An efficient and accurate method for robust inter‐dataset brain extraction and comparisons with 9 other methods

Abstract

Brain extraction is an important first step in many magnetic resonance neuroimaging studies. Due to variability in brain morphology and in the appearance of the brain due to differences in scanner acquisition parameters, the development of a generally applicable brain extraction algorithm has proven challenging. Learning-based brain extraction algorithms in particular perform well when the target and training images are sufficiently similar, but often perform worse when this condition is not met. In this study, we propose a new patch-based multi-atlas segmentation method for brain extraction which is specifically developed for accurate and robust processing across datasets. Using a diverse collection of labeled images from 5 different datasets, extensive comparisons were made with 9 other commonly used brain extraction methods, both before and after applying error correction (a machine learning method for automatically correcting segmentation errors) to each method. The proposed method performed equal to or better than the other methods in each of two segmentation scenarios: a challenging inter-dataset segmentation scenario in which no dataset-specific atlases were used (mean Dice coefficient 98.57%, volumetric correlation 0.994 across datasets following error correction), and an intra-dataset segmentation scenario in which only dataset-specific atlases were used (mean Dice coefficient 99.02%, volumetric correlation 0.998 across datasets following error correction). Furthermore, combined with error correction, the proposed method runs in less than one-tenth of the time required by the other top-performing methods in the challenging inter-dataset comparisons. Validation on an independent multi-centre dataset also confirmed the excellent performance of the proposed method.

Automated brain extraction of multisequence MRI using artificial neural networks

“…Moreover, a case‐specific tuning of parameters from these brain extraction algorithms may have allowed to improve their performance to some extent (Iglesias et al, ; Popescu, et al, ). This is particularly the case for BEaST, where a mismatch between source and target domain can result in a significant drop in performance (Eskildsen et al, ; Novosad & Collins, ). Dataset‐specific adaptations are, however, not a practical approach, especially in the context of high‐throughput processing.…”

“…For every test, we reported the absolute value of the Z-statistics [abs(Z)], the Bonferroni-adjusted p-value and the effect size [r] (with r values >.1 corresponding to a small effect, .3 to a medium effect, and .5 to a large effect size; Cohen, 1988 tuning of parameters from these brain extraction algorithms may have allowed to improve their performance to some extent (Iglesias et al, 2011;Popescu, et al, 2012). This is particularly the case for BEaST, where a mismatch between source and target domain can result in a significant drop in performance (Eskildsen et al, 2012;Novosad & Collins, 2018). Dataset-specific adaptations are, however, not a practical approach, especially in the context of high-throughput processing.…”

Accurate and robust segmentation of neuroanatomy in T1‐weighted MRI by combining spatial priors with deep convolutional neural networks

“…Some such methods have attempted to learn complex mappings between image features and labels using traditional machine-learning based classifiers (e.g., support vector machines (Boser, Guyon, & Vapnik, 1992) and random forests (Breiman, 2001)) combined with handcrafted feature sets (Morra et al, 2010;Zikic et al, 2012), while others have found success transferring labels using a combination of linear or nonlinear image registration with local and/or nonlocal label fusion (so-called "multiatlas segmentation" methods (Coupé et al, 2011, Heckemann, Hajnal, Aljabar, Rueckert, & Hammers, 2006, Iglesias & Sabuncu, 2015). Indeed, many state-of-the-art results (e.g., hippocampus segmentation (Zandifar, Fonov, Coupé, Pruessner, & Collins, 2017) and brain extraction (Novosad & Collins, 2018) exploit a complementary combination of both multiatlas segmentation and machine-learning methods (e.g., error correction (EC) (Wang et al, 2011)).…”

“…Indeed, many state-of-the-art results (e.g. hippocampus segmentation (Zandifar et al, 2017) and brain extraction (Novosad and Collins, 2018)) exploit a complementary combination of both multi-atlas segmentation and machine-learning methods (e.g. error correction (Wang et al, 2011)), though the computational time required for segmentation is usually correspondingly greater.…”