Advertisement

Uncertainty-Informed Detection of Epileptogenic Brain Malformations Using Bayesian Neural Networks

Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11767)

Abstract

Focal cortical dysplasia (FCD) is a prevalent surgically-amenable epileptogenic malformation of cortical development. On MRI, FCD typically presents with cortical thickening, hyperintensity, and blurring of the gray-white matter interface. These changes may be visible to the naked eye, or subtle and be easily overlooked. Despite advances in MRI analytics, current machine learning algorithms fail to detect FCD in up to 50% of cases. Moreover, the deterministic nature of current algorithms does not allow conducting risk assessments of such predictions, an essential step in clinical decision-making. Here, we propose an algorithm formulated on Bayesian convolutional neural networks (CNN) providing information on prediction uncertainty, while leveraging this information to improve classification performance. Our classifier was trained on a patch-based augmented dataset derived from 56 patients with histologically-validated FCD to distinguish the lesion from healthy tissue. The algorithm was trained and cross-validated on multimodal 3T MRI data. Compared to a non-Bayesian learner with the same network architecture and complexity, the uncertainty-informed Bayesian CNN classifiers showed significant improvement in sensitivity (89% vs 82%; p < 0.05) while specificity was high for both classifiers. We demonstrate empirically the effectiveness of our uncertainty-informed CNN algorithm, making it ideal for large-scale clinical diagnostics of FCD.

Keywords

Magnetic resonance imaging Clinical diagnostics Epilepsy Deep learning Uncertainty Dropout Monte Carlo Classification 

Supplementary material

490278_1_En_25_MOESM1_ESM.pdf (2.7 mb)
Supplementary material 1 (PDF 2748 kb)Supplementary material 1 (PDF 2748 kb)

References

  1. 1.
    Blümcke, I., et al.: The clinicopathologic spectrum of focal cortical dysplasias: a consensus classification proposed by an ad hoc Task Force of the ILAE Diagnostic Methods Commission. Epilepsia 52, 158–174 (2011)CrossRefGoogle Scholar
  2. 2.
    Fauser, S., et al.: Long-term seizure outcome in 211 patients with focal cortical dysplasia. Epilepsia. 56, 66–76 (2015)CrossRefGoogle Scholar
  3. 3.
    Bernasconi, A., Bernasconi, N., Bernhardt, B.C., Schrader, D.: Advances in MRI for “cryptogenic” epilepsies. Nat. Rev. Neurol. 7, 99–108 (2011)CrossRefGoogle Scholar
  4. 4.
    Kini, L.G., Gee, J.C., Litt, B.: Computational analysis in epilepsy neuroimaging: a survey of features and methods. NeuroImage Clin. 11, 515–529 (2016)CrossRefGoogle Scholar
  5. 5.
    Hong, S.-J., Kim, H., Schrader, D., Bernasconi, N., Bernhardt, B.C., Bernasconi, A.: Automated detection of cortical dysplasia type II in MRI-negative epilepsy. Neurology. 83, 48–55 (2014)CrossRefGoogle Scholar
  6. 6.
    Adler, S., et al.: Novel surface features for automated detection of focal cortical dysplasias in paediatric epilepsy. NeuroImage Clin. 14, 18–27 (2017)CrossRefGoogle Scholar
  7. 7.
    Gill, R.S., et al.: Automated detection of epileptogenic cortical malformations using multimodal MRI. In: Cardoso, M.J., et al. (eds.) DLMIA/ML-CDS 2017. LNCS, vol. 10553, pp. 349–356. Springer, Cham (2017).  https://doi.org/10.1007/978-3-319-67558-9_40CrossRefGoogle Scholar
  8. 8.
    LeCun, Y., Bengio, Y., Hinton, G.: Deep learning. Nature 521, 436–444 (2015)CrossRefGoogle Scholar
  9. 9.
    Litjens, G., et al.: A survey on deep learning in medical image analysis. Med. Image Anal. 42, 60–88 (2017)CrossRefGoogle Scholar
  10. 10.
    Guo, C., Pleiss, G., Sun, Y., Weinberger, K.Q.: On Calibration of Modern Neural Networks. http://arxiv.org/abs/1706.04599v2 (2017)
  11. 11.
    Nair, T., Precup, D., Arnold, D.L., Arbel, T.: Exploring Uncertainty Measures in Deep Networks for Multiple Sclerosis Lesion Detection and Segmentation. http://arxiv.org/abs/1808.01200v2 (2018)
  12. 12.
    Leibig, C., Allken, V., Ayhan, M.S., Berens, P., Wahl, S.: Leveraging uncertainty information from deep neural networks for disease detection. Sci. Rep. 7, 17816 (2017)CrossRefGoogle Scholar
  13. 13.
    Gal, Y., Ghahramani, Z.: Bayesian Convolutional Neural Networks with Bernoulli Approximate Variational Inference. http://arxiv.org/abs/1506.02158v6 (2015)
  14. 14.
    Kendall, A., Gal, Y.: What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision? http://arxiv.org/abs/1703.04977v2 (2017)
  15. 15.
    Sled, J.G., Zijdenbos, A.P., Evans, A.C.: A nonparametric method for automatic correction of intensity nonuniformity in MRI data. IEEE Trans. Med. Imaging 17, 87–97 (1998)CrossRefGoogle Scholar
  16. 16.
    Fonov, V.S., Evans, A.C., McKinstry, R.C., Almli, C.R., Collins, D.L.: Unbiased nonlinear average age-appropriate brain templates from birth to adulthood. Neuroimage 47, S102 (2009)CrossRefGoogle Scholar
  17. 17.
    Sisodiya, S.M., Fauser, S., Cross, J.H., Thom, M.: Focal cortical dysplasia type II: biological features and clinical perspectives. Lancet Neurol. 8, 830–843 (2009)CrossRefGoogle Scholar
  18. 18.
    Hong, S.-J., et al.: Multimodal MRI profiling of focal cortical dysplasia type II. Neurology 88, 734–742 (2017)CrossRefGoogle Scholar
  19. 19.
    Gill, R.S., et al.: Deep convolutional networks for automated detection of epileptogenic brain malformations. In: Frangi, A.F., Schnabel, J.A., Davatzikos, C., Alberola-López, C., Fichtinger, G. (eds.) MICCAI 2018. LNCS, vol. 11072, pp. 490–497. Springer, Cham (2018).  https://doi.org/10.1007/978-3-030-00931-1_56CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Neuroimaging of Epilepsy Laboratory, McConnell Brain Imaging CenterMontreal Neurological Institute (MNI)MontrealCanada

Personalised recommendations