Segmentation of Edema in HGG MR Images Using Convolutional Neural Networks

  • S. Poornachandra
  • C. Naveena
  • Manjunath Aradhya
  • K. B. Nagasundara
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 695)

Abstract

In this paper, we present the segmentation of edema subregion in the high-grade gliomas (HGGs) MR images We use the T1, T2, FLAIR, and T1c MRI modalities in our work and employ convolutional neural network approach (CNN) architecture for the segmentation task. Preprocessing was done in each case by correcting the inhomogeneities in MR images, equalizing histogram, and applying Z-score normalization to all the volumes. Here, we experimented with the convolutional layers, activation functions, and max-pooling layers. The CNN was trained on 3D patches extracted from patient volumes. Our experimentation has given promising results with mean dice score of 0.68, positive predicted value of 0.63, and sensitivity of 0.65 for edema segmentation.

Keywords

Gliomas CNN MRI Dice score PPV and sensitivity 

References

  1. 1.
    Menze, B.H., et al.: The multimodal brain tumor image segmentation benchmark (BRATS). IEEE Trans. Med. Imaging 34(10), 1993–2024 (2015)CrossRefGoogle Scholar
  2. 2.
    Brain Tumors. N. Engl. J. Med. (2001)Google Scholar
  3. 3.
    De Angelis L.M.: Brain tumors, N. Engl. Med. 344, 114–23 (2001)Google Scholar
  4. 4.
    Deimling, A.: Gliomas. Recent Results in Cancer Research. Springer, Berlin (2009)CrossRefGoogle Scholar
  5. 5.
    Havaei, M., et al.: Deep learning trends for focal brain pathology segmentation in MRI. In: Machine Learning for Health Informatics: State-of-the-Art and Future Challenges, pp. 125–148. Springer International Publishing (2016)Google Scholar
  6. 6.
    Malmi, E., et al.: CaBS: a cascaded brain tumor segmentation approach. In: Proceedings MICCAI Brain, Tumor Segmentation(BRATS), pp. 42–47 (2015)Google Scholar
  7. 7.
    Mikael A. et al.: Brain tumor segmentation by a generative model with a prior on tumor shape. In: Proceedings MICCAI-BRATS, pp. 1–4 (2015)Google Scholar
  8. 8.
    Festa, J., et al.: Automatic brain tumor segmentation of multi-sequence MR images using random decision forests, pp. 23–26. In: Proceedings MICCAI Brain, Tumor Segmentation(BRATS) (2013)Google Scholar
  9. 9.
    Tustison, N., et al.: ANTs and Arboles. In: Proceedings MICCAI Brain Tumor Segmentation (BRATS) (2013)Google Scholar
  10. 10.
    Clark, M., et al.: Automatic tumor segmentation using knowledge-based techniques. IEEE Trans. Med. Imaging 17, 187201 (1998)CrossRefGoogle Scholar
  11. 11.
    Prastawa, M., et al.: Automatic brain tumor segmentation by subject specific modification of atlas priors. J. Acad. Radiol. (2003)Google Scholar
  12. 12.
    Subbanna, N., Arbel, T.: Probabilistic gabor and markov random fields segmentation of brain tumours in mri volumes. In: Proceedings MICCAI Brain Tumor Segmentation Challenge (BRATS), pp. 28–31 (2012)Google Scholar
  13. 13.
    Shin, H.-C.: Hybrid clustering and logistic regression for multi-modal brain tumor segmentation. In: Proceedings of Workshops and Challanges in Medical Image Computing and Computer-Assisted Intervention (MICCAI12) (2012)Google Scholar
  14. 14.
    Song, B., et al.: Anatomy-guided brain tumor segmentation and classification. In: International Workshop on Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries. Springer, Cham (2016)CrossRefGoogle Scholar
  15. 15.
    Zikic, D., et al.: Decision forests for tissue-specific segmentation of high-grade gliomas in multi-channel MR. In: International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, Berlin, Heidelberg (2012)CrossRefGoogle Scholar
  16. 16.
    Abdel Maksoud, Eman A et al.: 3D Brain tumor segmentation based on hybrid clustering techniques using multi-views of MRI. In: Medical Imaging in Clinical Applications: Algorithmic and Computer-Based Approaches. Springer Publications (2016)Google Scholar
  17. 17.
    Tustison, N.J., et al.: N4ITK: improved N3 bias correction. IEEE Trans. Med. Imaging 29(6), 1310–1320 (2010)CrossRefGoogle Scholar
  18. 18.
    Nyl, L.G., Udupa, J.K., Zhang, X.: New variants of a method of MRI scale standardization. IEEE Trans. Med. Imaging 19(2) 143–150 (2000)CrossRefGoogle Scholar
  19. 19.
    Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: Adv. Neural Inf. Process. Syst. (2012)Google Scholar
  20. 20.
    Bergstra, J., et al.: Theano: A CPU and GPU math compiler in Python. In: Proceedings 9th Python in Science Conference (2010)Google Scholar
  21. 21.
    Pereira, S., et al.: Deep convolutional neural networks for the segmentation of gliomas in multi-sequence MRI. In: International Workshop on Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries. Springer, Cham (2015)Google Scholar
  22. 22.
    Kamnitsas, K., et al.: DeepMedic for brain tumor segmentation. In: International Workshop on Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries. Springer, Cham (2016)CrossRefGoogle Scholar
  23. 23.
    Rexilius, J., et al.: Multispectral brain tumor segmentation based on histogram model adaptation. In: Proceedings SPIE 6514, Medical Imaging (2007)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2018

Authors and Affiliations

  • S. Poornachandra
    • 1
  • C. Naveena
    • 2
  • Manjunath Aradhya
    • 3
  • K. B. Nagasundara
    • 4
  1. 1.Department of Computer Science and ApplicationsVTU RRCBelgaumIndia
  2. 2.Department of Computer Science and EngineeringSJB Institute of TechnologyBengaluruIndia
  3. 3.Department of MCASri Jayachamarajendra College of EngineeringMysoreIndia
  4. 4.Department of Computer Science and EngineeringJSS Academy of Technical EducationBengaluruIndia

Personalised recommendations