Higher-Order Block Term Decomposition for Spatially Folded fMRI Data
- 1.6k Downloads
The growing use of neuroimaging technologies generates a massive amount of biomedical data that exhibit high dimensionality. Tensor-based analysis of brain imaging data has been proved quite effective in exploiting their multiway nature. The advantages of tensorial methods over matrix-based approaches have also been demonstrated in the context of functional magnetic resonance imaging (fMRI) data analysis. However, such methods can become ineffective in demanding scenarios, involving, e.g., strong noise and/or significant overlapping of activated regions. This paper aims at investigating the possible gains that can be obtained from a better exploitation of the spatial dimension, through a higher (than 3)-order tensor modeling of the fMRI signals. In this context, a higher-order Block Term Decomposition (BTD) is applied, for the first time in fMRI analysis. Its effectiveness in handling strong instances of noise is demonstrated via extensive simulation results.
KeywordsfMRI Tensors BTD CPD TPICA
The authors would like to thank Profs. A. Stegeman and N. Helwig for providing the datasets used in  and , respectively, and Prof. S. Van Huffel for her critical comments on earlier version of this paper. Constructive comments from the reviewers are also gratefully acknowledged. This research has been funded by the European Union’s Seventh Framework Programme (H2020-MSCA-ITN-2014) under grant agreement No. 642685 MacSeNet.
- 2.Theodoridis, S.: Machine Learning: A Bayesian and Optimization Perspective. Academic Press, Boston (2015)Google Scholar
- 5.Andersen, K.W., Mørup, M., Siebner, H., Madsen, K.H., Hansen, L.K.: Identifying modular relations in complex brain networks. In: IEEE International Workshop on Machine Learning for Signal Processing (MLSP) (2012)Google Scholar
- 11.Harshman, R.A.: Foundations of the PARAFAC procedure: models and conditions for an “explanatory” multi-modal factor analysis. UCLA Work. Papers in Phonetics, pp. 1–84(1970)Google Scholar
- 13.Stegeman, A.: Comparing Independent Component Analysis and the PARAFAC model for artificial multi-subject fMRI data. Unpublished Technical report, University of Groningen (2007)Google Scholar
- 21.Norgaard, L.: Classification and prediction of quality and process parameters of thick juice and beet sugar by fluorescence spectroscopy and chemometrics. Zuckerindustrie 120(11), 970–981 (1995)Google Scholar
- 22.Phillips, N.C.: Gasthuisberg University Hospital raises fMRI to new level with Intera 3.0 T. http://netforum.healthcare.philips.com/
- 23.Hunyadi, B., Camps, D., Sorber, L., Van Paesschen, W., De Vos, M., Van Huffel, S., De Lathauwer, L.: Block term decomposition for modelling epileptic seizures. EURASIP J. Adv. Signal Process. (2014). doi: 10.1186/1687-6180-2014-139
- 24.Phan, A.H., Cichocki, A., Zdunek, R., Lehky, S.: From basis components to complex structural patterns. In: IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Vancouver (2013)Google Scholar
- 25.Brie, D., Miron, S., Caland, F., Mustin, C.: An uniqueness condition for the 4-way CANDECOMP/PARAFAC model with collinear loadings in three modes. In: IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Prague (2011)Google Scholar
- 27.Vervliet, N., Debals, O., Sorber, L., Van Barel, M., De Lathauwer, L.: Tensorlab user guide (2016). http://www.tensorlab.net