Abstract
Shape correlation of multi-object complexes in the human body can have significant implications in understanding the development of disease. While there exist geometric and statistical methods that aim for multi-object shape analysis, very little research can effectively extract shape correlation. It is especially difficult to extract the correlation when the involved objects have different variability in separate non-Euclidean spaces. To address these difficulties, this paper proposes geometric and statistical methods to extract the shape correlation from multi-object complexes. In particular, we focus on the shape correlation of the hippocampus and the caudate subject to the development of autism. The proposed methods are designed (1) to capture objects’ shape features (2) to capture shape correlation regardless of different variability between the two objects and (3) to provide interpretable shape correlation in multi-object complexes. In our experiments on synthetic data and autism data, the quantitative results and the qualitative visualization suggest that our methods are effective and robust.
Similar content being viewed by others
Notes
In this paper, “joint shape variation” is used interchangeably with “shape correlation”.
While Procrustes alignment is normally used to preprocess shape data, our simulation instead uses this alignment to produce a geometric transformation that is sensitive to NEUJIVE and that forms the group difference.
We create a random partition of the samples into 10 roughly equal-sized subsets of the ASD group and likewise with the non-ASD group. We set aside one of the subsets from each group for testing and use the remaining subsets for training and validation.
References
Abid, A., Zhang, M.J., Bagaria, V.K., Zou, J.: Exploring patterns enriched in a dataset with contrastive principal component analysis. Nat. Commun. 9(1), 1–7 (2018)
Björck, A., Golub, G.H.: Numerical methods for computing angles between linear subspaces. Math. Comput. 27(123), 579–594 (1973)
Bouhaddani, S.E., Uh, H.W., Jongbloed, G., Hayward, C., Klaric, L., Kielbasa, S.M., Houwing-Duistermaat, J.: Integrating omics datasets with the omicspls package. BMC Bioinf. 19(1), 1–9 (2018). https://doi.org/10.1186/s12859-018-2371-3
Cerrolaza, J., López Picazo, M., Humbert, L., Sato, Y., Rueckert, D., González Ballester, M.A., Linguraru, M.G.: Computational anatomy for multi-organ analysis in medical imaging: a review. Med. Image Anal. 56, 44–67 (2019)
Damon, J.: Smoothness and geometry of boundaries associated to skeletal structures I: sufficient conditions for smoothness. Ann. línst. Fourier 53, 1941–1985 (2003)
Damon, J., Marron, J.: Backwards principal component analysis and principal nested relations. J. Math. Imag. Vis. 50(1), 107–114 (2014)
Deutsch, F.: The angle between subspaces of a Hilbert space. In: Approximation theory, wavelets and applications, pp. 107–130. Springer (1995)
Dryden, I.L., Mardia, K.V.: Statistical shape analysis. Wiley, Chichester (1998)
Dryden, I.L., Mardia, K.V.: Statistical shape analysis: with applications in R, vol. 995. Wiley (2016)
Eilam-Stock, T., Wu, T., Spagna, A., Egan, L.J., Fan, J.: Neuroanatomical alterations in high-functioning adults with autism spectrum disorder. Front. Neurosci. 10, 237 (2016)
Eltzner, B., Jung, S., Huckemann, S.: Dimension reduction on polyspheres with application to skeletal representations, pp. 22–29. Springer (2015)
Feng, Q., Jiang, M., Hannig, J., Marron, J.S.: Angle-based joint and individual variation explained. J. Multivar. Anal. 166, 241–265 (2018)
Fletcher, P.T., Lu, C., Pizer, S.M., Joshi, S.: Principal geodesic analysis for the study of nonlinear statistics of shape. IEEE Trans. Med. Imag. 23(8), 995–1005 (2004)
Gorczowski, K., Styner, M., Jeong, J., Marron, J.S., Piven, J., Hazlett, H.C., Pizer, S.M., Gerig, G.: Statistical shape analysis of multi-object complexes. In: 2007 IEEE Conference on Computer Vision and Pattern Recognition, pp. 1–8 (2007)
Gorczowski, K., Styner, M., Jeong, J.Y., Marron, J., Piven, J., Hazlett, H.C., Pizer, S.M., Gerig, G.: Multi-object analysis of volume, pose, and shape using statistical discrimination. IEEE Trans. Pattern Anal. Mach. Intell. 32(4), 652–661 (2009)
Gori, P., Colliot, O., Marrakchi-Kacem, L., Worbe, Y., Poupon, C., Hartmann, A., Ayache, N., Durrleman, S.: A Bayesian framework for joint morphometry of surface and curve meshes in multi-object complexes. Med. Image Anal. 35, 458–474 (2017)
Hong, J.: Classification of neuroanatomical structures based on non-Euclidean geometric object properties. Ph.D. thesis. Computer Science dissertation, University of North Carolina at Chapel Hill (2019)
Hong, J., Vicory, J., Schulz, J., Styner, M., Marron, J., Pizer, S.: Non-Euclidean classification of medically imaged objects via s-reps. Med. Image Anal. 31, 37–45 (2016)
Hong, S., Fishbaugh, J., Gerig, G.: 4D continuous medial representation by geodesic shape regression. In: 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), pp. 1014–1017. IEEE (2018)
Hotelling, H.: Relations between two sets of variates. Biometrika 28(3–4), 321–377 (1936). https://doi.org/10.1093/biomet/28.3-4.321
Huckemann, S., Hotz, T., Munk, A.: Intrinsic shape analysis: Geodesic PCA for Riemannian manifolds modulo isometric Lie group actions. Statistica Sinica pp. 1–58 (2010)
Ipsen, I.C., Meyer, C.D.: The angle between complementary subspaces. Am. Math. Mon. 102(10), 904–911 (1995)
Jiang, M.: Statistical learning of integrative analysis. Ph.D. thesis, The University of North Carolina at Chapel Hill (2018)
Jung, S., Dryden, I.L., Marron, J.S.: Analysis of principal nested spheres. Biometrika (2012)
Katuwal, G.J., Cahill, N.D., Baum, S.A., Michael, A.M.: The predictive power of structural MRI in autism diagnosis. In: 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 4270–4273 (2015). https://doi.org/10.1109/EMBC.2015.7319338
van der Kloet, F.M., Sebastián-León, P., Conesa, A., Smilde, A.K., Westerhuis, J.A.: Separating common from distinctive variation. BMC Bioinf. 17(5), 271–286 (2016)
Knyazev, A.V., Argentati, M.E.: Majorization for changes in angles between subspaces, ritz values, and graph laplacian spectra. SIAM J. Matrix Anal. Appl. 29(1), 15–32 (2007)
Lindström, A., Pettersson, F., Almqvist, F., Berglund, A., Kihlberg, J., Linusson, A.: Hierarchical pls modeling for predicting the binding of a comprehensive set of structurally diverse protein- ligand complexes. J. Chem. Inf. Model. 46(3), 1154–1167 (2006)
Liu, Z.: Geometric and statistical models for multi-object shape analysis (chapter 2). Ph.D. thesis. Computer Science dissertation, University of North Carolina at Chapel Hill (2022)
Liu, Z., Damon, J., Marron, J.S., Pizer, S.: Geometric and statistical models for analysis of two-object complexes. Under review (2022)
Liu, Z., Hong, J., Vicory, J., Damon, J.N., Pizer, S.M.: Fitting unbranching skeletal structures to objects. Med. Image Anal. 70, 102020 (2021)
Lock, E.F., Hoadley, K.A., Marron, J.S., Nobel, A.B.: Joint and individual variation explained (JIVE) for integrated analysis of multiple data types. Ann. Appl. Stat. 7(1), 523 (2013)
Marron, J.S., Todd, M.J., Ahn, J.: Distance weighted discrimination. J. Am. Stat. Assoc. 102(480), 1267–1271 (2007)
Miolane, N., Caorsi, M., Lupo, U., Guerard, M., Guigui, N., Mathe, J., Cabanes, Y., Reise, W., Davies, T., Leitão, A., et al.: ICLR 2021 challenge for computational geometry & topology: design and results. arXiv preprint arXiv:2108.09810 (2021)
Murphy, C.M., Deeley, Q., Daly, E., Ecker, C., Obrien, F.: Anatomy and aging of the amygdala and hippocampus in autism spectrum disorder: an in vivo magnetic resonance imaging study of asperger syndrome. Autism Res. 5(1), 3–12 (2012)
Nicolson, R., DeVito, T.J., Vidal, C.N., Sui, Y., Hayashi, K.M., Drost, D.J., Williamson, P.C., Rajakumar, N., Toga, A.W., Thompson, P.M.: Detection and mapping of hippocampal abnormalities in autism. Psychiatr. Res. Neuroimaging 148(1), 11–21 (2006)
Pizer, S.M., Hong, J., Vicory, J., Liu, Z., Marron, J.S., et al.: Object shape representation via skeletal models (s-reps) and statistical analysis. Riemannian Geometric Statistics in Medical Image Analysis pp. 233–272 (2019)
Pizer, S.M., Jung, S., Goswami, D., Vicory, J., Zhao, X., Chaudhuri, R., Damon, J.N., Huckemann, S., Marron, J.: Nested sphere statistics of skeletal models. In: Innovations for shape analysis, pp. 93–115. Springer (2013)
Pizer, S.M., Marron, J., Damon, J., Vicory, J., Krishna, A., Liu, Z., Taheri, M.: Skeletons, object shape, statistics. Front. Comput. Sci. 4, 842637 (2022)
Qiu, A., Adler, M., Crocetti, D., Miller, M.I., Mostofsky, S.H.: Basal ganglia shapes predict social, communication, and motor dysfunctions in boys with autism spectrum disorder. J. Am. Acad. Child Adolesc. Psychiatr. 49(6), 539–551 (2010)
Richards, R., Greimel, E., Kliemann, D., Koerte, I.K., Schulte-Körne, G., Reuter, M., Wachinger, C.: Increased hippocampal shape asymmetry and volumetric ventricular asymmetry in autism spectrum disorder. NeuroImage Clin. 26, 102207 (2020)
Sagonas, C., Panagakis, Y., Leidinger, A., Zafeiriou, S.: Robust joint and individual variance explained. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 5267–5276 (2017)
Schulz, J., Pizer, S., Marron, J., Godtliebsen, F.: Nonlinear hypothesis testing of geometric object properties of shapes applied to hippocampi. J. Math. Imag. Vis. 54, 15–34 (2016)
Shen, C., Sun, M., Tang, M., Priebe, C.E.: Generalized canonical correlation analysis for classification. J. Multivar. Anal. 130, 310–322 (2014)
Styner, M., Oguz, I., Xu, S., Brechbühler, C., Pantazis, D., Levitt, J., Shenton, M., Gerig, G.: Statistical shape analysis of brain structures using SPHARM-PDM. Insight J. 1071, 242–250 (2006)
Torgerson, W.S.: Multidimensional scaling: I. theory and method. Psychometrika 17(4), 401–419 (1952)
Trygg, J., Wold, S.: O2-PLS, a two-block (X-Y) latent variable regression (LVR) method with an integral OSC filter. J. Chemom. 17, 53–64 (2003). https://doi.org/10.1002/cem.775
Tu, L., Styner, M., Vicory, J., et al.: Skeletal shape correspondence through entropy. IEEE Transactions on Medical Imaging (2018)
Van Deun, K., Van Mechelen, I., Thorrez, L., Schouteden, M., De Moor, B., Van Der Werf, M.J., De Lathauwer, L., Smilde, A.K., Kiers, H.A.: Disco-sca and properly applied gsvd as swinging methods to find common and distinctive processes. PLoS One 7(5), e37840 (2012)
Wang, J., Vachet, C., Rumple, A., Gouttard, S., Ouziel, C., Perrot, E., Du, G., Huang, X., Gerig, G., Styner, M.A.: Multi-atlas segmentation of subcortical brain structures via the AutoSeg software pipeline. Front. Neuroinf. 8, 7 (2014)
Wei, S., Lee, C., Wichers, L., Marron, J.: Direction-projection-permutation for high-dimensional hypothesis tests. J. Comput. Graph. Stat. 25(2), 549–569 (2016)
Westerhuis, J.A., Kourti, T., MacGregor, J.F.: Analysis of multiblock and hierarchical PCA and PLS models. J. Chemom. 12(5), 301–321 (1998)
Wold, H.: Partial least squares (2004). https://doi.org/10.1002/0471667196.ess1914
Wold, S., Geladi, P., Esbensen, K., Öhman, J.: Multi-way principal components and PLS analysis. J. Chemom. 1, 41–56 (2005). https://doi.org/10.1002/cem.1180010107
Wold, S., Kettaneh, N., Tjessem, K.: Hierarchical multiblock PLS and PC models for easier model interpretation and as an alternative to variable selection. J. Chemom. 10, 463–482 (1996)
Yushkevich, P., Fletcher, P.T., Joshi, S., Thall, A., Pizer, S.M.: Continuous medial representations for geometric object modeling in 2d and 3d. Image Vis. Comput. 21(1), 17–27 (2003)
Acknowledgements
This research is funded by NIH grants R01HD055741, R01HD059854 and R01HD088125. The ASD data was kindly provided by the IBIS network. We thank G. Gerig (NYU), SunHyung Kim (UNC), D. Louis Collins (McGill University), Vladimir Fonov (McGill University) and Heather Hazlett (UNC) for their efforts in processing the data. We are also grateful for useful discussion relating to this project with J. Fishbaugh, Xi Yang, Iain Carmichael and Eric Lock. We specially thank the reviewers for the insightful comments and suggestions.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
A Non-Euclidean Joint and Individual Variation Explained
A Non-Euclidean Joint and Individual Variation Explained
In this section, we show more comprehensive experimental analysis of NEUJIVE.
First, it is of interest to show all NEUJIVE components of the toy example discussed in Sect. 7.2, which has homogeneous circular data in each block. In Fig. 11 the first row shows both the joint and the individual structures from NEUJIVE. The individual structures are designed as random variables from multivariate Gaussian (see Eq. (11)). Moreover, the top left cell in Fig. 11 shows that the individual structures (shown as blue dots) from NEUJIVE are distributed around the PNS mean as expected. As a comparison, AJIVE fails to find significant individual components in either block, as shown in the top right cell. Instead, the blue dots are the residual components resulting from AJIVE.
Second, we show the NEUJIVE components when the two blocks are of notably different variability in the bottom row of Fig. 11. Different from Sect. 7.2, the first block is generated via
where \(p(\theta )\) is a straight line in the tangent space at the north pole. A point on this line has coordinates \( (\theta , 0.3\theta )\). The second block is still generated via Eq. (11). Moreover, to have Different Noise Levels (DNL) across the two blocks, we set different standard deviations of \(\epsilon _k\) across the two blocks. The results show that NEUJIVE can still give effective estimation of the joint components regardless of the different variability between the two blocks.
Third, we investigate the robustness of the joint structures estimated by NEUJIVE as we increase noise. We simulate two blocks of data via Eq. (11), increasing the standard deviation of \(\epsilon _k\) from 0.01 to 0.1. The simulated data under 3 different noise levels can be seen in Fig. 12 left, i.e., \(\sigma (\epsilon _k)=\{0.01,\, 0.05,\, 0.1\}\). For each noise level, we compute the joint structure from NEUJIVE. We measure the Root of Mean Square Error (RMSE) between the NEUJIVE joint structures and the actual joint structures. The curves of RMSE versus noise levels of the two blocks are shown in Fig. 12 right. This figure shows that the RMSE is almost linearly increasing along with the increasing noise level.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Liu, Z., Schulz, J., Taheri, M. et al. Analysis of Joint Shape Variation from Multi-Object Complexes. J Math Imaging Vis 65, 542–562 (2023). https://doi.org/10.1007/s10851-022-01136-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10851-022-01136-5