Abstract
We consider how to test for group differences of shapes given longitudinal data. In particular, we are interested in differences of longitudinal models of each group’s subjects. We introduce a generalization of principal geodesic analysis to the tangent bundle of a shape space. This allows the estimation of the variance and principal directions of the distribution of trajectories that summarize shape variations within the longitudinal data. Each trajectory is parameterized as a point in the tangent bundle. To study statistical differences in two distributions of trajectories, we generalize the Bhattacharyya distance in Euclidean space to the tangent bundle. This not only allows to take second-order statistics into account, but also serves as our test-statistic during permutation testing. Our method is validated on both synthetic and real data, and the experimental results indicate improved statistical power in identifying group differences. In fact, our study sheds new light on group differences in longitudinal corpus callosum shapes of subjects with dementia versus normal controls.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
- 2.
The threshold \(\epsilon \) varies with the application. In our experiments, we set it to 1e-6. Usually, the eigenvalues larger than \(\epsilon \) cover almost \(99\,\%\) of the variances.
- 3.
We use two geodesics to connect three given shapes and uniformly sample points on these two geodesics. Then, by connecting opposing points, we obtain new geodesics which are located within the triangle region to sample a population of shapes.
- 4.
The average of two generalized squared-Mahalanobis distances is related to the first term of the generalized Bhattacharyya distance in Eq. (4).
References
Bhattacharyya, A.: On a measure of divergence between two multinomial populations. Sankhyā Indian J. Stat. 7(4), 401–406 (1946)
Bickel, P.J., Levina, E.: Covariance regularization by thresholding. Ann. Stat. 36(6), 2577–2604 (2008)
Durrleman, S., Pennec, X., Trouvé, A., Braga, J., Gerig, G., Ayache, N.: Toward a comprehensive framework for the spatiotemporal statistical analysis of longitudinal shape data. IJCV 103(1), 22–59 (2013)
Fletcher, P.T.: Geodesic regression and the theory of least squares on Riemannian manifolds. IJCV 105(2), 171–185 (2013)
Fletcher, P.T., Lu, C., Pizer, S.M., Joshi, S.: Principal geodesic analysis for the study of nonlinear statistics of shape. IEEE TMI 23(8), 995–1005 (2004)
Gilmore, J.H., Shi, F., Woolson, S.L., Knickmeyer, R.C., Short, S.J., Lin, W., Zhu, H., Hamer, R.M., Styner, M., Shen, D.: Longitudinal development of cortical and subcortical gray matter from birth to 2 years. Cereb. Cortex 22(11), 2478–2485 (2012)
Hong, Y., Joshi, S., Sanchez, M., Styner, M., Niethammer, M.: Metamorphic geodesic regression. In: Ayache, N., Delingette, H., Golland, P., Mori, K. (eds.) MICCAI 2012, Part III. LNCS, vol. 7512, pp. 197–205. Springer, Heidelberg (2012)
Ledoit, O., Wolf, M.: A well-conditioned estimator for large-dimensional covariance matrices. J. Multivar. Anal. 88(2), 365–411 (2004)
Lee, J.: Introduction to Smooth Manifolds. Springer, New York (2012)
Mahalanobis, P.C.: On the generalized distance in statistics. Proc. Natl. Inst. Sci. (Calcutta) 2, 49–55 (1936)
Marcus, D.S., Fotenos, A.F., Csernansky, J.G., Morris, J.C., Buckner, R.L.: Open access series of imaging studies: longitudinal mri data in nondemented and demented older adults. J. Cogn. Neurosci. 22(12), 2677–2684 (2010)
Muralidharan, P., Fletcher, P.T.: Sasaki metrics for analysis of longitudinal data on manifolds. In: CVPR, pp. 1027–1034 (2012)
Niethammer, M., Huang, Y., Vialard, F.-X.: Geodesic regression for image time-series. In: Fichtinger, G., Martel, A., Peters, T. (eds.) MICCAI 2011, Part II. LNCS, vol. 6892, pp. 655–662. Springer, Heidelberg (2011)
Oliver, D.S.: Calculation of the inverse of the covariance. Math. Geol. 30(7), 911–933 (1998)
Sasaki, S.: On the differential geometry of tangent bundles of riemannian manifolds. TMJ 10(3), 338–354 (1958)
Singh, N., Hinkle, J., Joshi, S., Fletcher, P.T.: A hierarchical geodesic model for diffeomorphic longitudinal shape analysis. In: Gee, J.C., Joshi, S., Pohl, K.M., Wells, W.M., Zöllei, L. (eds.) IPMI 2013. LNCS, vol. 7917, pp. 560–571. Springer, Heidelberg (2013)
Su, J., Kurtek, S., Klassen, E., Srivastava, A.: Statistical analysis of trajectories on riemannian manifolds: bird migration, hurricane tracking and video surveillance. Ann. Appl. Stat. 8(1), 530–552 (2014)
Su, J., Srivastrava, A., de Souza, F., Sarkar, S.: Rate-invariant analysis of trajectories on riemannian manifolds with application in visual speech recognition. In: CVPR, pp. 620–627 (2014)
Acknowledgements
This work was supported by NSF EECS-1148870 and NSF EECS-0925875.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendices
Appendix
A Properties of the Generalized Bhattacharyya Distance
Non-negativity. In the first term of Eq. (4), \(D_M^{\mathcal {TM}}\) is the generalized squared-Mahalanobis distance which is non-negative; consequently, the first term in Eq. (4) is non-negative. Furthermore, the determinant of a covariance matrix in the second term is also non-negative, since it is the product of all non-negative eigenvalues. Besides, it is easy to demonstrate that \((|\varSigma _1|+|\varSigma _2|) / (2\sqrt{|\varSigma _1||\varSigma _2|}) \ge 1\), indicating the second term is non-negative. Hence, \(D_B^{\mathcal {TM}}(D_1, D_2) \ge 0\).
Identity of Indiscernibles. If \(D_1 = D_2\), i.e., \(\mu _1 = \mu _2\) and \(\varSigma _1 = \varSigma _2\), we see that (1) \({{\mathrm{Log}}}_{\mu _1} \mu _2\) and \({{\mathrm{Log}}}_{\mu _2} \mu _1\) are zero tangent vectors, and (2) \(|\varSigma _1| = |\varSigma _2|\). Hence, \(D_M^{\mathcal {TM}}(\mu _1, D_2) = D_M^{\mathcal {TM}}(\mu _2, D_1) = 0\), i.e., the first term of Eq. (4) is 0; also, the second term is 0. Now, if \(D_1 = D_2\) then \(D_B^{\mathcal {TM}}(D_1, D_2) = 0\). On the other hand, assuming \(D_B^{\mathcal {TM}}(D_1, D_2) = 0\), we can only obtain \(\mu _1 = \mu _2\) and \(|\varSigma _1| = |\varSigma _2|\), because of the non-negativity properties of the two terms in Eq. (4). But, we cannot draw the conclusion that the two covariance matrices are equal. Therefore, if \(D_1=D_2\) then \(D_B^{\mathcal {TM}}(D_1, D_2) = 0\), but it is possible that \(D_B^{\mathcal {TM}}(D_1, D_2) =0\) for some \(D_1 \not = D_2\), if \(\mu _1 = \mu _2\) and \(|\varSigma _1| = |\varSigma _2|\).
Symmetry. Because both terms of Eq. (4) are symmetric, the sum of them is also symmetric, i.e., \(D_B^{\mathcal {TM}}(D_1, D_2) = D_B^{\mathcal {TM}}(D_2, D_1)\).
Triangle Inequality. Since, Eq. (1) in \(\mathbb {R}^n\) does not satisfy the triangle inequality, our generalized variant will not satisfy it either.
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
Hong, Y., Singh, N., Kwitt, R., Niethammer, M. (2015). Group Testing for Longitudinal Data. In: Ourselin, S., Alexander, D., Westin, CF., Cardoso, M. (eds) Information Processing in Medical Imaging. IPMI 2015. Lecture Notes in Computer Science(), vol 9123. Springer, Cham. https://doi.org/10.1007/978-3-319-19992-4_11
Download citation
DOI: https://doi.org/10.1007/978-3-319-19992-4_11
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-19991-7
Online ISBN: 978-3-319-19992-4
eBook Packages: Computer ScienceComputer Science (R0)