Semantic Segmentation Using Deep Learning for Brain Tumor MRI via Fully Convolution Neural Networks

  • Sanjay Kumar
  • Ashish Negi
  • J. N. Singh
Conference paper
Part of the Smart Innovation, Systems and Technologies book series (SIST, volume 106)


In this paper, premature head lump recognition along with analysis is dangerous to clinic. Therefore, segmentation of paying attention to growth neighborhood desires near subsists precise, efficient, and robust. Convolution system is authoritative illustration model with the purpose of capitulate skin tone. Researchers explain to intricacy complex with taught continuous pixels and top condition and image in semantic. According to research contribution approaching, the make completely convolution system with the intention obtain participation of random dimension and manufacture correspondingly sized production with resourceful supposition and knowledge. We describe and element the breathing liberty and entirely convolution system clarify describe function toward special impenetrable estimate everyday jobs in addition rough copy family member and preceding reproduction. We are acclimatizing fashionable arrangement network which is keen on fully convolution networks with relocating their knowledgeable representation by modification to the segmentation assignment. We describe a bounce structural chart to facilitate collect semantic requirement starting with a profound uncouth deposit through exterior in sequence following low, well coating toward construct precise in addition and thorough segmentation. This is the FCN attain circumstance of the segmentation and 36% similar development toward 66.6% indicate lying 2015 NYUD with pass through a filter present although deduction take a smaller amount single fifth and succeeding on behalf of the characteristic picture. According to researches, they designed a three-dimensional fully convolution neural network for brain tumor segmentation. During training, researchers optimized our network alongside beating purpose based on gamble achieve results and researchers also used to assess the superiority of prediction twisted in this representation. In order to accommodate the massive memory requirements of three-dimensional convolutions, we cropped the images we fed into our network, and we used a UNET architecture that allowed us to achieve good results even with a relatively narrow and shallow neural network. Finally, we used post-processing in order to smooth out the segmentations produced by our model.


Image segmentation Classification MRI (magnetic resonance imaging) scan Fully convolution neural networks 



Researchers examine the BCN (Byse Code Normal) somewhat outperforms the FCN. This is the majority probable owing to the BCN (Byse Code Normal) relying less on the precise image skin tone than do the FCN. In the FCN, convolution layer squeeze the picture as elevated skin tone, then the volitional layer reconstruct the segmented picture on or after this covering tone [12]. While this is a potentially tremendously powerful structural plan, the unspoken supposition is that distant on top of the earth level skin for comparable imagery determination as well comparable [9]. Though, the dissimilarity in excellence and declaration of the Rembrandt imagery compare to the brat images income this supposition possibly will not hold extremely powerfully, and consequently see a go to sleep in segmentation excellence sandwiched sand wiched among the BraTS and Rembrandt datasets [13]. In conclusion, the consequences demonstrate that the segmentation superiority is very conflicting transversely the legalization set [8]. We scrutinize together exceptionally elevated and exceptionally short segmentation quality. The FCN model in meticulous have sample inside each container crossways the histogram. This show that at the same time as transport knowledge may be shows impending for application to segmentation, it is not automatically dependable and is motionless extremely reliant on picture excellence and declaration [8].


  1. 1.
    Shen, L., Anderson, T.: Multimodal Brain MRI Tumor Segmentation via Convolution Neural NetworksGoogle Scholar
  2. 2.
    Havaei, M., Davy, A., Warde-Farley, D., Biard, A., Courville, A., Bengio, Y., Pal, C., Jodoin, P.-M., Larochelle, H.: Brain tumor segmentation with deep neural networks. Med. Image Anal. 35, 18–31 (2017)Google Scholar
  3. 3.
    Sharma, K., Kaur, A., Gujral, S.: Brain tumor detection based on machine learning algorithms. Int. J. Comput. Appl. 103(1) (2014)Google Scholar
  4. 4.
    Xiao, Z., Huang, R., Ding, Y., Lan, T., Dong, R.F., Qin, Z., Zhang, X., Wang, W.: A deep learning-based segmentation method for brain tumor in MR images. In: 2016 IEEE 6th International Conference on Computational Advances in Bio and Medical Sciences (ICCABS), pp. 1–6. IEEE (2016)Google Scholar
  5. 5.
    Garcia-Garcia, A., Orts-Escolano, S., Oprea, S., Villena-Martinez, V., Garcia-Rodriguez, J.: A Review on Deep Learning Techniques Applied to Semantic Segmentation. arXiv:1704.06857 (2017)
  6. 6.
    Ciresan, D., Giusti, A., Gambardella, L., Schmidhuber, J.: Deep neural networks segment neuronal membranes in electron microscopy images. Nips, pp. 1–9 (2012)Google Scholar
  7. 7.
    Usman, K., Rajpoot, K.: Brain tumor classification from multi-modality MRI using wavelets and machine learning. Pattern Anal. Appl. 1–11 (2017)Google Scholar
  8. 8.
    Kayalibay, B., Jensen, G., van der Smagt, P.: CNN-based Segmentation of Medical Imaging Data (2017)Google Scholar
  9. 9.
    Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., Fei-Fei, L.: ImageNet: a large-scale hierarchical image database. In: CVPR (2009)Google Scholar
  10. 10.
    Bauer, S., Nolte, L.-P., Reyes, M.: Fully automatic segmentation of brain tumor images using support vector machine classification in combination with hierarchical conditional random field regularization. In: International Conference on Medical Image Computing and Computer-Assisted Intervention, pp. 354–361. Springer, Berlin, Heidelberg (2011)Google Scholar
  11. 11.
    Menze, B.H., et al.: The multimodal brain tumor image segmentation benchmark (BRATS). IEEE Trans. Med. Imag. 34(10), 1993–2024 (2015)CrossRefGoogle Scholar
  12. 12.
    Long, J., Shelhamer, E., Darrell, T.: Fully convolutional networks for semantic segmentation. In: CVPR (2015) (to appear)Google Scholar
  13. 13.
    Bahadure, N.B., Ray, A.K., Thethi, H.P.: Image analysis for MRI based brain tumor detection and feature extraction using biologically inspired BWT and SVM. Int. J. Biomed. Imag. (2017)Google Scholar
  14. 14.
    Bauer, S., Wiest, R., Nolte, L.-P., Reyes, M.: A survey of MRI-based medical image analysis for brain tumor studies. Phys. Med. Biol. 58(13), R97 (2013)Google Scholar
  15. 15.
    Dvorak, P., Menze, B.H.: Local structure prediction with convolutional neural networks for multimodal brain tumor segmentation. In: MCV@ MICCAI, pp. 59–71 (2015)Google Scholar
  16. 16.
    Noh, H., Hong, S., Han, B.: Learning deconvolution network for semantic segmentation. In: Proceedings of the IEEE International Conference on Computer Vision, 11–18 Dec 2015, pp. 1520–1528 (2016)Google Scholar
  17. 17.
    Reese, T.G., Heid, O., Weisskoff, R.M., Wedeen, V.J.: Reduction of eddy-current-induced distortion in diffusion MRI using a twice-refocused spin echo. Magn. Reason. Med. 49(1), 177–182 (2003)CrossRefGoogle Scholar
  18. 18.
    Long, J., Shelhamer, E., Darrell, T.: Fully convolutional networks for semantic segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 3431–3440 (2015)Google Scholar
  19. 19.
    Menze, B.H., Jakab, A., Bauer, S., Kalpathy-Cramer, J., Farahani, K., Kirby, K., Burren, Y., et al.: The multimodal brain tumor image segmentation benchmark (BRATS). IEEE Trans. Med. Imag. 34(10), 1993–2024 (2015)Google Scholar
  20. 20.
    Ronneberger, O., Fischer, P., Brox, T.: U-Net: Convolutional Networks for Biomedical Image Segmentation, pp. 1–8 (2015)Google Scholar
  21. 21.
    Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. CoRR, abs/1409.1556 (2014)Google Scholar
  22. 22.
    Rathi, V.P., Palani, S.: Brain tumor MRI image classification with feature selection and extraction using linear discriminant analysis. arXiv:1208.2128 (2012)
  23. 23.
    Kumar, S., Srivastava, S.: Image encryption using s-des based on Arnold cat map using logistic map. Int. J. Bus. Eng. Res. 8 (2014)Google Scholar
  24. 24.
    Scarpace, L., Flanders, A.E., R. Jain, T. Mikkelsen, Andrews, D.W.: Data from rembrandt (2015)Google Scholar
  25. 25.
    Ren, S., He, K., Girshick, R., Sun, J.: Faster R-CNN: to-wards concurrent object uncovering with district proposal network. Adv. Neural Seq. Process. Sys-t. (NIPS) (2015)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  1. 1.Uttarakhand Technical University DehradunSudhowalaIndia
  2. 2.G.B. Pant Engineering CollegePauri GarhwalIndia
  3. 3.Galgotias UniversityGreater NoidaIndia

Personalised recommendations