Article

Brain MAPS: An automated, accurate and robust brain extraction technique using a template library

Dementia Research Centre, UCL Institute of Neurology, Queen Square, London WC1N 3BG, UK.
NeuroImage (Impact Factor: 6.13). 12/2010; 55(3):1091-108. DOI: 10.1016/j.neuroimage.2010.12.067
Source: PubMed

ABSTRACT Whole brain extraction is an important pre-processing step in neuroimage analysis. Manual or semi-automated brain delineations are labour-intensive and thus not desirable in large studies, meaning that automated techniques are preferable. The accuracy and robustness of automated methods are crucial because human expertise may be required to correct any suboptimal results, which can be very time consuming. We compared the accuracy of four automated brain extraction methods: Brain Extraction Tool (BET), Brain Surface Extractor (BSE), Hybrid Watershed Algorithm (HWA) and a Multi-Atlas Propagation and Segmentation (MAPS) technique we have previously developed for hippocampal segmentation. The four methods were applied to extract whole brains from 682 1.5T and 157 3T T(1)-weighted MR baseline images from the Alzheimer's Disease Neuroimaging Initiative database. Semi-automated brain segmentations with manual editing and checking were used as the gold-standard to compare with the results. The median Jaccard index of MAPS was higher than HWA, BET and BSE in 1.5T and 3T scans (p<0.05, all tests), and the 1st to 99th centile range of the Jaccard index of MAPS was smaller than HWA, BET and BSE in 1.5T and 3T scans ( p<0.05, all tests). HWA and MAPS were found to be best at including all brain tissues (median false negative rate ≤0.010% for 1.5T scans and ≤0.019% for 3T scans, both methods). The median Jaccard index of MAPS were similar in both 1.5T and 3T scans, whereas those of BET, BSE and HWA were higher in 1.5T scans than 3T scans (p<0.05, all tests). We found that the diagnostic group had a small effect on the median Jaccard index of all four methods. In conclusion, MAPS had relatively high accuracy and low variability compared to HWA, BET and BSE in MR scans with and without atrophy.

0 Followers
 · 
131 Views
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents a skull stripping method to segment the brain from MRI human head scans using multi-seeded region growing technique. The proposed method has two stages. In Stage-1, the brain in the middle slice is segmented, the brains in the remaining slices are segmented in Stage-2. In each stage, the proposed method is required to identify the rough brain mask. The fine brain region in the rough brain mask is segmented using multi-seeded region growing approach. The proposed method uses multiple seed points which are selected automatically based on the intensity profile of grey matter (GM), white matter (WM) and cerebrospinal fluid (CSF) of the brain image. The proposed brain segmentation method using multi-seeded region growing (BSMRG) was validated using 100 volumes of T1, T2 and PD-weighted MR brain images obtained from Internet Brain Segmentation Repository (IBSR), LONI and Whole Brain Atlas (WBA). The best Dice (D) value of 0·971 and Jaccard (J) value of 0·944 were recorded by the proposed BSMRG method on IBSR dataset. For LONI dataset, the best values of D = 0·979 and J = 0·960 were obtained for the sagittal oriented images by the proposed method. The performance consistency of the proposed method was tested on the brain images of all types and orientation and have and produced better and stable results than the existing methods Brain Extraction Tool (BET), Brain Surface Extraction (BSE), Watershed Algorithm (WAT), Hybrid Watershed Algorithm (HWA) and Skull Stripping using Graph Cuts (GCUT).
    Imaging Science Journal The 06/2014; 62(5):273-284. DOI:10.1179/1743131X13Y.0000000068 · 0.32 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Multi-atlas segmentation (MAS), first introduced and popularized by the pioneering work of Rohlfing, Brandt, Menzel and Maurer Jr (2004), Klein, Mensh, Ghosh, Tourville and Hirsch (2005), and Heckemann, Hajnal, Aljabar, Rueckert and Hammers (2006), is becoming one of the most widely-used and successful image segmentation techniques in biomedical applications. By manipulating and utilizing the entire dataset of "atlases" (training images that have been previously labeled, e.g., manually by an expert), rather than some model-based average representation, MAS has the flexibility to better capture anatomical variation, thus offering superior segmentation accuracy. This benefit, however, typically comes at a high computational cost. Recent advancements in computer hardware and image processing software have been instrumental in addressing this challenge and facilitated the wide adoption of MAS. Today, MAS has come a long way and the approach includes a wide array of sophisticated algorithms that employ ideas from machine learning, probabilistic modeling, optimization, and computer vision, among other fields. This paper presents a survey of published MAS algorithms and studies that have applied these methods to various biomedical problems. In writing this survey, we have three distinct aims. Our primary goal is to document how MAS was originally conceived, later evolved, and now relates to alternative methods. Second, this paper is intended to be a detailed reference of past research activity in MAS, which roughly spans a decade (2003 - June 2014) and entails novel methodological developments and application-specific solutions. Finally, our goal is to also present a perspective on the future of MAS, which, we believe, will be one of the dominant approaches in biomedical image segmentation.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Stable MR acquisition is essential for reliable measurement of brain atrophy in longitudinal studies. One attractive recent advance in MRI is to speed up acquisition using parallel imaging (e.g. reducing volumetric T1-weighted acquisition scan times from around 9 to 5minutes). In some studies, a decision to change to an accelerated acquisition may have been deliberately taken, while in others repeat scans may occasionally be accidentally acquired with an accelerated acquisition. In ADNI, non-accelerated and accelerated scans were acquired in the same scanning session on each individual. We investigated the impact on brain atrophy as measured by k-means normalised boundary shift integral (KN-BSI) and deformation-based morphometry when changing from non-accelerated to accelerated MRI acquisitions over a 12-month interval using scans of 422 subjects from ADNI. KN-BSIs were calculated using both a non-accelerated baseline scan and non-accelerated 12-month scans (i.e. consistent acquisition), and a non-accelerated baseline scan and an accelerated 12-month scan (i.e. changed acquisition). Fluid-based non-rigid registration was also performed on those scans to estimate the brain atrophy rate. We found that the effect on KN-BSI and fluid-based non-rigid registration depended on the scanner manufacturer. For KN-BSI, in Philips and Siemens scanners, the change had very little impact on the measured atrophy rate (increase of 0.051% in Philips and -0.035% in Siemens from consistent acquisition to changed acquisition), whereas, in GE, the change caused a mean reduction of 0.65% in the brain atrophy rate. This is likely due to the difference in tissue contrast between grey matter and cerebrospinal fluid in the non-accelerated and accelerated scans in GE, which uses IR-FSPGR instead of MP-RAGE. For fluid-based non-rigid registration, the change caused a mean increase of 0.29% in the brain atrophy rate in the changed acquisition compared to consistent acquisition in Philips, whereas in GE and Siemens, the change had less impact on the mean atrophy rate (increase of 0.18% in GE and 0.049% in Siemens). Moving from non-accelerated baseline scans to accelerated scans for follow-up may have surprisingly little effect on computed atrophy rates depending on the exact sequence details and the scanner manufacturer; even accidentally inconsistent scans of this nature may still be useful. Copyright © 2014 Elsevier Inc. All rights reserved.
    NeuroImage 12/2014; 107. DOI:10.1016/j.neuroimage.2014.11.049 · 6.13 Impact Factor