Towards Large-Scale Fiber Orientation Models of the Brain – Automation and Parallelization of a Seeded Region Growing Segmentation of High-Resolution Brain Section Images

  • Anna LührsEmail author
  • Oliver Bücker
  • Markus Axer
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10087)


To understand the microscopical organization of the human brain including cellular and fiber architectures, it is a necessary prerequisite to build virtual models of the brain on a sound biological basis. 3D Polarized Light Imaging (3D-PLI) provides a window to analyze the fiber architecture and the fibers’ intricate inter-connections at microscopic resolutions. Considering the complexity and the pure size of the human brain with its nearly 86 billion nerve cells, 3D-PLI is challenging with respect to data handling and analysis in the TeraByte to PetaByte ranges, and inevitably requires supercomputing facilities. Parallelization and automation of image processing steps open up new perspectives to speed up the generation of new high resolution models of the human brain to provide groundbreaking insights into the brain’s three-dimensional micro architecture. Here, we will describe the implementation and the performance of a parallelized semi-automated seeded region growing algorithm used to classify tissue and background components in up to one million 3D-PLI images acquired from an entire human brain. This algorithm represents an important element of a complex UNICORE-based analysis workflow ultimately aiming at the extraction of spatial fiber orientations from 3D-PLI measurements.


Polarized light imaging Fiber architecture Human brain Workflow Region growing Segmentation Scaling Supercomputing 



This work was partially supported by the Helmholtz Association portfolio theme “Supercomputing and Modeling for the Human Brain” and by the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement no. 604102 (Human Brain Project).


  1. [AAG+11]
    Axer, M., Amunts, K., Gräßel, D., Palm, C., Dammers, J., Axer, H., Pietrzyk, U., Zilles, K.: A novel approach to the human connectome: ultra-high resolution mapping of fiber tracts in the brain. NeuroImage 54, 1091–1101 (2011)CrossRefGoogle Scholar
  2. [AB94]
    Adams, R., Bischof, L.: Seeded region growing. IEEE Trans. Pattern Anal. Mach. Intell. 16(6), 641–647 (1994)CrossRefGoogle Scholar
  3. [ABA14]
    Amunts, K., Bücker, O., Axer, M.: Towards a multiscale, high-resolution model of the human brain. In: Grandinetti, L., Lippert, T., Petkov, N. (eds.) BrainComp 2013. LNCS, vol. 8603, pp. 3–14. Springer, Heidelberg (2014). doi: 10.1007/978-3-319-12084-3_1 Google Scholar
  4. [AGK+11]
    Axer, M., Gräßel, D., Kleiner, M., Dammers, J., Dickscheid, T., Reckfort, J., Hütz, T., Eiben, B., Pietrzyk, U., Zilles, K., Amunts, K.: High-resolution fiber tract reconstruction in the human brain by means of three-dimensional polarized light imaging (3D-PLI). Front. Neuroinformatics 5, 34 (2011)Google Scholar
  5. [BJHD96]
    Bader, D.A., Jájá, J., Harwood, D., Davis, L.S.: Parallel algorithms for image enhancement and segmentation by region growing, with an experimental study. J. Supercomputing 10(2), 141–168 (1996)CrossRefGoogle Scholar
  6. [GSAW05]
    Grady, L., Schiwietz, T., Aharon, S., Westermann, R.: Random walks for interactive organ segmentation in two and three dimensions: implementation and validation. In: Duncan, J.S., Gerig, G. (eds.) MICCAI 2005. LNCS, vol. 3750, pp. 773–780. Springer, Heidelberg (2005). doi: 10.1007/11566489_95 CrossRefGoogle Scholar
  7. [HFB+05]
    Happ, P.N., Ferreira, R.S., Bentes, C., Costa, G., Feitosa, R.Q.: Multiresolution segmentation: a parallel approach for high resolution image segmentation in multicore architectures. In: The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, XXXVIII-4/C7, 28–32 (2005)Google Scholar
  8. [HFBF12]
    Happ, P.N., Feitosa, R.Q., Bentes, C., Farias, R.: A parallel image segmentation algorithm on GPUs. In: Proceedings of the 4th GEOBIA, pp. 580–585, May 2012Google Scholar
  9. [Hua92]
    Huang, C.-L.: Parallel image segmentation using modified Hopfield model. Pattern Recogn. Lett. 13(5), 345–353 (1992)CrossRefGoogle Scholar
  10. [HZ09]
    Hagan, A., Zhao, Y.: Parallel 3D Image segmentation of large data sets on a GPU cluster. In: Bebis, G., et al. (eds.) ISVC 2009. LNCS, vol. 5876, pp. 960–969. Springer, Heidelberg (2009). doi: 10.1007/978-3-642-10520-3_92 CrossRefGoogle Scholar
  11. [KB90]
    Khotanzad, A., Bouarfa, A.: Image segmentation by a parallel, non-parametric histogram based clustering algorithm. Pattern Recogn. 23(9), 961–973 (1990)CrossRefGoogle Scholar
  12. [MG98]
    Moga, A.N., Gabbouj, M.: Parallel marker-based image segmentation with watershed transformation. J. Parallel Distrib. Comput. 51(1), 27–45 (1998)CrossRefzbMATHGoogle Scholar
  13. [MS]
    ANL Mathematics and Computer Science. The Message Passing Interface (MPI) standard. Accessed 29 October 2015
  14. [NVI]
    NVIDIA. CUDA Parallel Computing. Accessed 29 October 2015
  15. [Wes14]
    Westhoff, A.: Hybrid parallelization of a seeded region growing segmentation of brain images for a GPU cluster. In: ARCS 2014: 27th International Conference on Architecture of Computing Systems - Workshop Proceedings, p. 8. Berlin, Lübeck (Germany), 25 Feb 2014–28 Feb 2014, VDE Verlag, February 2014Google Scholar
  16. [WMS09]
    Wassenberg, J., Middelmann, W., Sanders, P.: An efficient parallel algorithm for graph-based image segmentation. In: Jiang, X., Petkov, N. (eds.) CAIP 2009. LNCS, vol. 5702, pp. 1003–1010. Springer, Heidelberg (2009). doi: 10.1007/978-3-642-03767-2_122 CrossRefGoogle Scholar
  17. [ZCUM11]
    Zhuge, Y., Cao, Y., Udupa, J.K., Miller, R.W.: Parallel fuzzy connected image segmentation on GPU. Med. Phys. 38(7), 4365–4371 (2011)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2016

Authors and Affiliations

  1. 1.Simulation Lab Neuroscience - Bernstein Facility for Simulation and Database Technology, Institute for Advanced Simulation, Jülich Aachen Research AllianceForschungszentrum JülichJülichGermany
  2. 2.Jülich Supercomputing Centre, Institute for Advanced SimulationForschungszentrum JülichJülichGermany
  3. 3.Institute of Neuroscience and Medicine (INM-1)Forschungszentrum JülichJülichGermany

Personalised recommendations