Skip to main content
main-content
Top

Hint

Swipe to navigate through the chapters of this book

2018 | OriginalPaper | Chapter

Inherent Brain Segmentation Quality Control from Fully ConvNet Monte Carlo Sampling

Authors : Abhijit Guha Roy, Sailesh Conjeti, Nassir Navab, Christian Wachinger

Published in: Medical Image Computing and Computer Assisted Intervention – MICCAI 2018

Publisher: Springer International Publishing

share
SHARE

Abstract

We introduce inherent measures for effective quality control of brain segmentation based on a Bayesian fully convolutional neural network, using model uncertainty. Monte Carlo samples from the posterior distribution are efficiently generated using dropout at test time. Based on these samples, we introduce next to a voxel-wise uncertainty map also three metrics for structure-wise uncertainty. We then incorporate these structure-wise uncertainty in group analyses as a measure of confidence in the observation. Our results show that the metrics are highly correlated to segmentation accuracy and therefore present an inherent measure of segmentation quality. Furthermore, group analysis with uncertainty results in effect sizes closer to that of manual annotations. The introduced uncertainty metrics can not only be very useful in translation to clinical practice but also provide automated quality control and group analyses in processing large data repositories.
Literature
1.
go back to reference Jenkinson, M., Beckmann, C.F., Behrens, T.E., Woolrich, M.W., Smith, S.M.: FSL. Neuroimage 62(2), 782–790 (2012) CrossRef Jenkinson, M., Beckmann, C.F., Behrens, T.E., Woolrich, M.W., Smith, S.M.: FSL. Neuroimage 62(2), 782–790 (2012) CrossRef
2.
go back to reference Fischl, B., Salat, D.H., Busa, E., Albert, M., et al.: Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain. Neuron 33(3), 341–355 (2002) CrossRef Fischl, B., Salat, D.H., Busa, E., Albert, M., et al.: Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain. Neuron 33(3), 341–355 (2002) CrossRef
3.
go back to reference Keshavan, A.: Mindcontrol: a web application for brain segmentation quality control. NeuroImage 170, 365–372 (2018) CrossRef Keshavan, A.: Mindcontrol: a web application for brain segmentation quality control. NeuroImage 170, 365–372 (2018) CrossRef
4.
go back to reference Iglesias, J.E., Sabuncu, M.R., Van Leemput, K.: Improved inference in Bayesian segmentation using Monte Carlo sampling: application to hippocampal subfield volumetry. Med. Image Anal. 17(7), 766–778 (2013) CrossRef Iglesias, J.E., Sabuncu, M.R., Van Leemput, K.: Improved inference in Bayesian segmentation using Monte Carlo sampling: application to hippocampal subfield volumetry. Med. Image Anal. 17(7), 766–778 (2013) CrossRef
5.
go back to reference Wachinger, C., Fritscher, K., Sharp, G., Golland, P.: Contour-driven atlas-based segmentation. IEEE TMI 34(12), 2492–2505 (2015) Wachinger, C., Fritscher, K., Sharp, G., Golland, P.: Contour-driven atlas-based segmentation. IEEE TMI 34(12), 2492–2505 (2015)
6.
go back to reference Gal, Y., Ghahramani, Z.: Dropout as a Bayesian approximation: representing model uncertainty in deep learning. In: Proceedings of ICML, pp. 1050–1059 (2016) Gal, Y., Ghahramani, Z.: Dropout as a Bayesian approximation: representing model uncertainty in deep learning. In: Proceedings of ICML, pp. 1050–1059 (2016)
7.
go back to reference Kendall, A., Badrinarayanan, V., Cipolla, R.: Bayesian segnet: model uncertainty in deep convolutional encoder-decoder architectures for scene understanding. In: Proceedings of BMVC (2017) Kendall, A., Badrinarayanan, V., Cipolla, R.: Bayesian segnet: model uncertainty in deep convolutional encoder-decoder architectures for scene understanding. In: Proceedings of BMVC (2017)
9.
go back to reference Roy, A.G., Conjeti, S., Sheet, D., Katouzian, A., Navab, N., Wachinger, C.: Error corrective boosting for learning fully convolutional networks with limited data. In: Descoteaux, M., Maier-Hein, L., Franz, A., Jannin, P., Collins, D.L., Duchesne, S. (eds.) MICCAI 2017. LNCS, vol. 10435, pp. 231–239. Springer, Cham (2017). https://​doi.​org/​10.​1007/​978-3-319-66179-7_​27 CrossRef Roy, A.G., Conjeti, S., Sheet, D., Katouzian, A., Navab, N., Wachinger, C.: Error corrective boosting for learning fully convolutional networks with limited data. In: Descoteaux, M., Maier-Hein, L., Franz, A., Jannin, P., Collins, D.L., Duchesne, S. (eds.) MICCAI 2017. LNCS, vol. 10435, pp. 231–239. Springer, Cham (2017). https://​doi.​org/​10.​1007/​978-3-319-66179-7_​27 CrossRef
10.
go back to reference Bouix, S., et al.: On evaluating brain tissue classifiers without a ground truth. Neuroimage 36(4), 1207–1224 (2007) CrossRef Bouix, S., et al.: On evaluating brain tissue classifiers without a ground truth. Neuroimage 36(4), 1207–1224 (2007) CrossRef
12.
go back to reference Valindria, V., et al.: Reverse classification accuracy: predicting segmentation performance in the absence of ground truth. TMI 36(8), 1597–1606 (2017) Valindria, V., et al.: Reverse classification accuracy: predicting segmentation performance in the absence of ground truth. TMI 36(8), 1597–1606 (2017)
13.
go back to reference Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., Salakhutdinov, R.: Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15(1), 1929–1958 (2014) MathSciNetMATH Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., Salakhutdinov, R.: Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15(1), 1929–1958 (2014) MathSciNetMATH
14.
go back to reference Landman, B, Warfield, S.: Miccai workshop on multiatlas labeling. In: MICCAI Grand Challenge (2012) Landman, B, Warfield, S.: Miccai workshop on multiatlas labeling. In: MICCAI Grand Challenge (2012)
15.
go back to reference Jack, C.R., Bernstein, M.A., Fox, N.C., Thompson, P., et al.: The Alzheimer’s disease neuroimaging initiative (ADNI): MRI methods. JMRI 27(4), 685–691 (2008) CrossRef Jack, C.R., Bernstein, M.A., Fox, N.C., Thompson, P., et al.: The Alzheimer’s disease neuroimaging initiative (ADNI): MRI methods. JMRI 27(4), 685–691 (2008) CrossRef
16.
go back to reference Kennedy, D.N., Haselgrove, C., Hodge, S.M., Rane, P.S., Makris, N., Frazier, J.A.: CANDIShare: a resource for pediatric neuroimaging data (2012) Kennedy, D.N., Haselgrove, C., Hodge, S.M., Rane, P.S., Makris, N., Frazier, J.A.: CANDIShare: a resource for pediatric neuroimaging data (2012)
17.
go back to reference Thompson, P.M., Hayashi, K.M., et al.: Mapping hippocampal and ventricular change in Alzheimer disease. Neuroimage 22(4), 1754–1766 (2004) CrossRef Thompson, P.M., Hayashi, K.M., et al.: Mapping hippocampal and ventricular change in Alzheimer disease. Neuroimage 22(4), 1754–1766 (2004) CrossRef
18.
go back to reference Di Martino, A., Yan, C.G., Li, Q., Denio, E.: The autism brain imaging data exchange: towards a large-scale evaluation of the intrinsic brain architecture in autism. Mol. Psychiatry 19(6), 659 (2014) CrossRef Di Martino, A., Yan, C.G., Li, Q., Denio, E.: The autism brain imaging data exchange: towards a large-scale evaluation of the intrinsic brain architecture in autism. Mol. Psychiatry 19(6), 659 (2014) CrossRef
19.
go back to reference Van Rooij, D., Anagnostou, E., Arango, C., Auzias, G., et al.: Cortical and subcortical brain morphometry differences between patients with autism spectrum disorder and healthy individuals across the lifespan: results from the ENIGMA ASD working group. Am. J. Psychiatry 175(4), 359–369 (2018) CrossRef Van Rooij, D., Anagnostou, E., Arango, C., Auzias, G., et al.: Cortical and subcortical brain morphometry differences between patients with autism spectrum disorder and healthy individuals across the lifespan: results from the ENIGMA ASD working group. Am. J. Psychiatry 175(4), 359–369 (2018) CrossRef
Metadata
Title
Inherent Brain Segmentation Quality Control from Fully ConvNet Monte Carlo Sampling
Authors
Abhijit Guha Roy
Sailesh Conjeti
Nassir Navab
Christian Wachinger
Copyright Year
2018
DOI
https://doi.org/10.1007/978-3-030-00928-1_75

Premium Partner