Advertisement

Pattern Visualization and Recognition Using Tensor Factorization for Early Differential Diagnosis of Parkinsonism

  • Rui Li
  • Ping Wu
  • Igor Yakushev
  • Jian Wang
  • Sibylle I. Ziegler
  • Stefan Förster
  • Sung-Cheng Huang
  • Markus Schwaiger
  • Nassir Navab
  • Chuantao Zuo
  • Kuangyu ShiEmail author
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10435)

Abstract

Idiopathic Parkinsons disease (PD) and atypical parkinsonian syndromes may have similar symptoms at the early disease stage. Pattern recognition on metabolic imaging has been confirmed of distinct value in the early differential diagnosis of Parkinsonism. However, the principal component analysis (PCA) based method ends up with a unique probability score of each disease pattern. This restricts the exploration of heterogeneous characteristic features for differentiation. There is no visualization of the underlying mechanism to assist the radiologist/neurologist either. We propose a tensor factorization based method to extract the characteristic patterns of the diseases. By decomposing the 3D data, we can capture the intrinsic characteristic pattern in the data. In particular, the disease-related patterns can be visualized individually for the inspection by physicians. The test on PET images of 206 early parkinsonian patients has confirmed differential patterns on the visualized feature images using the proposed method. Computer-aided diagnosis based on multi-class support vector machine (SVM) shown improved diagnostic accuracy of Parkinsonism using the tensor-factorized feature images compared to the state-of-the-art PCA-based scores [Tang et al. Lancet Neurol. 2010].

Notes

Acknowledgement

The methodological development is based on the funding from German Research Foundation (DFG) Collaborative Research Centre 824 (SFB824). The international cooperation was supported by the Sino-German Insititue for Brain Molecular Imaging and Clinical Translation.

References

  1. 1.
    Brett W.B., Tamara, G.K., et al.: Matlab tensor toolbox version 2.6., February 2015Google Scholar
  2. 2.
    Bi, L., Kim, J., Feng, D., Fulham, M.: Multi-stage thresholded region classification for whole-body PET-CT lymphoma studies. In: Golland, P., Hata, N., Barillot, C., Hornegger, J., Howe, R. (eds.) MICCAI 2014. LNCS, vol. 8673, pp. 569–576. Springer, Cham (2014). doi: 10.1007/978-3-319-10404-1_71 CrossRefGoogle Scholar
  3. 3.
    Chen, K., Langbaum, J.B., Fleisher, A.S., Ayutyanont, N., Reschke, C., Lee, W., Liu, X., Bandy, D., Alexander, G.E., Thompson, P.M., Foster, N.L., Harvey, D.J., de Leon, M.J., Koeppe, R.A., Jagust, W.J., Weiner, M.W., Reiman, E.M.: Twelve-month metabolic declines in probable alzheimer’s disease and amnestic mild cognitive impairment assessed using an empirically pre-defined statistical region-of-interest: findings from the alzheimer’s disease neuroimaging initiative. Neuroimage 51(2), 654–664 (2010)CrossRefGoogle Scholar
  4. 4.
    Huang, H., Ding, C., Luo, D.J.: Tensor reduction error analysis-applications to video compression and classification. In: IEEE Conference on Computer Vision and Pattern Recognition (2008)Google Scholar
  5. 5.
    Eidelberg, D.: Metabolic brain networks in neurodegenerative disorders: a functional imaging approach. Trends Neurosci. 32(10), 548–557 (2009)CrossRefGoogle Scholar
  6. 6.
    Fahn, S., Oakes, D., Shoulson, I., Kieburtz, K., Rudolph, A., Lang, A., Olanow, C.W., Tanner, C., Marek, K.: Levodopa and the progression of Parkinson’s disease. New Engl. J. Med. 351(24), 2498–2508 (2004)CrossRefGoogle Scholar
  7. 7.
    Gao, F., Liu, H., Shi, P.: Patient-adaptive lesion metabolism analysis by dynamic PET images. In: Ayache, N., Delingette, H., Golland, P., Mori, K. (eds.) MICCAI 2012. LNCS, vol. 7512, pp. 558–565. Springer, Heidelberg (2012). doi: 10.1007/978-3-642-33454-2_69 CrossRefGoogle Scholar
  8. 8.
    Hellwig, S., Frings, L., Amtage, F., Buchert, R., Spehl, T.S., Rijntjes, M., Tuscher, O., Weiller, C., Weber, W.A., Vach, W., Meyer, P.T.: 18f-FDG PET is an early predictor of overall survival in suspected atypical Parkinsonism. J. Nucl. Med. 56(10), 1541–1546 (2015)CrossRefGoogle Scholar
  9. 9.
    Hitchcock, F.L.: The expression of a tensor or a polyadic as a sum of products. J. Math. Phys. 6(1), 164–189 (1927)CrossRefzbMATHGoogle Scholar
  10. 10.
    Hughes, A.J., Ben-Shlomo, Y., Daniel, S.E., Lees, A.J.: What features improve the accuracy of clinical diagnosis in parkinson’s disease: a clinicopathologic study. Neurology 57(10 Suppl 3), S34–S38 (2001)Google Scholar
  11. 11.
    Hughes, A.J., Daniel, S.E., Ben-Shlomo, Y., Lees, A.J.: The accuracy of diagnosis of parkinsonian syndromes in a specialist movement disorder service. Brain 125(Pt 4), 861–870 (2002)CrossRefGoogle Scholar
  12. 12.
    Jiao, J., Searle, G.E., Tziortzi, A.C., Salinas, C.A., Gunn, R.N., Schnabel, J.A.: Spatio-temporal pharmacokinetic model based registration of 4D PET neuroimaging data. NeuroImage 84, 225–235 (2014)CrossRefGoogle Scholar
  13. 13.
    Amatriain, X., Baltrunas, L., Karatzoglou, A., Oliver, N.: Multiverse recommendation: n-dimensional tensor factorization for context-aware collaborative filtering. In: Proceedings of the Fourth ACM conference on Recommender Systems, pp. 79–86 (2010)Google Scholar
  14. 14.
    Mitchell, T.M., Papalexakis, E.E., et al.: Turbo-SMT: accelerating coupled sparse matrix-tensor factorizations by 200x. In: SIAM International Conference on Data Mining. SIAM (2014)Google Scholar
  15. 15.
    Phan, A.-H., Cichocki, A.: Tensor decompositions for feature extraction and classification of high dimensional datasets. Nonlinear Theor. Appl. IEICE 1(1), 37–68 (2010)CrossRefGoogle Scholar
  16. 16.
    Tang, C.C., Poston, K.L., Eckert, T., Feigin, A., Frucht, S., Gudesblatt, M., Dhawan, V., Lesser, M., Vonsattel, J.P., Fahn, S., Eidelberg, D.: Differential diagnosis of Parkinsonism: a metabolic imaging study using pattern analysis. Lancet Neurol. 9(2), 149–158 (2010)CrossRefGoogle Scholar
  17. 17.
    Xu, Z., Bagci, U., Seidel, J., Thomasson, D., Solomon, J., Mollura, D.J.: Segmentation based denoising of PET images: an iterative approach via regional means and affinity propagation. In: Golland, P., Hata, N., Barillot, C., Hornegger, J., Howe, R. (eds.) MICCAI 2014. LNCS, vol. 8673, pp. 698–705. Springer, Cham (2014). doi: 10.1007/978-3-319-10404-1_87 CrossRefGoogle Scholar
  18. 18.
    Zhang, D., Shen, D.: Multi-modal multi-task learning for joint prediction of multiple regression and classification variables in alzheimer’s disease. Neuroimage 59(2), 895–907 (2012)MathSciNetCrossRefGoogle Scholar
  19. 19.
    Zhou, G.X., Zhao, Q.B., Cichocki, A., Zhang, Y., Wang, X.Y.: Fast nonnegative tensor factorization based on accelerated proximal gradient and low-rank approximation. Neurocomputing 198, 148–154 (2016)CrossRefGoogle Scholar
  20. 20.
    Zhou, L., Salvado, O., Dore, V., Bourgeat, P., Raniga, P., Villemagne, V.L., Rowe, C.C., Fripp, J.: MR-less surface-based amyloid estimation by subject-specific atlas selection and Bayesian fusion. In: Ayache, N., Delingette, H., Golland, P., Mori, K. (eds.) MICCAI 2012. LNCS, vol. 7511, pp. 220–227. Springer, Heidelberg (2012). doi: 10.1007/978-3-642-33418-4_28 CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • Rui Li
    • 1
    • 6
  • Ping Wu
    • 2
  • Igor Yakushev
    • 3
  • Jian Wang
    • 4
  • Sibylle I. Ziegler
    • 3
  • Stefan Förster
    • 3
  • Sung-Cheng Huang
    • 5
  • Markus Schwaiger
    • 3
  • Nassir Navab
    • 1
  • Chuantao Zuo
    • 2
  • Kuangyu Shi
    • 3
    Email author
  1. 1.Department of Computer ScienceTechnische Universität MünchenMunichGermany
  2. 2.Huashan Hospital, PET CenterFudan UniversityShanghaiChina
  3. 3.Department of Nuclear MedicineTechnische Universität MünchenMunichGermany
  4. 4.Department of Neurology, Huashan HospitalFudan UniversityShanghaiChina
  5. 5.Department of Molecular and Medical PharmacologyUniversity of CaliforniaLos AngelesUSA
  6. 6.Alibaba CloudHangzhouChina

Personalised recommendations