Abstract
Diagnosing brain dysconnectivity disorders at an early stage amounts to understanding the evolution of such abnormal connectivities over time. Ideally, without resorting to collecting more connectomic data over time, one would predict the disease evolution with high accuracy. At this point, generative learning models from limited data can come into play to predict brain connectomic evolution over time from a single acquisition timepoint. Here, we aim to bridge the gap between data scarcity and brain connectomic evolution prediction by proposing our novel Few-shot LeArning Training Network (FLAT-Net), the first framework leveraging the few-shot learning paradigm for brain connectivity evolution prediction from baseline timepoint. To do so, we introduce the concept of learning from representative connectional brain templates (CBTs), which encode the most centered and representative features (i.e., connectivities) for a given population of brain networks. Such CBTs capture well the data heterogeneity and diversity, hence they can train our predictive model in a frugal but generalizable manner. More specifically, our FLAT-Net starts by clustering the data into k clusters using the renowned K-means method. Then, for each cluster of homogenous brain networks, we create a CBT, which we call cluster specific-CBT (cs-CBT). We solely use each cs-CBT to train a distinct geometric generative adversarial network (gGAN) (i.e., for k clusters, we extract k cs-CBTs, and we train k gGANs (sub-model) each for a distinct cs-CBT) to learn the cs-CBT evolution over time. At the testing stage, we compute the Euclidean distance between the testing subject and each cs-CBT, and we select the gGAN model trained on the closest cs-CBT to the testing subject for prediction. A series of benchmarks against variants and excised interpretations of our framework showed that the proposed FLAT-Net, training strategy, and sub-model selection are promising strategies for predicting longitudinal brain alterations from only a few representative templates. Our FLAT-Net code is available at https://github.com/basiralab/FLAT-Net.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Price, R.B., Paul, B., Schneider, W., Siegle, G.J.: Neural correlates of three neurocognitive intervention strategies: a preliminary step towards personalized treatment for psychological disorders. Cogn. Ther. Res. 37, 657–672 (2013)
Tan, L., Jiang, T., Tan, L., Yu, J.T.: Toward precision medicine in neurological diseases. Ann. Transl. Med. 4, 104 (2016)
Ezzine, B.E., Rekik, I.: Learning-guided infinite network atlas selection for predicting longitudinal brain network evolution from a single observation. In: Shen, D., et al. (eds.) MICCAI 2019. LNCS, vol. 11765, pp. 796–805. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-32245-8_88
Nebli, A., Kaplan, U.A., Rekik, I.: Deep EvoGraphNet architecture for time-dependent brain graph data synthesis from a single timepoint. In: Rekik, I., Adeli, E., Park, S.H., Valdés Hernández, M.C. (eds.) PRIME 2020. LNCS, vol. 12329, pp. 144–155. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-59354-4_14
Bessadok, A., Mahjoub, M.A., Rekik, I.: Graph neural networks in network neuroscience (2021)
Goodfellow, I.J., et al.: Generative adversarial networks (2014)
Gurbuz, M.B., Rekik, I.: Deep graph normalizer: a geometric deep learning approach for estimating connectional brain templates. In: Martel, A.L., et al. (eds.) MICCAI 2020. LNCS, vol. 12267, pp. 155–165. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-59728-3_16
Macqueen, J.: Some methods for classification and analysis of multivariate observations. In: 5th Berkeley Symposium on Mathematical Statistics and Probability, pp. 281–297 (1967)
Mejia, A.F., Nebel, M.B., Eloyan, A., Caffo, B., Lindquist, M.A.: PCA leverage: outlier detection for high-dimensional functional magnetic resonance imaging data. Biostatistics 18, 521–536 (2017)
Snell, J., Swersky, K., Zemel, R.S.: Prototypical networks for few-shot learning. arXiv preprint arXiv:1703.05175 (2017)
Kolay, S., Ray, K.: K+ means: an enhancement over k-means clustering algorithm (2017)
Simonovsky, M., Komodakis, N.: Dynamic edge-conditioned filters in convolutional neural networks on graphs (2017)
Marcus, D., Fotenos, A., Csernansky, J., Morris, J., Buckner, R.: Open access series of imaging studies: longitudinal MRI data in nondemented and demented older adults. J. Cogn. Neurosci. 22, 2677–2684 (2010)
Mahjoub, I., Mahjoub, M., Rekik, I.: Brain multiplexes reveal morphological connectional biomarkers fingerprinting late brain dementia states. Sci. Rep. 8 (2018)
Nebli, A., Rekik, I.: Gender differences in cortical morphological networks. Brain Imaging Behav. 14(5), 1831–1839 (2019). https://doi.org/10.1007/s11682-019-00123-6
Fischl, B., et al.: Automatically parcellating the human cerebral cortex. Cereb. Cortex 14, 11–22 (2004)
Oldham, S., Fulcher, B., Parkes, L., Arnatkeviciūtė, A., Suo, C., Fornito, A.: Consistency and differences between centrality measures across distinct classes of networks. PLOS ONE 14, e0220061 (2019)
Loshchilov, I., Hutter, F.: Fixing weight decay regularization in adam (2017)
Murtagh, F., Contreras, P.: Methods of hierarchical clustering (2011)
Koelewijn, L., et al.: Alzheimer’s disease disrupts alpha and beta-band resting-state oscillatory network connectivity. Clin. Neurophysiol. 128, 2347–2357 (2017)
Binnewijzend, M., et al.: Brain network alterations in Alzheimer’s disease measured by eigenvector centrality in fMRI are related to cognition and cerebrospinal fluid biomarkers. Alzheimer’s Dementia 9, P684 (2013)
Acknowledgments
This work was funded by generous grants from the European H2020 Marie Sklodowska-Curie action (grant no. 101003403, http://basira-lab.com/normnets/) to I.R. and the Scientific and Technological Research Council of Turkey to I.R. under the TUBITAK 2232 Fellowship for Outstanding Researchers (no. 118C288, http://basira-lab.com/reprime/). However, all scientific contributions made in this project are owned and approved solely by the authors.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Özen, G., Nebli, A., Rekik, I. (2021). FLAT-Net: Longitudinal Brain Graph Evolution Prediction from a Few Training Representative Templates. In: Rekik, I., Adeli, E., Park, S.H., Schnabel, J. (eds) Predictive Intelligence in Medicine. PRIME 2021. Lecture Notes in Computer Science(), vol 12928. Springer, Cham. https://doi.org/10.1007/978-3-030-87602-9_25
Download citation
DOI: https://doi.org/10.1007/978-3-030-87602-9_25
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-87601-2
Online ISBN: 978-3-030-87602-9
eBook Packages: Computer ScienceComputer Science (R0)