Analysis of Dynamic Brain Connectivity Through Geodesic Clustering
- 1k Downloads
Analysis of dynamic functional connectivity allows for studying the time variant behavior of brain connectivity during specific tasks or at rest. There is, however, a debate around the significance of studies analyzing the dynamic connectivity, as it is usually estimated using short subsequences of the entire time-series. Therefore, a question that naturally arises is whether the dynamic connectivity information is robust enough to compare connectivity matrices. In this paper we investigate the importance of the choice of metric on the space of graphs to answer this question, using a dataset of twins under the assumption that twins connectivity is more similar than in any other pair of unrelated subjects. Specifically, the problem was formulated as a classification task between twin and non-twin pairs. The approach described in the paper relies on geodesic clustering of dynamic connectivity matrices to find a subset of brain states, which were then used to encode the pairwise connectivity similarities between subjects. Experiments were performed to compare the use of Euclidean distance in a vectorial space and a geodesic distance in the Riemannian space of symmetric positive definite matrices. We showed that the geodesic distance provided a better classification of twins subjects, suggesting this use of this distance can robustly compare dynamic connectivity matrices.
KeywordsDynamic functional connectivity Geodesic clustering Connectomes Task-based fMRI SVM Symmetric positive definite matrices
The authors acknowledge Cigdem Beyan and Muhammad Shahid for the helpful discussions.
- 13.Yamin, A., et al.: Comparison of brain connectomes using geodesic distance on manifold: a twin’s study. In: International Symposium on Biomedical Imaging 2019, Venice, 8–11 April 2019Google Scholar
- 15.Yamin, A., et al.: Investigating the impact of genetic background on brain dynamic functional connectivity through machine learning: a twins study. In: IEEE-EMBS International Conference on Biomedical and Health Informatics, Chicago, IL, USA, 19–22 May 2019Google Scholar
- 17.Tejwani, R., Liska, A., You, H.: Autism Classification Using Brain Functional Connectivity Dynamics and Machine Learning (2019). https://arxiv.org/pdf/1712.08041.pdf
- 18.Poffenberger, A.T.: Reaction Time to Retinal Stimulation, with Special Reference to the Time Lost in Conduction Through Nerve Centers. The Science Press, New York (1912)Google Scholar
- 21.Varoquaux, G., Gramfort, A., Poline, J.B., Thirion, B., Zemel, R, Shawe-Taylor, J.: Brain covariance selection: better individual functional connectivity models using population prior. In: Advances in Neural Information Processing Systems, Vancouver, Canada (2010)Google Scholar
- 24.Rashid, B., Damaraju, E., Pearlson, G.D., Calhoun, V.D.: Dynamic connectivity states estimated from resting fMRI identify differences among Schizophrenia, bipolar disorder, and healthy control subjects. Front. Hum. Neurosci. 8, 897 (2014). https://doi.org/10.3389/fnhum.2014.00897CrossRefGoogle Scholar
- 25.Dodero, L., Minh, H.Q., Biagio, M.S., Murino, V., Sona, D.: Kernel-based classification for brain connectivity graphs on the Riemannian manifold of positive definite matrices. In: ISBI 2015, 16–19 April 2015Google Scholar
- 26.Dodero, L., Sambataro, F., Murino, V., Sona, D.: Kernel-based analysis of functional brain connectivity on Grassmann manifold. In: Navab, N., Hornegger, J., Wells, W., Frangi, A. (eds.) MICCAI 2015. LNCS, vol. 9351, pp. 604–611. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-24574-4_72CrossRefGoogle Scholar
- 29.Yang, X., Song, Q., Cao, A.: Weighted support vector machine for data classification. In: Proceedings of 2005 IEEE International Joint Conference on Neural Networks, Montreal, Quebec, vol. 2, pp. 859–864 (2005). https://doi.org/10.1109/IJCNN.2005.1555965