Abstract
Deep learning models are being increasingly used in precision medicine thanks to their ability to provide accurate predictions of clinical outcome from large-scale datasets of patient’s records. However, in many cases data scarcity has forced the adoption of simpler (linear) feature extraction methods, which are less prone to overfitting. In this work, we exploit data augmentation and transfer learning techniques to show that deep, non-linear autoencoders can in fact extract relevant features from resting state functional connectivity matrices of stroke patients, even when the available data is modest. The latent representations extracted by the autoencoders can then be given as input to regularized regression methods to predict neurophsychological scores, significantly outperforming recently proposed methods based on linear feature extraction.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
Note that although the extracted features were obtained using the synthetic data, the model performance was always measured on the final stroke dataset.
- 2.
It should be pointed out that for these simulations we did not implement an exhaustive hyper-parameter optimization, as in the previous cases.
References
Greicius, M., Supekar, K., Menon, V., Dougherty, R.: Resting-state functional connectivity reflects structural connectivity in the default mode network. Cereb. Cortex 19, 72–8 (2008)
Meskaldji, D.E., et al.: Prediction of long-term memory scores in MCI based on resting-state FMRI. NeuroImage Clin. 12, 785–795 (2016)
Siegel, J.S., et al.: Disruptions of network connectivity predict impairment in multiple behavioral domains after stroke. Proc. Natl. Acad. Sci. 113(30), E4367–E4376 (2016)
Salvalaggio, A., De Filippo De Grazia, M., Zorzi, M., Thiebaut de Schotten, M., Corbetta, M.: Post-stroke deficit prediction from lesion and indirect structural and functional disconnection. Brain 143(7), 2173–2188 (2020)
Calesella, F., Testolin, A., De Filippo De Grazia, M., Zorzi, M.: A comparison of feature extraction methods for prediction of neuropsychological scores from functional connectivity data of stroke patients. Brain Inform. 8, 1–13 (2021)
Zorzi, M., De Filippo De Grazia, M., Blini, E., Testolin, A.: Assessment of machine learning pipelines for prediction of behavioral deficits from brain disconnectomes. In: Mahmud, M., Kaiser, M.S., Vassanelli, S., Dai, Q., Zhong, N. (eds.) BI 2021. LNCS (LNAI), vol. 12960, pp. 211–222. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-86993-9_20
Jollans, L., et al.: Quantifying performance of machine learning methods for neuroimaging data. Neuroimage 199, 351–365 (2019)
Bank, D., Koenigstein, N., Giryes, R.: Autoencoders. arXiv abs/2003.05991 (2020)
Kim, J.H., Zhang, Y., Han, K., Wen, Z., Choi, M., Liu, Z.: Representation learning of resting state fMRI with variational autoencoder. NeuroImage 241, 118423 (2021)
Huang, H., et al.: Modeling task fMRI data via deep convolutional autoencoder. IEEE Trans. Med. Imaging 37(7), 1551–1561 (2017)
Pinaya, W., Mechelli, A., Sato, J.: Using deep autoencoders to identify abnormal brain structural patterns in neuropsychiatric disorders: a large-scale multi-sample study. Hum. Brain Mapp. 40, 944–954 (2018)
GENG, X.F., Xu, J.: Application of autoencoder in depression diagnosis. DEStech Trans. Comput. Sci. Eng. (2017)
Van Essen, D.C., et al.: The WU-Minn human connectome project: an overview. Neuroimage 80, 62–79 (2013)
Cui, Z., Gong, G.: The effect of machine learning regression algorithms and sample size on individualized behavioral prediction with functional connectivity features. Neuroimage 178, 622–637 (2018)
Shalev-Shwartz, S., Ben-David, S.: Understanding Machine Learning - From Theory to Algorithms. Cambridge University Press, Cambridge (2014)
Cai, B., et al.: Functional connectome fingerprinting: identifying individuals and predicting cognitive functions via autoencoder. Hum. Brain Mapp. 42, 2691–2705 (2021)
Pedrycz, W., Chen, S.-M. (eds.): Deep Learning: Algorithms and Applications. SCI, vol. 865. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-31760-7
Scholz, M., Vigário, R.: Nonlinear PCA: a new hierarchical approach. In: ESANN (2002)
Masci, J., Meier, U., Cireşan, D., Schmidhuber, J.: Stacked convolutional auto-encoders for hierarchical feature extraction. In: Honkela, T., Duch, W., Girolami, M., Kaski, S. (eds.) ICANN 2011. LNCS, vol. 6791, pp. 52–59. Springer, Heidelberg (2011). https://doi.org/10.1007/978-3-642-21735-7_7
Akiba, T., Sano, S., Yanase, T., Ohta, T., Koyama, M.: Optuna: a next-generation hyperparameter optimization framework. In: Proceedings of the 25rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (2019)
Shorten, C., Khoshgoftaar, T.: A survey on image data augmentation for deep learning. J. Big Data 6, 1–48 (2019)
Zhang, H., Cissé, M., Dauphin, Y., Lopez-Paz, D.: Mixup: beyond empirical risk minimization. arXiv abs/1710.09412 (2018)
Isaksson, L., et al.: Mixup (sample pairing) can improve the performance of deep segmentation networks. J. Artif. Intell. Soft Comput. Res. 12, 29–39 (2022)
Hoerl, A., Kennard, R.: Ridge regression: biased estimation for nonorthogonal problems. Technometrics 12, 55–67 (2012)
Tibshirani, R.: Regression shrinkage selection via the lasso. J. Roy. Stat. Soc. Ser. B 73, 273–282 (2011)
Zou, H., Hastie, T.: regularization and variable selection via the elastic net. J. Roy. Stat. Soc. B 67(2), 301–320 (2005)
Baldi, P., Hornik, K.: Neural networks and principal component analysis: learning from examples without local minima. Neural Netw. 2(1), 53–58 (1989)
Singh, A., Sengupta, S., Lakshminarayanan, V.: Explainable deep learning models in medical image analysis. J. Imaging 6(6), 52 (2020)
Acknowledgements
This work was supported by grants from the Italian Ministry of Health (RF-2019-02359306 to MZ, Ricerca Corrente to IRCCS Ospedale San Camillo) and by MIUR (Dipartimenti di Eccellenza DM 11/05/2017 n. 262 to the Department of General Psychology). We are grateful to Prof. Maurizio Corbetta for providing the stroke dataset, which was collected in a study funded by grants R01 HD061117-05 and R01 NS095741. Healthy adults rs-fMRI data were provided by the Human Connectome Project, WU-Minn Consortium (Principal Investigators: David Van Essen and Kamil Ugurbil; 1U54MH091657) funded by the 16 NIH Institutes and Centers that support the NIH Blueprint for Neuroscience Research; and by the McDonnell Center for Systems Neuroscience at Washington University.
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Switzerland AG
About this paper
Cite this paper
Irarte, D., Testolin, A., De Filippo De Grazia, M., Zorzi, M. (2022). Prediction of Neuropsychological Scores from Functional Connectivity Matrices Using Deep Autoencoders. In: Mahmud, M., He, J., Vassanelli, S., van Zundert, A., Zhong, N. (eds) Brain Informatics. BI 2022. Lecture Notes in Computer Science(), vol 13406. Springer, Cham. https://doi.org/10.1007/978-3-031-15037-1_12
Download citation
DOI: https://doi.org/10.1007/978-3-031-15037-1_12
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-15036-4
Online ISBN: 978-3-031-15037-1
eBook Packages: Computer ScienceComputer Science (R0)