BrainWeb Dataset Research Articles

Adaptive multiresolution non-local means filter for three-dimensional magnetic resonance image denoising

In this study, an adaptive multiresolution version of the blockwise non-local (NL)-means filter is presented for three-dimensional (3D) magnetic resonance (MR) images. On the basis of an adaptive soft wavelet coefficient mixing, the proposed filter implicitly adapts the amount of denoising according to the spatial and frequency information contained in the image. Two versions of the filter are described for Gaussian and Rician noise. Quantitative validation was carried out on BrainWeb datasets by using several quality metrics. The results show that the proposed multiresolution filter obtained competitive performance compared with recently proposed Rician NL-means filters. Finally, qualitative experiments on anatomical and diffusion-weighted MR images show that the proposed filter efficiently removes noise while preserving fine structures in classical and very noisy cases. The impact of the proposed denoising method on fibre tracking is also presented on a HARDI dataset.

Skull‐stripping method for brain MRI using a 3D level set with a speedup operator

To extract the brain region from brain magnetic resonance (MR) images using a fast 3D level set method and a refinement process. The proposed method introduces a speedup operator to the conventional 3D level set method in order to accelerate the level set evolution. While the processing time for brain extraction is reduced by the speedup operator, the accuracy of brain extraction is also improved by adopting a refinement process. The speedup operator yielded a 75% reduction in the total iteration numbers for the synthesized volume. The proposed method was applied to several datasets and compared with previous methods, ie, BrainVisa, BET, and FreeSurfer. The proposed method provided a Jaccard index of 0.971 ± 0.0114 for the BrainWeb dataset, 0.864 ± 0.035 for the IBSR dataset, and 0.9414 ± 0.0517 for a self-produced dataset acquired with a 3T MRI system. Utilizing a speedup operator, the proposed method reduced the evolution time. Robust and accurate results for various datasets were obtained in experiments.

An Open Source Multivariate Framework for n-Tissue Segmentation with Evaluation on Public Data

We introduce Atropos, an ITK-based multivariate n-class open source segmentation algorithm distributed with ANTs ( http://www.picsl.upenn.edu/ANTs). The Bayesian formulation of the segmentation problem is solved using the Expectation Maximization (EM) algorithm with the modeling of the class intensities based on either parametric or non-parametric finite mixtures. Atropos is capable of incorporating spatial prior probability maps (sparse), prior label maps and/or Markov Random Field (MRF) modeling. Atropos has also been efficiently implemented to handle large quantities of possible labelings (in the experimental section, we use up to 69 classes) with a minimal memory footprint. This work describes the technical and implementation aspects of Atropos and evaluates its performance on two different ground-truth datasets. First, we use the BrainWeb dataset from Montreal Neurological Institute to evaluate three-tissue segmentation performance via (1) K-means segmentation without use of template data; (2) MRF segmentation with initialization by prior probability maps derived from a group template; (3) Prior-based segmentation with use of spatial prior probability maps derived from a group template. We also evaluate Atropos performance by using spatial priors to drive a 69-class EM segmentation problem derived from the Hammers atlas from University College London. These evaluation studies, combined with illustrative examples that exercise Atropos options, demonstrate both performance and wide applicability of this new platform-independent open source segmentation tool.

Improved tensor scale computation with application to medical image interpolation

Tensor scale (t-scale) is a parametric representation of local structure morphology that simultaneously describes its orientation, shape and isotropic scale. At any image location, t-scale represents the largest ellipse (an ellipsoid in three dimensions) centered at that location and contained in the same homogeneous region. Here, we present an improved algorithm for t-scale computation and study its application to image interpolation. Specifically, the t-scale computation algorithm is improved by: (1) enhancing the accuracy of identifying local structure boundary and (2) combining both algebraic and geometric approaches in ellipse fitting. In the context of interpolation, a closed form solution is presented to determine the interpolation line at each image location in a gray level image using t-scale information of adjacent slices. At each location on an image slice, the method derives normal vector from its t-scale that yields trans-orientation of the local structure and points to the closest edge point. Normal vectors at the matching two-dimensional locations on two adjacent slices are used to compute the interpolation line using a closed form equation. The method has been applied to BrainWeb data sets and to several other images from clinical applications and its accuracy and response to noise and other image-degrading factors have been examined and compared with those of current state-of-the-art interpolation methods. Experimental results have established the superiority of the new t-scale based interpolation method as compared to existing interpolation algorithms. Also, a quantitative analysis based on the paired t-test of residual errors has ascertained that the improvements observed using the t-scale based interpolation are statistically significant.

Automatic Identification of the Reference System Based on the Fourth Ventricular Landmarks in T1-weighted MR Images

The reference system based on the fourth ventricular landmarks (including the fastigial point and ventricular floor plane) is used in medical image analysis of the brain stem. The objective of this study was to develop a rapid, robust, and accurate method for the automatic identification of this reference system on T1-weighted magnetic resonance images. The fully automated method developed in this study consisted of four stages: preprocessing of the data set, expectation-maximization algorithm-based extraction of the fourth ventricle in the region of interest, a coarse-to-fine strategy for identifying the fastigial point, and localization of the base point. The method was evaluated on 27 Brain Web data sets qualitatively and 18 Internet Brain Segmentation Repository data sets and 30 clinical scans quantitatively. The results of qualitative evaluation indicated that the method was robust to rotation, landmark variation, noise, and inhomogeneity. The results of quantitative evaluation indicated that the method was able to identify the reference system with an accuracy of 0.7 +/- 0.2 mm for the fastigial point and 1.1 +/- 0.3 mm for the base point. It took <6 seconds for the method to identify the related landmarks on a personal computer with an Intel Core 2 6300 processor and 2 GB of random-access memory. The proposed method for the automatic identification of the reference system based on the fourth ventricular landmarks was shown to be rapid, robust, and accurate. The method has potentially utility in image registration and computer-aided surgery.

Image filtering via generalized scale.

In medical imaging, low signal-to-noise ratio (SNR) and/or contrast-to-noise ratio (CNR) often cause many image processing algorithms to perform poorly. Postacquisition image filtering is an important off-line image processing approach widely employed to enhance the SNR and CNR. A major drawback of many filtering techniques is image degradation by diffusing/blurring edges and/or fine structures. In this paper, we introduce a scale-based filtering method that employs scale-dependent diffusion conductance to perform filtering. This approach utilizes novel object scale information via a concept called generalized scale, which imposes no shape, size, or anisotropic constraints unlike previously published ball scale-based filtering strategies. The object scale allows us to better control the filtering process by constraining smoothing in regions with fine details and in the vicinity of boundaries while permitting effective smoothing in the interior of homogeneous regions. A new quantitative evaluation strategy that captures the SNR to CNR trade-off behavior of filtering methods is presented. The evaluations based on the Brainweb data sets show superior performance of generalized scale-based diffusive filtering over two existing methods, namely, ball scale-based and nonlinear complex diffusion processes. Qualitative experiments based on both phantom and patient magnetic resonance images demonstrate that the generalized scale-based approach leads to better preservation of fine details and edges.

A knowledge-driven algorithm for a rapid and automatic extraction of the human cerebral ventricular system from MR neuroimages

A knowledge-driven algorithm for a rapid, robust, accurate, and automatic extraction of the human cerebral ventricular system from MR neuroimages is proposed. Its novelty is in combination of neuroanatomy, radiological properties, and variability of the ventricular system with image processing techniques. The ventricular system is divided into six 3D regions: bodies and inferior horns of the lateral ventricles, third ventricle, and fourth ventricle. Within each ventricular region, a 2D region of interest (ROI) is defined based on anatomy and variability. Each ventricular region is further subdivided into subregions, and conditions detecting and preventing leakage into the extra-ventricular space are specified for each subregion. The algorithm extracts the ventricular system by (1) processing each ROI (to calculate its local statistics, determine local intensity ranges of cerebrospinal fluid and gray and white matters, set a seed point within the ROI, grow region directionally in 3D, check anti-leakage conditions, and correct growing if leakage occurred) and (2) connecting all unconnected regions grown by relaxing growing conditions. The algorithm was validated qualitatively on 68 and quantitatively on 38 MRI normal and pathological cases (30 clinical, 20 MGH Brain Repository, and 18 MNI BrainWeb data sets). It runs successfully for normal and pathological cases provided that the slice thickness is less than 3.0 mm in axial and less than 2.0 mm in coronal directions, and the data do not have a high inter-slice intensity variability. The algorithm also works satisfactorily in the presence of up to 9% noise and up to 40% RF inhomogeneity for the BrainWeb data. The running time is less than 5 s on a Pentium 4, 2.0 GHz PC. The best overlap metric between the results of a radiology expert and the algorithm is 0.9879 and the worst 0.9527; the mean and standard deviation of the overlap metric are 0.9723 and 0.01087, respectively.