Abstract
Connectomics has emerged as a powerful tool in neuroimaging and has spurred recent advancements in statistical and machine learning methods for connectivity data. Despite connectomes inhabiting a matrix manifold, most analytical frameworks ignore the underlying data geometry. This is largely because simple operations, such as mean estimation, do not have easily computable closed-form solutions. We propose a geometrically aware neural framework for connectomes, i.e., the mSPD-NN, designed to estimate the geodesic mean of a collections of symmetric positive definite (SPD) matrices. The mSPD-NN is comprised of bilinear fully connected layers with tied weights and utilizes a novel loss function to optimize the matrix-normal equation arising from Fréchet mean estimation. Via experiments on synthetic data, we demonstrate the efficacy of our mSPD-NN against common alternatives for SPD mean estimation, providing competitive performance in terms of scalability and robustness to noise. We illustrate the real-world flexibility of the mSPD-NN in multiple experiments on rs-fMRI data and demonstrate that it uncovers stable biomarkers associated with subtle network differences among patients with ADHD-ASD comorbidities and healthy controls.
Keywords
- Functional Connectomics
- SPD Manifolds
- Fréchet Mean Estimation
- Geometry-Aware Neural Networks
This is a preview of subscription content, access via your institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Banerjee, M., Chakraborty, R., Ofori, E., Vaillancourt, D., Vemuri, B.C.: Nonlinear regression on Riemannian manifolds and its applications to neuro-image analysis. In: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (eds.) MICCAI 2015. LNCS, vol. 9349, pp. 719–727. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-24553-9_88
Bessadok, A., Mahjoub, M.A., Rekik, I.: Graph neural networks in network neuroscience. IEEE Trans. Pattern Anal. Mach. Intell. 45, 5833–5848 (2022)
Congedo, M., Afsari, B., Barachant, A., Moakher, M.: Approximate joint diagonalization and geometric mean of symmetric positive definite matrices. PLoS ONE 10(4), e0121423 (2015)
Congedo, M., Barachant, A., Koopaei, E.K.: Fixed point algorithms for estimating power means of positive definite matrices. IEEE Trans. Signal Process. 65(9), 2211–2220 (2017)
Dong, Z., et al.: Deep manifold learning of symmetric positive definite matrices with application to face recognition. In: Thirty-First AAAI Conference on Artificial Intelligence (2017)
Duerden, E.G., Tannock, R., Dockstader, C.: Altered cortical morphology in sensorimotor processing regions in adolescents and adults with attention-deficit/hyperactivity disorder. Brain Res. 1445, 82–91 (2012)
D’Souza, N.S., Nebel, M.B., Wymbs, N., Mostofsky, S., Venkataraman, A.: Integrating neural networks and dictionary learning for multidimensional clinical characterizations from functional connectomics data. In: Shen, D., et al. (eds.) MICCAI 2019. LNCS, vol. 11766, pp. 709–717. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-32248-9_79
D’Souza, N.S., et al.: A joint network optimization framework to predict clinical severity from resting state functional MRI data. Neuroimage 206, 116314 (2020)
D’Souza, N.S., Nebel, M.B., Crocetti, D., Robinson, J., Mostofsky, S., Venkataraman, A.: A matrix autoencoder framework to align the functional and structural connectivity manifolds as guided by behavioral phenotypes. In: de Bruijne, M., et al. (eds.) MICCAI 2021. LNCS, vol. 12907, pp. 625–636. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-87234-2_59
Fornito, A., Zalesky, A., Breakspear, M.: Graph analysis of the human connectome: promise, progress, and pitfalls. Neuroimage 80, 426–444 (2013)
Fox, M.D., et al.: Spontaneous fluctuations in brain activity observed with functional magnetic resonance imaging. Nat. Rev. Neuro. 8(9), 700 (2007)
Jeuris, B.: Riemannian optimization for averaging positive definite matrices (2015)
Jolliffe, I.T., Cadima, J.: Principal component analysis: a review and recent developments. Philos. Trans. R. Soc. A: Math. Phys. Eng. Sci. 374(2065), 20150202 (2016)
Khosla, M., et al.: Machine learning in resting-state fMRI analysis. Magn. Reson. Imaging 64, 101–121 (2019)
Leitner, Y.: The co-occurrence of autism and attention deficit hyperactivity disorder in children-what do we know? Front. Hum. Neurosci. 8, 268 (2014)
Lindquist, M.A.: The statistical analysis of fMRI data. Stat. Sci. 23(4), 439–464 (2008)
Moakher, M.: A differential geometric approach to the geometric mean of symmetric positive-definite matrices. SIAM J. Matrix Anal. Appl. 26(3), 735–747 (2005)
Nandakumar, N., et al.: A multi-task deep learning framework to localize the eloquent cortex in brain tumor patients using dynamic functional connectivity. In: Kia, S.M., et al. (eds.) MLCN/RNO-AI -2020. LNCS, vol. 12449, pp. 34–44. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-66843-3_4
Nguyen, X.S., et al.: A neural network based on SPD manifold learning for skeleton-based hand gesture recognition. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 12036–12045 (2019)
Pennec, X., Fillard, P., Ayache, N.: A Riemannian framework for tensor computing. Int. J. Comput. Vision 66(1), 41–66 (2006)
Pham, D.T.: Joint approximate diagonalization of positive definite Hermitian matrices. SIAM J. Matrix Anal. Appl. 22(4), 1136–1152 (2001)
Pouw, L.B., et al.: The link between emotion regulation, social functioning, and depression in boys with ASD. Res. Autism Spectr. Disord. 7(4), 549–556 (2013)
Schirmer, M.D., et al.: Neuropsychiatric disease classification using functional connectomics-results of the connectomics in neuroimaging transfer learning challenge. Med. Image Anal. 70, 101972 (2021)
Zalesky, A., Fornito, A., Bullmore, E.T.: Network-based statistic: identifying differences in brain networks. Neuroimage 53(4), 1197–1207 (2010)
Zhang, T.: A majorization-minimization algorithm for the karcher mean of positive definite matrices. arXiv preprint arXiv:1312.4654 (2013)
Acknowledgements
This work is supported by the National Science Foundation CAREER award 1845430 (PI Venkataraman), the National Institute of Health R01HD108790 (PI Venkataraman) and R01EB029977 (PI Caffo).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
D’Souza, N.S., Venkataraman, A. (2023). mSPD-NN: A Geometrically Aware Neural Framework for Biomarker Discovery from Functional Connectomics Manifolds. In: Frangi, A., de Bruijne, M., Wassermann, D., Navab, N. (eds) Information Processing in Medical Imaging. IPMI 2023. Lecture Notes in Computer Science, vol 13939. Springer, Cham. https://doi.org/10.1007/978-3-031-34048-2_5
Download citation
DOI: https://doi.org/10.1007/978-3-031-34048-2_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-34047-5
Online ISBN: 978-3-031-34048-2
eBook Packages: Computer ScienceComputer Science (R0)