Abstract
Identification of neonatal brain abnormalities on MRI requires expert knowledge of both brain development and pathologies particular to this age group. To aid this process we propose an automated technique to highlight abnormal brain tissue while accommodating normal developmental changes. To train a developmental model, we used 185 T2 weighted neuroimaging datasets from healthy controls and preterm infants without obvious lesions on MRI age range = 27 + 5 − 51 (median = 40 + 5) weeks + days post menstrual age (PMA). We then tested the model on 39 preterm subjects with known pathology age range = 25 + 1 − 37 + 5 (median = 33) weeks PMA + days. We used voxel-wise Gaussian processes (GP) to model age (PMA) and sex against voxel intensity, where each GP outputs a predicted mean intensity and variance for that location. All GP outputs were combined to synthesize a ‘normal’ T2 image and corresponding variance map for age and sex. A Z-score map was created by calculating the difference between the neonate’s actual image and their synthesized image and scaling by the standard deviation (SD). With a threshold of 3 SD the model highlighted pathologies including germinal matrix, intraventricular and cerebellar hemorrhage, cystic periventricular leukomalacia and punctate lesions. Statistical analysis of the abnormality detection produces an average AUC value of 0.956 and 0.943 against two raters’ manual segmentations. The proposed method is effective at highlighting different abnormalities across the whole brain in the perinatal period while limiting false positives.
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
References
Imai, K., et al.: MRI changes in the thalamus and basal ganglia of full-term neonates with perinatal asphyxia. Neonatology 114, 253–260 (2018)
Kline-Fath, B.M., et al.: Conventional MRI scan and DTI imaging show more severe brain injury in neonates with hypoxic-ischemic encephalopathy and seizures. Early Hum. Dev. 122, 8–14 (2018)
Chaturvedi, A., et al.: Mechanical birth-related trauma to the neonate: an imaging perspective. Insights Imaging 9(1), 103–118 (2018). https://doi.org/10.1007/s13244-017-0586-x
Bogner, M.S.: Human Error in Medicine. CRC Press, Boca Raton (2018)
Pereira, S., et al.: Automatic brain tissue segmentation in MR images using random forests and conditional random fields. J. Neurosci. Methods 270, 111–123 (2016)
Bahadure, N.B., et al.: Image analysis for MRI based brain tumor detection and feature extraction using biologically inspired BWT and SVM. Int. J. Biomed. Imaging 2017, 12 pages (2017)
Rezaei, M., et al.: Brain abnormality detection by deep convolutional neural network. arXiv preprint arXiv:1708.05206(2017)
El Azami, M., et al.: Detection of lesions underlying intractable epilepsy on T1-weighted MRI as an outlier detection problem. PLoS ONE 11(9), e0161498 (2016)
Chen, X., et al.: Unsupervised detection of lesions in brain MRI using constrained adversarial auto-encoders. arXiv preprint arXiv:1806.04972(2018)
Bowles, C., et al.: Brain lesion segmentation through image synthesis and outlier detection. NeuroImage Clin. 16, 643–658 (2017)
O’Muircheartaigh, J., et al. “Modelling brain development to detect white matter injury in term and preterm born neonates. Brain, 143, 467–479 (2020)
Avants, B.B., et al.: Advanced normalization tools (ANTS). Insight J. 2, 1–35 (2009)
Smith, S.M.: Fast robust automated brain extraction. Hum. Brain Mapp. 17(3), 143–155 (2002)
Makropoulos, A., et al.: The developing human connectome project: a minimal processing pipeline for neonatal cortical surface reconstruction. Neuroimage 173, 88–112 (2018)
Jenkinson, M.: A global optimisation method for robust affine registration of brain images. Med. Image Anal. 5(2), 143–156 (2001)
Jenkinson, M., et al.: Improved optimisation for the robust and accurate linear registration and motion correction of brain images. NeuroImage 17(2), 825–841 (2002)
Rasmussen, C.E.: Gaussian processes in machine learning. In: Bousquet, O., von Luxburg, U., Rätsch, G. (eds.) ML-2003. LNCS (LNAI), vol. 3176, pp. 63–71. Springer, Heidelberg (2004). https://doi.org/10.1007/978-3-540-28650-9_4
Kingma, D.P., et al.: Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)
Gardner, J., et al.: GPyTorch: blackbox matrix-matrix Gaussian process inference with GPU acceleration. In: Advances in Neural Information Processing Systems (2018)
Acknowledgments
This work was supported by The Wellcome/EPSRC Centre for Medical Engineering at Kings College London (WT 203148/Z/16/Z), the NIHR Clinical Research Facility (CRF) at Guy’s and St Thomas’ and by the National Institute for Health Research Biomedical Research Centres based at Guy’s and St Thomas’ NHS Foundation Trust, and South London, Maudsley NHS Foundation Trust. The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health. R.M’s PhD is supported by the EPSRC Centre for Doctoral Training in Smart Medical Imaging at King’s College London. J.O. is supported by a Sir Henry Dale Fellowship jointly funded by the Wellcome Trust and the Royal Society (grant 206675/Z/17/Z). J.O. received support from the Medical Research Council Centre for Neurodevelopmental Disorders, King’s College London (grant MR/N026063/1). The project includes data from a programme of research funded by the NIHR Programme Grants for Applied Research Programme (RP-PG-0707-10154). The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, MRIC, CCF, NETSCC, the Programme Grants for Applied Research programme or the Department of Health.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Macleod, R., O’Muircheartaigh, J., Edwards, A.D., Carmichael, D., Rutherford, M., Counsell, S.J. (2020). Automatic Detection of Neonatal Brain Injury on MRI. In: Hu, Y., et al. Medical Ultrasound, and Preterm, Perinatal and Paediatric Image Analysis. ASMUS PIPPI 2020 2020. Lecture Notes in Computer Science(), vol 12437. Springer, Cham. https://doi.org/10.1007/978-3-030-60334-2_32
Download citation
DOI: https://doi.org/10.1007/978-3-030-60334-2_32
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-60333-5
Online ISBN: 978-3-030-60334-2
eBook Packages: Computer ScienceComputer Science (R0)
