Automated brain extraction of multisequence MRI using artificial neural networks

Brain extraction is a critical preprocessing step in the analysis of neuroimaging studies conducted with magnetic resonance imaging (MRI) and influences the accuracy of downstream analyses. The majority of brain extraction algorithms are, however, optimized for processing healthy brains and thus frequently fail in the presence of pathologically altered brain or when applied to heterogeneous MRI datasets. Here we introduce a new, rigorously validated algorithm (termed HD‐BET) relying on artificial neural networks that aim to overcome these limitations. We demonstrate that HD‐BET outperforms six popular, publicly available brain extraction algorithms in several large‐scale neuroimaging datasets, including one from a prospective multicentric trial in neuro‐oncology, yielding state‐of‐the‐art performance with median improvements of +1.16 to +2.50 points for the Dice coefficient and −0.66 to −2.51 mm for the Hausdorff distance. Importantly, the HD‐BET algorithm, which shows robust performance in the presence of pathology or treatment‐induced tissue alterations, is applicable to a broad range of MRI sequence types and is not influenced by variations in MRI hardware and acquisition parameters encountered in both research and clinical practice. For broader accessibility, the HD‐BET prediction algorithm is made freely available (www.neuroAI-HD.org) and may become an essential component for robust, automated, high‐throughput processing of MRI neuroimaging data.

[1]  Jerry L. Prince,et al.  Cortical reconstruction using implicit surface evolution: Accuracy and precision analysis , 2006, NeuroImage.

[2]  Patrick van der Smagt,et al.  CNN-based Segmentation of Medical Imaging Data , 2017, ArXiv.

[3]  Konstantinos Kamnitsas,et al.  Efficient multi‐scale 3D CNN with fully connected CRF for accurate brain lesion segmentation , 2016, Medical Image Anal..

[4]  Jeffrey N. Chiang,et al.  Optimized Brain Extraction for Pathological Brains (optiBET) , 2014, PloS one.

[5]  Allan Hanbury,et al.  An Efficient Algorithm for Calculating the Exact Hausdorff Distance , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  Li Wang,et al.  Level set segmentation of brain magnetic resonance images based on local Gaussian distribution fitting energy , 2010, Journal of Neuroscience Methods.

[7]  Ulla Ruotsalainen,et al.  Automatic cerebral and cerebellar hemisphere segmentation in 3D MRI: Adaptive disconnection algorithm , 2010, Medical Image Anal..

[8]  Heinz-Otto Peitgen,et al.  The Skull Stripping Problem in MRI Solved by a Single 3D Watershed Transform , 2000, MICCAI.

[9]  Nikolaos Papanikolopoulos,et al.  Imperfect Segmentation Labels: How Much Do They Matter? , 2018, CVII-STENT/LABELS@MICCAI.

[10]  Guido Gerig,et al.  User-guided 3D active contour segmentation of anatomical structures: Significantly improved efficiency and reliability , 2006, NeuroImage.

[11]  H. Haidar,et al.  Measurement of Cortical Thickness in 3D Brain MRI Data: Validation of the Laplacian Method , 2006, Journal of neuroimaging : official journal of the American Society of Neuroimaging.

[12]  W. A. Hanson,et al.  Interactive 3D segmentation of MRI and CT volumes using morphological operations. , 1992, Journal of computer assisted tomography.

[13]  R. Leahy,et al.  Magnetic Resonance Image Tissue Classification Using a Partial Volume Model , 2001, NeuroImage.

[14]  Xu Han,et al.  Brain extraction from normal and pathological images: A joint PCA/Image-Reconstruction approach , 2017, NeuroImage.

[15]  Peter König,et al.  Data augmentation instead of explicit regularization , 2018, ArXiv.

[16]  Bruce Fischl,et al.  Accurate and robust brain image alignment using boundary-based registration , 2009, NeuroImage.

[17]  Jacoline E. C. Bromberg,et al.  Phase II part of EORTC study 26101: The sequence of bevacizumab and lomustine in patients with first recurrence of a glioblastoma. , 2016 .

[18]  Mark W. Woolrich,et al.  Bayesian analysis of neuroimaging data in FSL , 2009, NeuroImage.

[19]  D Louis Collins,et al.  An efficient and accurate method for robust inter‐dataset brain extraction and comparisons with 9 other methods , 2018, Human brain mapping.

[20]  Alan C. Evans,et al.  Automated 3-D Extraction of Inner and Outer Surfaces of Cerebral Cortex from MRI , 2000, NeuroImage.

[21]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[22]  Martin Klein,et al.  Lomustine and Bevacizumab in Progressive Glioblastoma , 2017, The New England journal of medicine.

[23]  Snehashis Roy,et al.  Robust skull stripping using multiple MR image contrasts insensitive to pathology , 2017, NeuroImage.

[24]  Seyed-Ahmad Ahmadi,et al.  V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation , 2016, 2016 Fourth International Conference on 3D Vision (3DV).

[25]  Bruce R. Rosen,et al.  DeepNeuro: an open-source deep learning toolbox for neuroimaging , 2018, Neuroinformatics.

[26]  Jacob Cohen Statistical Power Analysis for the Behavioral Sciences , 1969, The SAGE Encyclopedia of Research Design.

[27]  Stephen M. Smith,et al.  Segmentation of brain MR images through a hidden Markov random field model and the expectation-maximization algorithm , 2001, IEEE Transactions on Medical Imaging.

[28]  D. Louis Collins,et al.  BEaST: Brain extraction based on nonlocal segmentation technique , 2012, NeuroImage.

[29]  Anders M. Dale,et al.  Cortical Surface-Based Analysis I. Segmentation and Surface Reconstruction , 1999, NeuroImage.

[30]  Ludwig Kappos,et al.  Correlation between brain volume loss and clinical and MRI outcomes in multiple sclerosis , 2015, Neurology.

[31]  J. Mazziotta,et al.  MRI‐PET Registration with Automated Algorithm , 1993, Journal of computer assisted tomography.

[32]  Arthur W. Toga,et al.  Construction of a 3D probabilistic atlas of human cortical structures , 2008, NeuroImage.

[33]  Klaus H. Maier-Hein,et al.  Deep MRI brain extraction: A 3D convolutional neural network for skull stripping , 2016, NeuroImage.

[34]  Guido Gerig,et al.  Elastic model-based segmentation of 3-D neuroradiological data sets , 1999, IEEE Transactions on Medical Imaging.

[35]  Stephen M Smith,et al.  Fast robust automated brain extraction , 2002, Human brain mapping.

[36]  Stephen M. Smith,et al.  A global optimisation method for robust affine registration of brain images , 2001, Medical Image Anal..

[37]  L. R. Dice Measures of the Amount of Ecologic Association Between Species , 1945 .

[38]  Nick C Fox,et al.  The clinical use of structural MRI in Alzheimer disease , 2010, Nature Reviews Neurology.

[39]  Jian Sun,et al.  Identity Mappings in Deep Residual Networks , 2016, ECCV.

[40]  Gregory G. Brown,et al.  Quantitative evaluation of automated skull‐stripping methods applied to contemporary and legacy images: Effects of diagnosis, bias correction, and slice location , 2006, Human brain mapping.

[41]  Deniz Erdogmus,et al.  Auto-Context Convolutional Neural Network (Auto-Net) for Brain Extraction in Magnetic Resonance Imaging , 2017, IEEE Transactions on Medical Imaging.

[42]  Richard M. Leahy,et al.  BrainSuite: An Automated Cortical Surface Identification Tool , 2000, MICCAI.

[43]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[44]  Thomas Brox,et al.  3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation , 2016, MICCAI.

[45]  Yi Hong,et al.  CompNet: Complementary Segmentation Network for Brain MRI Extraction , 2018, MICCAI.

[46]  P. Kalavathi,et al.  Methods on Skull Stripping of MRI Head Scan Images—a Review , 2016, Journal of Digital Imaging.

[47]  R Cameron Craddock,et al.  The preprocessed connectomes project repository of manually corrected skull-stripped T1-weighted anatomical MRI data , 2016, bioRxiv.

[48]  Sébastien Ourselin,et al.  Generalised Dice overlap as a deep learning loss function for highly unbalanced segmentations , 2017, DLMIA/ML-CDS@MICCAI.

[49]  Richard Frayne,et al.  An open, multi-vendor, multi-field-strength brain MR dataset and analysis of publicly available skull stripping methods agreement , 2017, NeuroImage.

[50]  Christopher Joseph Pal,et al.  Brain tumor segmentation with Deep Neural Networks , 2015, Medical Image Anal..

[51]  Jian Chen,et al.  Brain extraction using the watershed transform from markers , 2013, Front. Neuroinform..

[52]  Wiro J. Niessen,et al.  Accuracy and reproducibility study of automatic MRI brain tissue segmentation methods , 2010, NeuroImage.

[53]  David Bonekamp,et al.  Automated brain extraction of multisequence MRI using artificial neural networks , 2019, Human brain mapping.

[54]  Satrajit S. Ghosh,et al.  Evaluation of volume-based and surface-based brain image registration methods , 2010, NeuroImage.

[55]  R W Cox,et al.  AFNI: software for analysis and visualization of functional magnetic resonance neuroimages. , 1996, Computers and biomedical research, an international journal.

[56]  Rita G. Nunes,et al.  Reconstruction of white matter fibre tracts using diffusion kurtosis tensor imaging at 1.5T: Pre-surgical planning in patients with gliomas , 2018, European journal of radiology open.

[57]  S. Heiland,et al.  Automated quantitative tumour response assessment of MRI in neuro-oncology with artificial neural networks: a multicentre, retrospective study. , 2019, The Lancet. Oncology.

[58]  Fang Liu,et al.  Bayesian convolutional neural network based MRI brain extraction on nonhuman primates , 2018, NeuroImage.

[59]  Paul M. Thompson,et al.  Robust Brain Extraction Across Datasets and Comparison With Publicly Available Methods , 2011, IEEE Transactions on Medical Imaging.

[60]  Mark Jenkinson,et al.  Optimizing parameter choice for FSL-Brain Extraction Tool (BET) on 3D T1 images in multiple sclerosis , 2012, NeuroImage.

[61]  Klaus H. Maier-Hein,et al.  Brain Tumor Segmentation and Radiomics Survival Prediction: Contribution to the BRATS 2017 Challenge , 2017, BrainLes@MICCAI.

[62]  Brian B. Avants,et al.  The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) , 2015, IEEE Transactions on Medical Imaging.