Robust skull stripping using multiple MR image contrasts insensitive to pathology

Abstract

Automatic skull-stripping or brain extraction of magnetic resonance (MR) images is often a fundamental step in many neuroimage processing pipelines. The accuracy of subsequent image processing relies on the accuracy of the skull-stripping. Although many automated stripping methods have been proposed in the past, it is still an active area of research particularly in the context of brain pathology. Most stripping methods are validated on T1-w MR images of normal brains, especially because high resolution T1-w sequences are widely acquired and ground truth manual brain mask segmentations are publicly available for normal brains. However, different MR acquisition protocols can provide complementary information about the brain tissues, which can be exploited for better distinction between brain, cerebrospinal fluid, and unwanted tissues such as skull, dura, marrow, or fat. This is especially true in the presence of pathology, where hemorrhages or other types of lesions can have similar intensities as skull in a T1-w image. In this paper, we propose a sparse patch based Multi-cONtrast brain STRipping method (MONSTR)1, where non-local patch information from one or more atlases, which contain multiple MR sequences and reference delineations of brain masks, are combined to generate a target brain mask. We compared MONSTR with four state-of-the-art, publicly available methods: BEaST, SPECTRE, ROBEX, and OptiBET. We evaluated the performance of these methods on 6 datasets consisting of both healthy subjects and patients with various pathologies. Three datasets (ADNI, MRBrainS, NAMIC) are publicly available, consisting of 44 healthy volunteers and 10 patients with schizophrenia. Other three in-house datasets, comprising 87 subjects in total, consisted of patients with mild to severe traumatic brain injury, brain tumors, and various movement disorders. A combination of T1-w, T2-w were used to skull-strip these datasets. We show significant improvement in stripping over the competing methods on both healthy and pathological brains. We also show that our multi-contrast framework is robust and maintains accurate performance across different types of acquisitions and scanners, even when using normal brains as atlases to strip pathological brains, demonstrating that our algorithm is applicable even when reference segmentations of pathological brains are not available to be used as atlases.

TBI contusion segmentation from MRI using convolutional neural networks

“…FLAIR and T 2 -w images were rigidly registered [16] to the MPRAGE. Then for every subject, the MPRAGE was stripped [17] and the same brainmask was applied to the other contrasts. All of the stripped images were corrected for any intensity inhomogeneity [18].…”

Automatic falx cerebri and tentorium cerebelli segmentation from magnetic resonance images

“…All data was acquired on a Siemens Biograph mMR 3T imaging platform. The five patients’ T1-w MRI first underwent standard neuroimaging preprocessing, which includes inhomogeneity correction, 12 skull stripping, 13 and affine registration to an MNI atlas at 0.8 mm isotropic resolution. To quantify the performance of our proposed method, we use two metrics that measure surface distance between the automatic result and manual delineation: Hausdorff distance (HD) 14 and mean surface distance (MSD).…”

Unsupervised brain lesion segmentation from MRI using a convolutional autoencoder

“…The images were preprocessed using standard preprocessing procedures: Resampling to 0.8×0.8×0.8 mm 3 voxel size, correction for signal non-uniformity 13 in the T1-w and T2-w images (see subsection 2.3 for why FLAIR was omitted), rigid registration to the MNI-ICBM152 template, 14 and skull removal. 15…”