Advertisement

Diffeomorphic Metric Learning and Template Optimization for Registration-Based Predictive Models

  • Ayagoz Mussabayeva
  • Maxim Pisov
  • Anvar Kurmukov
  • Alexey Kroshnin
  • Yulia Denisova
  • Li Shen
  • Shan Cong
  • Lei Wang
  • Boris GutmanEmail author
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11846)

Abstract

We present a method for metric optimization and template construction in the Large Deformation Diffeomorphic Metric Mapping (LDDMM) framework. The construction treats the Riemannian metric on the space of diffeomorphisms as a data-embedding kernel in the context of predictive modeling, here Kernel Logistic Regression (KLR). The task is then to optimize kernel parameters, including the LDDMM metric parameters as well as the registration template, resulting in a parameterized argminimum optimization. In practice, this leads to a group-wise registration problem with the goal of improving predictive performance, for example by focusing the metric and template on discriminating patient and control populations. We validate our algorithm using two discriminative problems on a synthetic data set as well as 3D subcortical shapes from the SchizConnect cohort. Though secondary to the template and kernel optimization, accuracy of schizophrenia classification is improved by LDDMM-KLR compared to linear and RBF-KLR.

Keywords

Image registration Machine learning Subcortical shape Metric learning LDDMM 

Notes

Acknowledgements

The research was conducted in the IITP RAS and solely supported by the Russian Science Foundation grant (project 17-11-01390).

References

  1. 1.
    Beg, M.F., Miller, M.I., Trouvé, A., Younes, L.: Computing large deformation metric mappings via geodesic flows of diffeomorphisms. Int. J. Comput. Vision 61(2), 139–157 (2005)CrossRefGoogle Scholar
  2. 2.
    Mussabayeva, A., et al.: Image registration and predictive modeling: learning the metric on the space of diffeomorphisms. In: Reuter, M., Wachinger, C., Lombaert, H., Paniagua, B., Lüthi, M., Egger, B. (eds.) ShapeMI 2018. LNCS, vol. 11167, pp. 160–168. Springer, Cham (2018).  https://doi.org/10.1007/978-3-030-04747-4_15CrossRefGoogle Scholar
  3. 3.
    Ceritoglu, C., et al.: Computational analysis of LDDMM for brain mapping. Front. Neurosci. 7, 151 (2013)CrossRefGoogle Scholar
  4. 4.
    Cho, Y., Saul, L.K.: Kernel methods for deep learning. In: Advances in Neural Information Processing Systems, pp. 342–350 (2009)Google Scholar
  5. 5.
    Ayat, N.-E., Cheriet, M., Suen, C.Y.: Optimization of the SVM kernels using an empirical error minimization scheme. In: Lee, S.-W., Verri, A. (eds.) SVM 2002. LNCS, vol. 2388, pp. 354–369. Springer, Heidelberg (2002).  https://doi.org/10.1007/3-540-45665-1_28CrossRefGoogle Scholar
  6. 6.
    Huang, J., Yuen, P.C., Chen, W.S., Lai, J.H.: Kernel subspace LDA with optimized kernel parameters on face recognition. In: Proceedings of Sixth IEEE International Conference on Automatic Face and Gesture Recognition, pp. 327–332. IEEE (2004)Google Scholar
  7. 7.
    Niethammer, M., Kwitt, R., Vialard, F.X.: Metric learning for image registration. In: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2019Google Scholar
  8. 8.
    Zhang, M., Singh, N., Fletcher, P.T.: Bayesian estimation of regularization and atlas building in diffeomorphic image registration. In: Gee, J.C., Joshi, S., Pohl, K.M., Wells, W.M., Zöllei, L. (eds.) IPMI 2013. LNCS, vol. 7917, pp. 37–48. Springer, Heidelberg (2013).  https://doi.org/10.1007/978-3-642-38868-2_4CrossRefGoogle Scholar
  9. 9.
    Gould, S., Fernando, B., Cherian, A., Anderson, P., Cruz, R.S., Guo, E.: On differentiating parameterized argmin and argmax problems with application to bi-level optimization. arXiv preprint arXiv:1607.05447 (2016)
  10. 10.
    Wang, L., et al.: SchizConnect: mediating neuroimaging databases on schizophrenia and related disorders for large-scale integration. NeuroImage 124, 1155–1167 (2016). Sharing the wealth: Brain Imaging Repositories in 2015CrossRefGoogle Scholar
  11. 11.
    Fischl, B.: FreeSurfer. Neuroimage 62(2), 774–781 (2012)CrossRefGoogle Scholar
  12. 12.
    Roshchupkin, G.V., et al.: Heritability of the shape of subcortical brain structures in the general population. Nat. Commun. 7, 13738 (2016)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Ayagoz Mussabayeva
    • 1
  • Maxim Pisov
    • 1
    • 3
  • Anvar Kurmukov
    • 1
    • 4
  • Alexey Kroshnin
    • 1
    • 4
  • Yulia Denisova
    • 1
  • Li Shen
    • 5
  • Shan Cong
    • 6
  • Lei Wang
    • 7
  • Boris Gutman
    • 1
    • 2
    Email author
  1. 1.The Institute for Information Transmission ProblemsMoscowRussia
  2. 2.Department of Biomedical EngineeringIllinois Institute of TechnologyChicagoUSA
  3. 3.Moscow Institute of Physics and Technology State UniversityMoscowRussia
  4. 4.Higher School of EconomicsMoscowRussia
  5. 5.Department of Biostatistics, Epidemiology and InformaticsUniversity of PennsylvaniaPhiladelphiaUSA
  6. 6.Indiana UniversityIndianapolisUSA
  7. 7.Department of Psychiatry and Behavioral SciencesNorthwestern University Feinberg School of MedicineChicagoUSA

Personalised recommendations