An Automatic Brain Tumor Segmentation Approach Based on Affinity Clustering

  • C. RamirezEmail author
  • V. Gomez
  • I. De la Pava
  • A. Alvarez
  • J. Echeverry
  • J. Rios
  • A. Orozco
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11401)


The computational methods for tumor segmentation from Magnetic Resonance Images (MRI) are useful tools that aim to support medical specialists in tumor detection. High-grade brain tumors present extremely heterogeneous properties and are associated with elevated mortality rates. Therefore, their segmentation constitutes a challenging and relevant research task. However, most unsupervised methodologies for brain tumor segmentation rely on adequate parameter initialization by the user to achieve satisfactory results. Here, we propose a novel automatic brain tumor core segmentation methodology based on affinity clustering that overcomes that shortcoming. Obtained results in a public database show that our approach is competitive with the state-of-the-art regarding the Dice scores between the segmented tumors and their corresponding Ground Truth images.


Affinity Propagation Brain tumor Automatic segmentation 



Under grants provided by the project 1110-744-55860, funded by Colciencias. C. Ramirez was funded under the project E6-18-10 funded by the VIIE and by the Master in Electrical Engineering-Universidad Tecnológica de Pereira. V. Gomez and I. de la Pava was supported by the program “Doctorado Nacional en Empresa - Convoctoria 758 de 2016”, funded by Colciencias. Also, J. Echeverry is supported by the project Metodología para el reconocimiento y la traducción de señas aisladas en la Lengua de Señas Colombiana utilizando técnicas de visión por computador-6-16-4, funded by the VIIE-Universidad Tecnologica de Pereira.


  1. 1.
    Achanta, R., et al.: Slic superpixels compared to state-of-the-art superpixel methods. IEEE Trans. Pattern Anal. Mach. Intell. 34(11), 2274–2282 (2012)CrossRefGoogle Scholar
  2. 2.
    Angulakshmi, M., Priya, G.L.: Brain tumour segmentation from MRI using superpixels based spectral clustering. J. King Saud Univ.-Comput. Inf. Sci. (2018)Google Scholar
  3. 3.
    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)CrossRefGoogle Scholar
  4. 4.
    Bezdek, J., Hall, L., Clark, M., Goldgof, D.B., Clarke, L.: Medical image analysis with fuzzy models. Stat. Method Med. Res. 6(3), 191–214 (1997)CrossRefGoogle Scholar
  5. 5.
    Boughattas, N., Berar, M., Hamrouni, K., Ruan, S.: Feature selection and classification using multiple kernel learning for brain tumor segmentation. In: 2018 4th International Conference on Advanced Technologies for Signal and Image Processing (ATSIP), pp. 1–5. IEEE (2018)Google Scholar
  6. 6.
    Dueck, D., Frey, B.J.: Non-metric affinity propagation for unsupervised image categorization. In: 2007 IEEE 11th International Conference on Computer Vision, ICCV 2007, pp. 1–8. IEEE (2007)Google Scholar
  7. 7.
    Emblem, K.E., Nedregaard, B., Hald, J.K., Nome, T., Due-Tonnessen, P., Bjornerud, A.: Automatic glioma characterization from dynamic susceptibility contrast imaging: brain tumor segmentation using knowledge-based fuzzy clustering. J. Magn. Reson. Imaging 30(1), 1–10 (2009)CrossRefGoogle Scholar
  8. 8.
    Gordillo, N., Montseny, E., Sobrevilla, P.: State of the art survey on MRI brain tumor segmentation. Magn. Reson. Imaging 31(8), 1426–1438 (2013)CrossRefGoogle Scholar
  9. 9.
    Menze, B., et al.: The multimodal brain tumor image segmentation benchmark (BRATS). IEEE Trans. Med. Imaging, 33 (2014).
  10. 10.
    Nabizadeh, N., Kubat, M.: Automatic tumor segmentation in single-spectral MRI using a texture-based and contour-based algorithm. Expert Syst. Appl. 77, 1–10 (2017)CrossRefGoogle Scholar
  11. 11.
    Paul, T.U., Bandhyopadhyay, S.K.: Segmentation of brain tumor from brain MRI images reintroducing k-means with advanced dual localization method. Int. J. Eng. Res. Appl. 2(3), 226–231 (2012)Google Scholar
  12. 12.
    Vishnuvarthanan, A., Rajasekaran, M.P., Govindaraj, V., Zhang, Y., Thiyagarajan, A.: An automated hybrid approach using clustering and nature inspired optimization technique for improved tumor and tissue segmentation in magnetic resonance brain images. Appl. Soft Comput. 57, 399–426 (2017)CrossRefGoogle Scholar
  13. 13.
    Zhao, L., Wu, W., Corso, J.J.: Brain tumor segmentation based on GMM and active contour method with a model-aware edge map. In: BRATS MICCAI, pp. 19–23 (2012)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • C. Ramirez
    • 1
    Email author
  • V. Gomez
    • 1
  • I. De la Pava
    • 1
  • A. Alvarez
    • 1
  • J. Echeverry
    • 1
  • J. Rios
    • 1
  • A. Orozco
    • 1
  1. 1.Automatic Research Group, Faculty of EngineeringsUniversidad Tecnológica de PereiraPereiraColombia

Personalised recommendations