution neural network. In particular, it has been developed using image enhancement, segmentation, and classification techniques. The U-Net architecture is used for segmenting the tumor followed by fully CNN to classify the extracted portion to improve the performance. The outcomes are recorded and compared with existing techniques and found the proposed framework gives better performance by 10%. This method can be extended in 3D images using standard brain tumor database.

Keywords

Medical image processing Deep learning U-Net architecture Brain tumor segmentation Fully convolutional neural networks 

References

  1. 1.
    El-Dahshan, E.A., Mohsen, H.M., Revett, K., Salem, A.M.: Computer-aided diagnosis of human brain tumor through MRI: a survey and a new algorithm. Expert Syst. Appl. 41, 5526–5545 (2014)Google Scholar
  2. 2.
    Angulakshmi, M., LaxmiPriya, G.: Automated brain tumour segmentation techniques—a review. Int. J. Imaging Syst. Technol. 27, 66–77 (2017)Google Scholar
  3. 3.
    Jiang, Y., Chi1, Z: A scale-invariant framework for image classification with deep learning. In: IEEE International Conference on Systems, Man, and Cybernetics (SMC), pp. 1019–1024. Banff Center, Canada, 5–8 October (2017)Google Scholar
  4. 4.
    Bernal, J., Kushibar, K., Cabezas, M., Valverde, S., Oliver, A., Llad, X.: Quantitative analysis of patch-based fully convolutional neural networks for tissue segmentation on brain magnetic resonance imaging (2018). arXiv:1801.06457
  5. 5.
    Aghdam, H.H., Heravi, E.J.: Guide to Convolutional Neural Networks, pp. 247–258 (2017)Google Scholar
  6. 6.
    Elnoor, H., Abdalla, M. Esmail, M.: Brain tumor detection by using artificial neural network. In: International Conference on Computer, Control, Electrical, and Electronics Engineering (2018)Google Scholar
  7. 7.
    Telrandhe, S.V., Chikate, D., Banode, P.: Automated brain tumor detection using back propagation neural network. Int, J, Soft Comput. Artif. Intell. 3 (2015)Google Scholar
  8. 8.
    Zin, S., Khaing, A.S.: Brain tumor detection and segmentation using watershed segmentation and morphological operation. Int. J. Res. Eng. Technol. 3, 367–374 (2014)Google Scholar
  9. 9.
    Wankhade, A., Malviya, A.V.: Brain tumor detection using K-mean clustering and SVM. Int. Res. J. Eng. Technol 5, 3186–3194 (2018)Google Scholar
  10. 10.
    Dong, H., Yang, G., Liu, F., Mo1, Y., Guo, Y.: Automatic brain tumor detection and segmentation using U-Net based fully convolutional networks. In: Conference on Medical Image Understanding and Analysis (2017)Google Scholar
  11. 11.
    Agarap, A.: Deep learning using rectified linear units (ReLU) (2018). https://arxiv.org/pdf/1803.08375.pdf
  12. 12.
    Bakas, S., Akbari, H., Sotiras, A., Bilello, M., Rozyscki, M., Kirby, S., Davatzikos, C.: Segmentation labels and radiomic features for the pre-operative scans from the TCGA-GBM collection. The Cancer Imaging Archive (2017)Google Scholar
  13. 13.
    Otsu, N.: A threshold selection method from gray-level histograms. IEEE Trans. Syst. Man Cybern. 9, 62–66 (2007)Google Scholar
  14. 14.
    Noh, H., Hong, S., Han, B.: Learning deconvolution network for semantic segmentation. In: IEEE International Conference on Computer Vision, pp. 1520–1528. Santiago, Chile (2015)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2020

Authors and Affiliations

  1. 1.KIIT Deemed to be UniversityBhubaneswarIndia

Personalised recommendations