Qualitative and quantitative evaluation of six algorithms for correcting intensity nonuniformity effects

Deep learning based correction of RF field induced inhomogeneities for T2w prostate imaging at 7 T

At ultrahigh field strengths images of the body are hampered by B1 -field inhomogeneities. These present themselves as inhomogeneous signal intensity and contrast, which is regarded as a "bias field" to the ideal image. Current bias field correction methods, such as the N4 algorithm, assume a low frequency bias field, which is not sufficiently valid for T2w images at 7 T. In this work we propose a deep learning based bias field correction method to address this issue for T2w prostate images at 7 T. By combining simulated B1 -field distributions of a multi-transmit setup at 7 T with T2w prostate images at 1.5 T, we generated artificial 7 T images for which the homogeneous counterpart was available. Using these paired data, we trained a neural network to correct the bias field. We predicted either a homogeneous image (t-Image neural network) or the bias field (t-Biasf neural network). In addition, we experimented with the single-channel images of the receive array and the corresponding sum of magnitudes of this array as the input image. Testing was carried out on four datasets: the test split of the synthetic training dataset, volunteer and patient images at 7 T, and patient images at 3 T. For the test split, the performance was evaluated using the structural similarity index measure, Wasserstein distance, and root mean squared error. For all other test data, the features Homogeneity and Energy derived from the gray level co-occurrence matrix (GLCM) were used to quantify the improvement. For each test dataset, the proposed method was compared with the current gold standard: the N4 algorithm. Additionally, a questionnaire was filled out by two clinical experts to assess the homogeneity and contrast preservation of the 7 T datasets. All four proposed neural networks were able to substantially reduce the B1 -field induced inhomogeneities in T2w 7 T prostate images. By visual inspection, the images clearly look more homogeneous, which is confirmed by the increase in Homogeneity and Energy in the GLCM, and the questionnaire scores from two clinical experts. Occasionally, changes in contrast within the prostate were observed, although much less for the t-Biasf network than for the t-Image network. Further, results on the 3 T dataset demonstrate that the proposed learning based approach is on par with the N4 algorithm. The results demonstrate that the trained networks were capable of reducing the B1 -field induced inhomogeneities for prostate imaging at 7 T. The quantitative evaluation showed that all proposed learning based correction techniques outperformed the N4 algorithm. Of the investigated methods, the single-channel t-Biasf neural network proves most reliable for bias field correction.

A segmentation-based method improving the performance of N4 bias field correction on T2weighted MR imaging data of the prostate

Magnetic Resonance (MR) images suffer from spatial inhomogeneity, known as bias field corruption. The N4ITK filter is a state-of-the-art method used for correcting the bias field to optimize MR-based quantification. In this study, a novel approach is presented to quantitatively evaluate the performance of N4 bias field correction for pelvic prostate imaging. An exploratory analysis, regarding the different values of convergence threshold, shrink factor, fitting level, number of iterations and use of mask, is performed to quantify the performance of N4 filter in pelvic MR images. The performance of a total of 240 different N4 configurations is examined using the Full Width at Half Maximum (FWHM) of the segmented periprostatic fat distribution as evaluation metric. Phantom T2weighted images were used to assess the performance of N4 for a uniform test tissue mimicking material, excluding factors such as patient related susceptibility and anatomy heterogeneity. Moreover, 89 and 204 T2weighted patient images from two public datasets acquired by scanners with a combined surface and endorectal coil at 1.5 T and a surface coil at 3 T, respectively, were utilized and corrected with a variable set of N4 parameters. Furthermore, two external public datasets were used to validate the performance of the N4 filter in T2weighted patient images acquired by various scanning conditions with different magnetic field strengths and coils. The results show that the set of N4 parameters, converging to optimal representations of fat in the image, were: convergence threshold 0.001, shrink factor 2, fitting level 6, number of iterations 100 and the use of default mask for prostate images acquired by a combined surface and endorectal coil at both 1.5 T and 3 T. The corresponding optimal N4 configuration for MR prostate images acquired by a surface coil at 1.5 T or 3 T was: convergence threshold 0.001, shrink factor 2, fitting level 5, number of iterations 25 and the use of default mask. Hence, periprostatic fat segmentation can be used to define the optimal settings for achieving T2weighted prostate images free from bias field corruption to provide robust input for further analysis.

vol2Brain: A New Online Pipeline for Whole Brain MRI Analysis

Automatic and reliable quantitative tools for MR brain image analysis are a very valuable resource for both clinical and research environments. In the past few years, this field has experienced many advances with successful techniques based on label fusion and more recently deep learning. However, few of them have been specifically designed to provide a dense anatomical labeling at the multiscale level and to deal with brain anatomical alterations such as white matter lesions (WML). In this work, we present a fully automatic pipeline (vol2Brain) for whole brain segmentation and analysis, which densely labels (N > 100) the brain while being robust to the presence of WML. This new pipeline is an evolution of our previous volBrain pipeline that extends significantly the number of regions that can be analyzed. Our proposed method is based on a fast and multiscale multi-atlas label fusion technology with systematic error correction able to provide accurate volumetric information in a few minutes. We have deployed our new pipeline within our platform volBrain (www.volbrain.upv.es), which has been already demonstrated to be an efficient and effective way to share our technology with the users worldwide.

Recommendations of Choice of Head Coil and Prescan Normalize Filter Depend on Region of Interest and Task

The performance of MRI head coils together with the influence of the prescan normalize filter in different brain regions was evaluated. Functional and structural data were recorded from 26 participants performing motor, auditory, and visual tasks in different conditions: with the 20- and 64-channel Siemens head/neck coil and the prescan normalize filter turned ON or OFF. Data were analyzed with the MRIQC tool to evaluate data quality differences. The functional data were statistically evaluated by comparison of the β estimates and the time-course signal-to-noise ratio (tSNR) in four regions of interest, i.e., the auditory, visual, and motor cortices and the thalamus. The MRIQC tool indicated a better data quality for both functional and structural data with the prescan normalize filter, with an advantage for the 20-channel head coil in functional data and an advantage for the 64-channel head coil in structural measurements. Nevertheless, recommendations for the functional data regarding choice of head coils and prescan normalize filter depend on the brain regions of interest. Higher β estimates and tSNR values occurred in the auditory cortex and thalamus with the prescan normalize filter, whereas the contrary was true for the visual and motor cortices. Due to higher β estimates in the visual cortex in the 64-channel head coil, this head coil is recommended for studies investigating the visual cortex. For most of the research questions, the 20-channel head coil is better suited for functional experiments, with the prescan normalize filter, especially when investigating deep brain areas. For anatomical studies, the 64-channel head coil seemed to be the better choice.

Safety, Efficacy, and Feasibility of Intranasal Insulin for the Treatment of Mild Cognitive Impairment and Alzheimer Disease Dementia: A Randomized Clinical Trial

Importance

Insulin modulates aspects of brain function relevant to Alzheimer disease and can be delivered to the brain using intranasal devices. To date, the use of intranasal insulin to treat persons with mild cognitive impairment and Alzheimer's disease dementia remains to be examined in a multi-site trial.

Objective

To examine the feasibility, safety, and efficacy of intranasal insulin for the treatment of persons with mild cognitive impairment and Alzheimer disease dementia in a phase 2/3 multisite clinical trial.

Design, Setting, and Participants

A randomized (1:1) double-blind clinical trial was conducted between 2014 and 2018. Participants received 40 IU of insulin or placebo for 12 months during the blinded phase, which was followed by a 6-month open-label extension phase. The clinical trial was conducted at 27 sites of the Alzheimer's Therapeutic Research Institute. A total of 432 adults were screened, and 144 adults were excluded. Inclusion criteria included adults aged 55 to 85 years with a diagnosis of amnestic mild cognitive impairment or Alzheimer disease (based on National Institute on Aging-Alzheimer Association criteria), a score of 20 or higher on the Mini-Mental State Examination, a clinical dementia rating of 0.5 or 1.0, and a delayed logical memory score within a specified range. A total of 289 participants were randomized. Among the first 49 participants, the first device (device 1) used to administer intranasal insulin treatment had inconsistent reliability. A new device (device 2) was used for the remaining 240 participants, who were designated the primary intention-to-treat population. Data were analyzed from August 2018 to March 2019.

Interventions

Participants received 40 IU of insulin (Humulin-RU-100; Lilly) or placebo (diluent) daily for 12 months (blinded phase) followed by a 6-month open-label extension phase. Insulin was administered with 2 intranasal delivery devices.

Main Outcomes and Measures

The primary outcome (mean score change on the Alzheimer Disease Assessment Scale-cognitive subscale 12) was evaluated at 3-month intervals. Secondary clinical outcomes were assessed at 6-month intervals. Cerebrospinal fluid collection and magnetic resonance imaging scans occurred at baseline and 12 months.

Results

A total of 289 participants (155 men [54.6%]; mean [SD] age, 70.9 [7.1] years) were randomized. Of those, 260 participants completed the blinded phase, and 240 participants completed the open-label extension phase. For the first 49 participants, the first device used to administer treatment had inconsistent reliability. A second device was used for the remaining 240 participants (123 men [51.3%]; mean [SD] age, 70.8 [7.1] years), who were designated the primary intention-to-treat population. No differences were observed between treatment arms for the primary outcome (mean score change on ADAS-cog-12 from baseline to month 12) in the device 2 ITT cohort (0.0258 points; 95% CI, -1.771 to 1.822 points; P = .98) or for the other clinical or cerebrospinal fluid outcomes in the primary (second device) intention-to-treat analysis. No clinically important adverse events were associated with treatment.

Conclusions and Relevance

In this study, no cognitive or functional benefits were observed with intranasal insulin treatment over a 12-month period among the primary intention-to-treat cohort.

Trial Registration

ClinicalTrials.gov Identifier: NCT01767909.

AssemblyNet: A large ensemble of CNNs for 3D whole brain MRI segmentation

Whole brain segmentation of fine-grained structures using deep learning (DL) is a very challenging task since the number of anatomical labels is very high compared to the number of available training images. To address this problem, previous DL methods proposed to use a single convolution neural network (CNN) or few independent CNNs. In this paper, we present a novel ensemble method based on a large number of CNNs processing different overlapping brain areas. Inspired by parliamentary decision-making systems, we propose a framework called AssemblyNet, made of two "assemblies" of U-Nets. Such a parliamentary system is capable of dealing with complex decisions, unseen problem and reaching a relevant consensus. AssemblyNet introduces sharing of knowledge among neighboring U-Nets, an "amendment" procedure made by the second assembly at higher-resolution to refine the decision taken by the first one, and a final decision obtained by majority voting. During our validation, AssemblyNet showed competitive performance compared to state-of-the-art methods such as U-Net, Joint label fusion and SLANT. Moreover, we investigated the scan-rescan consistency and the robustness to disease effects of our method. These experiences demonstrated the reliability of AssemblyNet. Finally, we showed the interest of using semi-supervised learning to improve the performance of our method.

Deep convolutional neural networks for brain image analysis on magnetic resonance imaging: a review

In recent years, deep convolutional neural networks (CNNs) have shown record-shattering performance in a variety of computer vision problems, such as visual object recognition, detection and segmentation. These methods have also been utilised in medical image analysis domain for lesion segmentation, anatomical segmentation and classification. We present an extensive literature review of CNN techniques applied in brain magnetic resonance imaging (MRI) analysis, focusing on the architectures, pre-processing, data-preparation and post-processing strategies available in these works. The aim of this study is three-fold. Our primary goal is to report how different CNN architectures have evolved, discuss state-of-the-art strategies, condense their results obtained using public datasets and examine their pros and cons. Second, this paper is intended to be a detailed reference of the research activity in deep CNN for brain MRI analysis. Finally, we present a perspective on the future of CNNs in which we hint some of the research directions in subsequent years.

The Effect of Single-Scan and Scan-Pair Intensity Inhomogeneity Correction Methods on Repeatability of Voxel-Based Morphometry With Multiple Magnetic Resonance Scanners

OBJECTIVE

The aim of this study was to investigate the effects of single-scan and scan-pair intensity inhomogeneity correction methods on the repeatability of voxel-based morphometry (VBM) using images acquired with multiple magnetic resonance (MR) scanners.

METHODS

Three-dimensional T1-weighed MR images of the brain were obtained from 22 healthy participants using each of 5 MR scanners, yielding 110 images (5 scanners × 22 subjects) in total. Six patterns of intensity inhomogeneity corrections (no correction, single-scan corrections, and scan-pair correction, and their combinations) were applied in the VBM procedure to investigate the effect of the corrections on the repeatability of gray and white matter volume measurements.

RESULTS

Single-scan and scan-pair intensity inhomogeneity corrections significantly reduced the variance in spatially normalized gray and white matter volumes. However, combining the 2 methods did not significantly improve the repeatability when evaluated as whole brain.

CONCLUSIONS

Single-scan and scan-pair intensity inhomogeneity corrections improved the repeatability of gray and white matter volumes obtained by multiple MR scanners and assessed by VBM.

Intensity inhomogeneity correction of SD-OCT data using macular flatspace

Images of the retina acquired using optical coherence tomography (OCT) often suffer from intensity inhomogeneity problems that degrade both the quality of the images and the performance of automated algorithms utilized to measure structural changes. This intensity variation has many causes, including off-axis acquisition, signal attenuation, multi-frame averaging, and vignetting, making it difficult to correct the data in a fundamental way. This paper presents a method for inhomogeneity correction by acting to reduce the variability of intensities within each layer. In particular, the N3 algorithm, which is popular in neuroimage analysis, is adapted to work for OCT data. N3 works by sharpening the intensity histogram, which reduces the variation of intensities within different classes. To apply it here, the data are first converted to a standardized space called macular flat space (MFS). MFS allows the intensities within each layer to be more easily normalized by removing the natural curvature of the retina. N3 is then run on the MFS data using a modified smoothing model, which improves the efficiency of the original algorithm. We show that our method more accurately corrects gain fields on synthetic OCT data when compared to running N3 on non-flattened data. It also reduces the overall variability of the intensities within each layer, without sacrificing contrast between layers, and improves the performance of registration between OCT images.

Adverse effects of metallic artifacts on voxel-wise analysis and tract-based spatial statistics in diffusion tensor imaging

Background Diffusion tensor imaging (DTI) is a magnetic resonance imaging (MRI) technique that reflects the Brownian motion of water molecules constrained within brain tissue. Fractional anisotropy (FA) is one of the most commonly measured DTI parameters, and can be applied to quantitative analysis of white matter as tract-based spatial statistics (TBSS) and voxel-wise analysis. Purpose To show an association between metallic implants and the results of statistical analysis (voxel-wise group comparison and TBSS) for fractional anisotropy (FA) mapping, in DTI of healthy adults. Material and Methods Sixteen healthy volunteers were scanned with 3-Tesla MRI. A magnetic keeper type of dental implant was used as the metallic implant. DTI was acquired three times in each participant: (i) without a magnetic keeper (FAnon1); (ii) with a magnetic keeper (FAimp); and (iii) without a magnetic keeper (FAnon2) as reproducibility of FAnon1. Group comparisons with paired t-test were performed as FAnon1 vs. FAnon2, and as FAnon1 vs. FAimp. Results Regions of significantly reduced and increased local FA values were revealed by voxel-wise group comparison analysis (a P value of less than 0.05, corrected with family-wise error), but not by TBSS. Conclusion Metallic implants existing outside the field of view produce artifacts that affect the statistical analysis (voxel-wise group comparisons) for FA mapping. When statistical analysis for FA mapping is conducted by researchers, it is important to pay attention to any dental implants present in the mouths of the participants.

Comparative evaluation of performance measures for shading correction in time-lapse fluorescence microscopy

Time-lapse fluorescence microscopy is a valuable technology in cell biology, but it suffers from the inherent problem of intensity inhomogeneity due to uneven illumination or camera nonlinearity, known as shading artefacts. This will lead to inaccurate estimates of single-cell features such as average and total intensity. Numerous shading correction methods have been proposed to remove this effect. In order to compare the performance of different methods, many quantitative performance measures have been developed. However, there is little discussion about which performance measure should be generally applied for evaluation on real data, where the ground truth is absent. In this paper, the state-of-the-art shading correction methods and performance evaluation methods are reviewed. We implement 10 popular shading correction methods on two artificial datasets and four real ones. In order to make an objective comparison between those methods, we employ a number of quantitative performance measures. Extensive validation demonstrates that the coefficient of joint variation (CJV) is the most applicable measure in time-lapse fluorescence images. Based on this measure, we have proposed a novel shading correction method that performs better compared to well-established methods for a range of real data tested.

Quantitative Evaluation of Intensity Inhomogeneity Correction Methods for Structural MR Brain Images

The correction of intensity non-uniformity (INU) in magnetic resonance (MR) images is extremely important to ensure both within-subject and across-subject reliability. Here we tackled the problem of objectively comparing INU correction techniques for T1-weighted images, which are the most commonly used in structural brain imaging. We focused our investigations on the methods integrated in widely used software packages for MR data analysis: FreeSurfer, BrainVoyager, SPM and FSL. We used simulated data to assess the INU fields reconstructed by those methods for controlled inhomogeneity magnitudes and noise levels. For each method, we evaluated a wide range of input parameters and defined an enhanced configuration associated with best reconstruction performance. By comparing enhanced and default configurations, we found that the former often provide much more accurate results. Accordingly, we used enhanced configurations for a more objective comparison between methods. For different levels of INU magnitude and noise, SPM and FSL, which integrate INU correction with brain segmentation, generally outperformed FreeSurfer and BrainVoyager, whose methods are exclusively dedicated to INU correction. Nonetheless, accurate INU field reconstructions can be obtained with FreeSurfer on images with low noise and with BrainVoyager for slow and smooth inhomogeneity profiles. Our study may prove helpful for an accurate selection of the INU correction method to be used based on the characteristics of actual MR data.

Intensity Inhomogeneity Correction of Structural MR Images: A Data-Driven Approach to Define Input Algorithm Parameters

Intensity non-uniformity (INU) in magnetic resonance (MR) imaging is a major issue when conducting analyses of brain structural properties. An inaccurate INU correction may result in qualitative and quantitative misinterpretations. Several INU correction methods exist, whose performance largely depend on the specific parameter settings that need to be chosen by the user. Here we addressed the question of how to select the best input parameters for a specific INU correction algorithm. Our investigation was based on the INU correction algorithm implemented in SPM, but this can be in principle extended to any other algorithm requiring the selection of input parameters. We conducted a comprehensive comparison of indirect metrics for the assessment of INU correction performance, namely the coefficient of variation of white matter (CVWM), the coefficient of variation of gray matter (CVGM), and the coefficient of joint variation between white matter and gray matter (CJV). Using simulated MR data, we observed the CJV to be more accurate than CVWM and CVGM, provided that the noise level in the INU-corrected image was controlled by means of spatial smoothing. Based on the CJV, we developed a data-driven approach for selecting INU correction parameters, which could effectively work on actual MR images. To this end, we implemented an enhanced procedure for the definition of white and gray matter masks, based on which the CJV was calculated. Our approach was validated using actual T1-weighted images collected with 1.5 T, 3 T, and 7 T MR scanners. We found that our procedure can reliably assist the selection of valid INU correction algorithm parameters, thereby contributing to an enhanced inhomogeneity correction in MR images.

A randomized, double-blind, placebo-controlled trial of resveratrol for Alzheimer disease

OBJECTIVE

A randomized, placebo-controlled, double-blind, multicenter 52-week phase 2 trial of resveratrol in individuals with mild to moderate Alzheimer disease (AD) examined its safety and tolerability and effects on biomarker (plasma Aβ40 and Aβ42, CSF Aβ40, Aβ42, tau, and phospho-tau 181) and volumetric MRI outcomes (primary outcomes) and clinical outcomes (secondary outcomes).

METHODS

Participants (n = 119) were randomized to placebo or resveratrol 500 mg orally once daily (with dose escalation by 500-mg increments every 13 weeks, ending with 1,000 mg twice daily). Brain MRI and CSF collection were performed at baseline and after completion of treatment. Detailed pharmacokinetics were performed on a subset (n = 15) at baseline and weeks 13, 26, 39, and 52.

RESULTS

Resveratrol and its major metabolites were measurable in plasma and CSF. The most common adverse events were nausea, diarrhea, and weight loss. CSF Aβ40 and plasma Aβ40 levels declined more in the placebo group than the resveratrol-treated group, resulting in a significant difference at week 52. Brain volume loss was increased by resveratrol treatment compared to placebo.

CONCLUSIONS

Resveratrol was safe and well-tolerated. Resveratrol and its major metabolites penetrated the blood-brain barrier to have CNS effects. Further studies are required to interpret the biomarker changes associated with resveratrol treatment.

CLASSIFICATION OF EVIDENCE

This study provides Class II evidence that for patients with AD resveratrol is safe, well-tolerated, and alters some AD biomarker trajectories. The study is rated Class II because more than 2 primary outcomes were designated.

Functional Magnetic Resonance Imaging Methods

Since its inception in 1992, Functional Magnetic Resonance Imaging (fMRI) has become an indispensible tool for studying cognition in both the healthy and dysfunctional brain. FMRI monitors changes in the oxygenation of brain tissue resulting from altered metabolism consequent to a task-based evoked neural response or from spontaneous fluctuations in neural activity in the absence of conscious mentation (the "resting state"). Task-based studies have revealed neural correlates of a large number of important cognitive processes, while fMRI studies performed in the resting state have demonstrated brain-wide networks that result from brain regions with synchronized, apparently spontaneous activity. In this article, we review the methods used to acquire and analyze fMRI signals.

Graphics Processing Unit-Accelerated Nonrigid Registration of MR Images to CT Images During CT-Guided Percutaneous Liver Tumor Ablations

RATIONALE AND OBJECTIVES

Accuracy and speed are essential for the intraprocedural nonrigid magnetic resonance (MR) to computed tomography (CT) image registration in the assessment of tumor margins during CT-guided liver tumor ablations. Although both accuracy and speed can be improved by limiting the registration to a region of interest (ROI), manual contouring of the ROI prolongs the registration process substantially. To achieve accurate and fast registration without the use of an ROI, we combined a nonrigid registration technique on the basis of volume subdivision with hardware acceleration using a graphics processing unit (GPU). We compared the registration accuracy and processing time of GPU-accelerated volume subdivision-based nonrigid registration technique to the conventional nonrigid B-spline registration technique.

MATERIALS AND METHODS

Fourteen image data sets of preprocedural MR and intraprocedural CT images for percutaneous CT-guided liver tumor ablations were obtained. Each set of images was registered using the GPU-accelerated volume subdivision technique and the B-spline technique. Manual contouring of ROI was used only for the B-spline technique. Registration accuracies (Dice similarity coefficient [DSC] and 95% Hausdorff distance [HD]) and total processing time including contouring of ROIs and computation were compared using a paired Student t test.

RESULTS

Accuracies of the GPU-accelerated registrations and B-spline registrations, respectively, were 88.3 ± 3.7% versus 89.3 ± 4.9% (P = .41) for DSC and 13.1 ± 5.2 versus 11.4 ± 6.3 mm (P = .15) for HD. Total processing time of the GPU-accelerated registration and B-spline registration techniques was 88 ± 14 versus 557 ± 116 seconds (P < .000000002), respectively; there was no significant difference in computation time despite the difference in the complexity of the algorithms (P = .71).

CONCLUSIONS

The GPU-accelerated volume subdivision technique was as accurate as the B-spline technique and required significantly less processing time. The GPU-accelerated volume subdivision technique may enable the implementation of nonrigid registration into routine clinical practice.

Segmentation of brain magnetic resonance images based on multi-atlas likelihood fusion: testing using data with a broad range of anatomical and photometric profiles

We propose a hierarchical pipeline for skull-stripping and segmentation of anatomical structures of interest from T1-weighted images of the human brain. The pipeline is constructed based on a two-level Bayesian parameter estimation algorithm called multi-atlas likelihood fusion (MALF). In MALF, estimation of the parameter of interest is performed via maximum a posteriori estimation using the expectation-maximization (EM) algorithm. The likelihoods of multiple atlases are fused in the E-step while the optimal estimator, a single maximizer of the fused likelihoods, is then obtained in the M-step. There are two stages in the proposed pipeline; first the input T1-weighted image is automatically skull-stripped via a fast MALF, then internal brain structures of interest are automatically extracted using a regular MALF. We assess the performance of each of the two modules in the pipeline based on two sets of images with markedly different anatomical and photometric contrasts; 3T MPRAGE scans of pediatric subjects with developmental disorders vs. 1.5T SPGR scans of elderly subjects with dementia. Evaluation is performed quantitatively using the Dice overlap as well as qualitatively via visual inspections. As a result, we demonstrate subject-level differences in the performance of the proposed pipeline, which may be accounted for by age, diagnosis, or the imaging parameters (particularly the field strength). For the subcortical and ventricular structures of the two datasets, the hierarchical pipeline is capable of producing automated segmentations with Dice overlaps ranging from 0.8 to 0.964 when compared with the gold standard. Comparisons with other representative segmentation algorithms are presented, relative to which the proposed hierarchical pipeline demonstrates comparative or superior accuracy.

Extending semiautomatic ventilation defect analysis for hyperpolarized (129)Xe ventilation MRI

RATIONALE AND OBJECTIVES

Clinical deployment of hyperpolarized (129)Xe magnetic resonance imaging requires accurate quantification and visualization of the ventilation defect percentage (VDP). Here, we improve the robustness of our previous semiautomated analysis method to reduce operator dependence, correct for B1 inhomogeneity and vascular structures, and extend the analysis to display multiple intensity clusters.

MATERIALS AND METHODS

Two segmentation methods were compared-a seeded region-growing method, previously validated by expert reader scoring, and a new linear-binning method that corrects the effects of bias field and vascular structures. The new method removes nearly all operator interventions by rescaling the (129)Xe magnetic resonance images to the 99th percentile of the cumulative distribution and applying fixed thresholds to classify (129)Xe voxels into four clusters: defect, low, medium, and high intensity. The methods were applied to 24 subjects including patients with chronic obstructive pulmonary disease (n = 8), age-matched controls (n = 8), and healthy normal subjects (n = 8).

RESULTS

Linear-binning enabled a faster and more reproducible workflow and permitted analysis of an additional 0.25 ± 0.18 L of lung volume by accounting for vasculature. Like region-growing, linear-binning VDP correlated strongly with reader scoring (R(2) = 0.93, P < .0001), but with less systematic bias. Moreover, linear-binning maps clearly depict regions of low and high intensity that may prove useful for phenotyping subjects with chronic obstructive pulmonary disease.

CONCLUSIONS

Corrected linear-binning provides a robust means to quantify (129)Xe ventilation images yielding VDP values that are indistinguishable from expert reader scores, while exploiting the entire dynamic range to depict multiple image clusters.

A unified Bayesian hierarchical model for MRI tissue classification

Various works have used magnetic resonance imaging (MRI) tissue classification extensively to study a number of neurological and psychiatric disorders. Various noise characteristics and other artifacts make this classification a challenging task. Instead of splitting the procedure into different steps, we extend a previous work to develop a unified Bayesian hierarchical model, which addresses both the partial volume effect and intensity non-uniformity, the two major acquisition artifacts, simultaneously. We adopted a normal mixture model with the means and variances depending on the tissue types of voxels to model the observed intensity values. We modeled the relationship among the components of the index vector of tissue types by a hidden Markov model, which captures the spatial similarity of voxels. Furthermore, we addressed the partial volume effect by construction of a higher resolution image in which each voxel is divided into subvoxels. Finally, We achieved the bias field correction by using a Gaussian Markov random field model with a band precision matrix designed in light of image filtering. Sparse matrix methods and parallel computations based on conditional independence are exploited to improve the speed of the Markov chain Monte Carlo simulation. The unified model provides more accurate tissue classification results for both simulated and real data sets.

A Novel Protocol of Continuous Navigation Guidance for Endoscopic Third Ventriculostomy

BACKGROUND:

Although considered a standard neurosurgical procedure, endoscopic third ventriculostomy (ETV) is associated with a relatively high complication rate that is predominantly related to malpositioning of the trajectory.

OBJECTIVE:

To develop an advanced navigation protocol for ETV, assess its possible benefits over commonly used ETV trajectories, and apply this protocol during surgery.

METHODS:

After development of our advanced protocol, the imaging data of 59 patients who underwent ETV without navigation guidance was transferred to our navigation software. An individualized endoscope trajectory was created according to our protocol in all cases. This trajectory was compared with 2 standard trajectories, especially with regard to the distance to relevant neuronal structures: a trajectory manually measured on preoperative radiological images, as performed in all 59 cases, and a trajectory resulting from a commonly used fixed coronal burr hole. Subsequently, we applied the protocol in 15 ETVs to assess the feasibility and procedural complications.

RESULTS:

Our individualized trajectory resulted in a significantly greater distance to the margins of the foramen of Monro, and the burr hole was located more posteriorly from the coronal suture in comparison with the standard trajectories. The advanced ETV technique was feasible in all 15 procedures, and no major complications occurred in any procedure. In 1 patient, a fornix contusion without clinical correlation was observed.

CONCLUSION:

Our data indicate that the proposed navigation protocol for ETV optimizes the distance of the endoscope to important neuronal structures. Continuous endoscope and puncture device guidance may further add to the safety of this procedure.

Incremental learning with SVM for multimodal classification of prostatic adenocarcinoma

Robust detection of prostatic cancer is a challenge due to the multitude of variants and their representation in MR images. We propose a pattern recognition system with an incremental learning ensemble algorithm using support vector machines (SVM) tackling this problem employing multimodal MR images and a texture-based information strategy. The proposed system integrates anatomic, texture, and functional features. The data set was preprocessed using B-Spline interpolation, bias field correction and intensity standardization. First- and second-order angular independent statistical approaches and rotation invariant local phase quantization (RI-LPQ) were utilized to quantify texture information. An incremental learning ensemble SVM was implemented to suit working conditions in medical applications and to improve effectiveness and robustness of the system. The probability estimation of cancer structures was calculated using SVM and the corresponding optimization was carried out with a heuristic method together with a 3-fold cross-validation methodology. We achieved an average sensitivity of 0.844 ± 0.068 and a specificity of 0.780 ± 0.038, which yielded superior or similar performance to current state of the art using a total database of only 41 slices from twelve patients with histological confirmed information, including cancerous, unhealthy non-cancerous and healthy prostate tissue. Our results show the feasibility of an ensemble SVM being able to learn additional information from new data while preserving previously acquired knowledge and preventing unlearning. The use of texture descriptors provides more salient discriminative patterns than the functional information used. Furthermore, the system improves selection of information, efficiency and robustness of the classification. The generated probability map enables radiologists to have a lower variability in diagnosis, decrease false negative rates and reduce the time to recognize and delineate structures in the prostate.

New phase-based B1 mapping method using two-dimensional spin-echo imaging with hyperbolic secant pulses

PURPOSE

To propose a new phase-based B1-mapping method that exploits phase information created by hyperbolic secant (HS) pulses in conventional 2D spin-echo imaging.

METHODS

In this B1-mapping method, HS pulses are used to accomplish π/2 excitation and π refocusing in standard multislice spin-echo imaging. When setting the ratio of pulse lengths of the π/2 and π HS pulses to 2:1, the spin-echo phase is independent of offset frequency and varies as a function of B1 strength. To eliminate undesired phase accumulations induced by unknown factors other than the B1 strength, two spin-echo images are acquired using HS pulses applied with opposite frequency-sweep directions, and the resulting phase images are subtracted from each other. To demonstrate the performance of the proposed method, phantom and in vivo experiments were performed using a surface coil and a volume coil.

RESULTS

The B1 maps obtained by using the proposed method were in accordance with the B1 maps obtained using previous methods in both phantom and in vivo experiments.

CONCLUSION

The proposed method is easy to implement without any sequence modification, is insensitive to B0 inhomogeneity and chemical shift, and is robust in a reasonably wide range of B1 field strength.

Comparison between brain CT and MRI for voxel-based morphometry of Alzheimer's disease

The voxel-based morphometry (VBM) technique using brain magnetic resonance imaging (MRI) objectively maps gray matter loss on a voxel-by-voxel basis after anatomic standardization. In patients with Alzheimer's disease (AD), reductions of gray matter volume, mainly in the medial temporal structures, have been reported; however, inhomogeneity and geometric distortion of the field intensity hampers the reproducibility of MRI. In the present study, we developed a novel computed tomography (CT)-based VBM method and used this technique to detect volume loss in AD patients as compared with normal controls. The results were compared with MRI-based VBM using the same subjects. Pittsburgh Compound B ((11)C-PIB) positron emission tomography (PET)/CT was performed and two experts in neuro-nuclear medicine judged whether regional amyloid β load was consistent with a diagnosis of AD. Before the injection of (11)C-PIB, high-quality CT scans were obtained using the same PET/CT equipment. MRI was performed within a mean interval of 25.1 ± 8.2 days before the PET/CT scan. Using statistical parametric mapping 8 (SPM8), the extracted gray matter images from CT and MRI were spatially normalized using a gray matter template and smoothed using a Gaussian kernel. Group comparisons were performed using SPM8 between five (11)C-PIB-positive patients with probable AD and seven (11)C-PIB-negative age-matched controls with normal cognition. Gray matter volumes in the bilateral medial temporal areas were reduced in the AD group as compared with the cognitively normal group in both CT-based VBM (in the left; P < 0.0001, cluster size 2776 and in the right; P < 0.0001, cluster size 630) and MRI-based VBM (in the left; P < 0.0001, cluster size 381 and in the right, P < 0.0001, cluster size 421). This newly developed CT-based VBM technique can detect significant atrophy in the entorhinal cortex in probable AD patients as previously reported using MRI-based VBM. However, CT-VBM was more sensitive and revealed larger areas of significant atrophy than MR-VBM.

Diffeomorphic Anatomical Registration Through Exponentiated Lie Algebra provides reduced effect of scanner for cortex volumetry with atlas-based method in healthy subjects

INTRODUCTION

This study aimed to investigate whether the effect of scanner for cortex volumetry with atlas-based method is reduced using Diffeomorphic Anatomical Registration Through Exponentiated Lie Algebra (DARTEL) normalization compared with standard normalization.

METHODS

Three-dimensional T1-weighted magnetic resonance images (3D-T1WIs) of 21 healthy subjects were obtained and evaluated for effect of scanner in cortex volumetry. 3D-T1WIs of the 21 subjects were obtained with five MRI systems. Imaging of each subject was performed on each of five different MRI scanners. We used the Voxel-Based Morphometry 8 tool implemented in Statistical Parametric Mapping 8 and WFU PickAtlas software (Talairach brain atlas theory). The following software default settings were used as bilateral region-of-interest labels: "Frontal Lobe," "Hippocampus," "Occipital Lobe," "Orbital Gyrus," "Parietal Lobe," "Putamen," and "Temporal Lobe."

RESULTS

Effect of scanner for cortex volumetry using the atlas-based method was reduced with DARTEL normalization compared with standard normalization in Frontal Lobe, Occipital Lobe, Orbital Gyrus, Putamen, and Temporal Lobe; was the same in Hippocampus and Parietal Lobe; and showed no increase with DARTEL normalization for any region of interest (ROI).

CONCLUSION

DARTEL normalization reduces the effect of scanner, which is a major problem in multicenter studies.

Influence of signal intensity non-uniformity on brain volumetry using an atlas-based method

Objective

Many studies have reported pre-processing effects for brain volumetry; however, no study has investigated whether non-parametric non-uniform intensity normalization (N3) correction processing results in reduced system dependency when using an atlas-based method. To address this shortcoming, the present study assessed whether N3 correction processing provides reduced system dependency in atlas-based volumetry.

Materials and Methods

Contiguous sagittal T1-weighted images of the brain were obtained from 21 healthy participants, by using five magnetic resonance protocols. After image preprocessing using the Statistical Parametric Mapping 5 software, we measured the structural volume of the segmented images with the WFU-PickAtlas software. We applied six different bias-correction levels (Regularization 10, Regularization 0.0001, Regularization 0, Regularization 10 with N3, Regularization 0.0001 with N3, and Regularization 0 with N3) to each set of images. The structural volume change ratio (%) was defined as the change ratio (%) = (100 × [measured volume - mean volume of five magnetic resonance protocols] / mean volume of five magnetic resonance protocols) for each bias-correction level.

Results

A low change ratio was synonymous with lower system dependency. The results showed that the images with the N3 correction had a lower change ratio compared with those without the N3 correction.

Conclusion

The present study is the first atlas-based volumetry study to show that the precision of atlas-based volumetry improves when using N3-corrected images. Therefore, correction for signal intensity non-uniformity is strongly advised for multi-scanner or multi-site imaging trials.

MRI non-uniformity correction through interleaved bias estimation and B-spline deformation with a template

We propose a template-based method for correcting field inhomogeneity biases in magnetic resonance images (MRI) of the human brain. At each algorithm iteration, the update of a B-spline deformation between an unbiased template image and the subject image is interleaved with estimation of a bias field based on the current template-to-image alignment. The bias field is modeled using a spatially smooth thin-plate spline interpolation based on ratios of local image patch intensity means between the deformed template and subject images. This is used to iteratively correct subject image intensities which are then used to improve the template-to-image deformation. Experiments on synthetic and real data sets of images with and without Alzheimer's disease suggest that the approach may have advantages over the popular N3 technique for modeling bias fields and narrowing intensity ranges of gray matter, white matter, and cerebrospinal fluid. This bias field correction method has the potential to be more accurate than correction schemes based solely on intrinsic image properties or hypothetical image intensity distributions.

A NOVEL FUZZY C-MEANS ALGORITHM AND ITS APPLICATION

The accurate and effective algorithm for segmenting image is very useful in many fields, especially in medical images. In this paper we introduced a novel method that focuses on segmenting the brain MR Image that is important for neural diseases. Because of many noises embedded in the acquiring procedure, such as eddy currents, susceptibility artifacts, rigid body motion and intensity inhomogeneity, segmenting the brain MR image is a difficult work. In this algorithm, we overcame the inhomogeneity shortage, by modifying the objective function by compensating its immediate neighborhood effect using Gaussian smooth method for decreasing the influence of the inhomogeneity and increasing the segmenting accuracy. Using simulate image and clinical MRI data, experiments show that our proposed algorithm is effective.

Bias field inconsistency correction of motion-scattered multislice MRI for improved 3D image reconstruction

A common solution to clinical MR imaging in the presence of large anatomical motion is to use fast multislice 2D studies to reduce slice acquisition time and provide clinically usable slice data. Recently, techniques have been developed which retrospectively correct large scale 3D motion between individual slices allowing the formation of a geometrically correct 3D volume from the multiple slice stacks. One challenge, however, in the final reconstruction process is the possibility of varying intensity bias in the slice data, typically due to the motion of the anatomy relative to imaging coils. As a result, slices which cover the same region of anatomy at different times may exhibit different sensitivity. This bias field inconsistency can induce artifacts in the final 3D reconstruction that can impact both clinical interpretation of key tissue boundaries and the automated analysis of the data. Here we describe a framework to estimate and correct the bias field inconsistency in each slice collectively across all motion corrupted image slices. Experiments using synthetic and clinical data show that the proposed method reduces intensity variability in tissues and improves the distinction between key tissue types.

Mapping fetal brain development in utero using magnetic resonance imaging: the Big Bang of brain mapping

The development of tools to construct and investigate probabilistic maps of the adult human brain from magnetic resonance imaging (MRI) has led to advances in both basic neuroscience and clinical diagnosis. These tools are increasingly being applied to brain development in adolescence and childhood, and even to neonatal and premature neonatal imaging. Even earlier in development, parallel advances in clinical fetal MRI have led to its growing use as a tool in challenging medical conditions. This has motivated new engineering developments encompassing optimal fast MRI scans and techniques derived from computer vision, the combination of which allows full 3D imaging of the moving fetal brain in utero without sedation. These promise to provide a new and unprecedented window into early human brain growth. This article reviews the developments that have led us to this point, examines the current state of the art in the fields of fast fetal imaging and motion correction, and describes the tools to analyze dynamically changing fetal brain structure. New methods to deal with developmental tissue segmentation and the construction of spatiotemporal atlases are examined, together with techniques to map fetal brain growth patterns.

Brain MRI segmentation with multiphase minimal partitioning: a comparative study

This paper presents the implementation and quantitative evaluation of a multiphase three-dimensional deformable model in a level set framework for automated segmentation of brain MRIs. The segmentation algorithm performs an optimal partitioning of three-dimensional data based on homogeneity measures that naturally evolves to the extraction of different tissue types in the brain. Random seed initialization was used to minimize the sensitivity of the method to initial conditions while avoiding the need for a priori information. This random initialization ensures robustness of the method with respect to the initialization and the minimization set up. Postprocessing corrections with morphological operators were applied to refine the details of the global segmentation method. A clinical study was performed on a database of 10 adult brain MRI volumes to compare the level set segmentation to three other methods: “idealized” intensity thresholding, fuzzy connectedness, and an expectation maximization classification using hidden Markov random fields. Quantitative evaluation of segmentation accuracy was performed with comparison to manual segmentation computing true positive and false positive volume fractions. A statistical comparison of the segmentation methods was performed through a Wilcoxon analysis of these error rates and results showed very high quality and stability of the multiphase three-dimensional level set method.

Focal liver lesions at 3.0 T: lesion detectability and image quality with T2-weighted imaging by using conventional and dual-source parallel radiofrequency transmission

PURPOSE

To prospectively compare T2-weighted single-shot turbo spin-echo (TSE) sequences performed with parallel and conventional radiofrequency (RF) transmission at 3.0 T for liver lesion detection, image quality, lesion conspicuity, and lesion contrast.

MATERIALS AND METHODS

After written informed consent and institutional review board approval, 52 consecutive patients (32 men, 20 women; mean age, 56.6 years ± 13.7 [standard deviation]) underwent routine magnetic resonance (MR) imaging with a clinical 3.0-T unit. Two independent readers reviewed images acquired with conventional and dual-source parallel RF transmission for detection of focal liver lesions, with separate reading of a third radiologist, including all available imaging findings, clinical history, and histopathologic findings, as reference. Image quality and lesion conspicuity were rated on five- and three-point evaluation scales, respectively. Contrast ratios between focal liver lesions and adjacent liver parenchyma were calculated. Significance was determined by using nonparametric Wilcoxon signed-rank and marginal homogeneity tests.

RESULTS

With the reference standard, 106 index lesions were identified in 22 patients. Detection rate significantly improved from 87% (92 of 106) to 97% (103 of 106) (reader 1) and from 85% (90 of 106) to 96% (102 of 106) (reader 2) with parallel RF transmission (reader 1, P = .0078; reader 2, P = .002). Quality of parallel RF transmission images was assigned scores significantly higher, compared with quality of conventional RF transmission images (mean for reader 1, 2.88 ± 0.73 vs 4.04 ± 0.44; mean for reader 2, 2.81 ± 0.72 vs 4.04 ± 0.39; P < .0001 for both). Lesion conspicuity scores were significantly higher on parallel RF transmission images, compared with conventional RF transmission images (mean for reader 1, 2.02 ± 0.64 vs 2.92 ± 0.27; mean for reader 2, 2.06 ± 0.67 vs 2.90 ± 0.30; P < .0001 for both). Contrast ratios were significantly higher with parallel RF transmission (P < .05).

CONCLUSION

Compared with conventional RF transmission, parallel RF transmission significantly improved liver lesion detection rate, image quality, lesion conspicuity, and lesion contrast.

SUPPLEMENTAL MATERIAL

http://radiology.rsna.org/lookup/suppl/doi:10.1148/radiol.11101429/-/DC1.

N4ITK: improved N3 bias correction

A variant of the popular nonparametric nonuniform intensity normalization (N3) algorithm is proposed for bias field correction. Given the superb performance of N3 and its public availability, it has been the subject of several evaluation studies. These studies have demonstrated the importance of certain parameters associated with the B-spline least-squares fitting. We propose the substitution of a recently developed fast and robust B-spline approximation routine and a modified hierarchical optimization scheme for improved bias field correction over the original N3 algorithm. Similar to the N3 algorithm, we also make the source code, testing, and technical documentation of our contribution, which we denote as "N4ITK," available to the public through the Insight Toolkit of the National Institutes of Health. Performance assessment is demonstrated using simulated data from the publicly available Brainweb database, hyperpolarized (3)He lung image data, and 9.4T postmortem hippocampus data.

A time-saving and facilitating approach for segmentation of anatomically defined cortical regions: MRI volumetry

In this study, we present an accurate, reliable, robust, and time-efficient technique for a semi-automatic segmentation of neuroanatomically defined cortical structures in Magnetic Resonance Imaging (MRI) scans. It involves manual drawing of the border of a region of interest (ROI), supported by three-dimensional (3D) visualization techniques (rendering), and a subsequent automatic tracing of the gray matter voxels inside the ROI by means of an automatic tissue classifier. The approach has been evaluated on a set of MRI scans of 75 participants selected from the Maastricht Aging Study (MAAS) and applied to cortical brain structures for both the left and right hemispheres, viz., the inferior prefrontal cortex (PFC); the orbital PFC; the dorsolateral PFC; the anterior cingulate cortex; and the posterior cingulate cortex. The use of a 3D surface-rendered brain can be rotated in any direction was invaluable in identifying anatomical landmarks on the basis of gyral and sulcal topography. This resulted in a high accuracy (anatomical correctness) and reliability: the intra-rater intra-class correlation coefficient (ICC) was between 0.96 and 0.99. Furthermore, the obtained time savings were substantial, i.e., up to a factor of 7.5 compared with fully manual segmentations.

Cerebellum, language, and cognition in autism and specific language impairment

We performed cerebellum segmentation and parcellation on magnetic resonance images from right-handed boys, aged 6-13 years, including 22 boys with autism [16 with language impairment (ALI)], 9 boys with Specific Language Impairment (SLI), and 11 normal controls. Language-impaired groups had reversed asymmetry relative to unimpaired groups in posterior-lateral cerebellar lobule VIIIA (right side larger in unimpaired groups, left side larger in ALI and SLI), contralateral to previous findings in inferior frontal cortex language areas. Lobule VIIA Crus I was smaller in SLI than in ALI. Vermis volume, particularly anterior I-V, was decreased in language-impaired groups. Language performance test scores correlated with lobule VIIIA asymmetry and with anterior vermis volume. These findings suggest ALI and SLI subjects show abnormalities in neurodevelopment of fronto-corticocerebellar circuits that manage motor control and the processing of language, cognition, working memory, and attention.

A robust method to estimate the intracranial volume across MRI field strengths (1.5T and 3T)

As population-based studies may obtain images from scanners with different field strengths, a method to normalize regional brain volumes according to intracranial volume (ICV) independent of field strength is needed. We found systematic differences in ICV estimation, tested in a cohort of healthy subjects (n = 5) that had been imaged using 1.5T and 3T scanners, and confirmed in two independent cohorts. This was related to systematic differences in the intensity of cerebrospinal fluid (CSF), with higher intensities for CSF located in the ventricles compared with CSF in the cisterns, at 3T versus 1.5T, which could not be removed with three different applied bias correction algorithms. We developed a method based on tissue probability maps in MNI (Montreal Neurological Institute) space and reverse normalization (reverse brain mask, RBM) and validated it against manual ICV measurements. We also compared it with alternative automated ICV estimation methods based on Statistical Parametric Mapping (SPM5) and Brain Extraction Tool (FSL). The proposed RBM method was equivalent to manual ICV normalization with a high intraclass correlation coefficient (ICC = 0.99) and reliable across different field strengths. RBM achieved the best combination of precision and reliability in a group of healthy subjects, a group of patients with Alzheimer's disease (AD) and mild cognitive impairment (MCI) and can be used as a common normalization framework.

Improvement of brain segmentation accuracy by optimizing non-uniformity correction using N3

Smoothly varying and multiplicative intensity variations within MR images that are artifactual, can reduce the accuracy of automated brain segmentation. Fortunately, these can be corrected. Among existing correction approaches, the nonparametric non-uniformity intensity normalization method N3 (Sled, J.G., Zijdenbos, A.P., Evans, A.C., 1998. Nonparametric method for automatic correction of intensity nonuniformity in MRI data. IEEE Trans. Med. Imag. 17, 87-97.) is one of the most frequently used. However, at least one recent study (Boyes, R.G., Gunter, J.L., Frost, C., Janke, A.L., Yeatman, T., Hill, D.L.G., Bernstein, M.A., Thompson, P.M., Weiner, M.W., Schuff, N., Alexander, G.E., Killiany, R.J., DeCarli, C., Jack, C.R., Fox, N.C., 2008. Intensity non-uniformity correction using N3 on 3-T scanners with multichannel phased array coils. NeuroImage 39, 1752-1762.) suggests that its performance on 3 T scanners with multichannel phased-array receiver coils can be improved by optimizing a parameter that controls the smoothness of the estimated bias field. The present study not only confirms this finding, but additionally demonstrates the benefit of reducing the relevant parameter values to 30-50 mm (default value is 200 mm), on white matter surface estimation as well as the measurement of cortical and subcortical structures using FreeSurfer (Martinos Imaging Centre, Boston, MA). This finding can help enhance precision in studies where estimation of cerebral cortex thickness is critical for making inferences.

MR image segmentation using a power transformation approach

This study proposes a segmentation method for brain MR images using a distribution transformation approach. The method extends traditional Gaussian mixtures expectation-maximization segmentation to a power transformed version of mixed intensity distributions, which includes Gaussian mixtures as a special case. As MR intensities tend to exhibit non-Gaussianity due to partial volume effects, the proposed method is designed to fit non-Gaussian tissue intensity distributions. One advantage of the method is that it is intuitively appealing and computationally simple. To avoid performance degradation caused by intensity inhomogeneity, different methods for correcting bias fields were applied prior to image segmentation, and their correction effects on the segmentation results were examined in the empirical study. The partitions of brain tissues (i.e., gray and white matter) resulting from the method were validated and evaluated against manual segmentation results based on 38 real T1-weighted image volumes from the internet brain segmentation repository, and 18 simulated image volumes from BrainWeb. The Jaccard and Dice similarity indexes were computed to evaluate the performance of the proposed approach relative to the expert segmentations. Empirical results suggested that the proposed segmentation method yielded higher similarity measures for both gray matter and white matter as compared with those based on the traditional segmentation using the Gaussian mixtures approach.

Online resource for validation of brain segmentation methods

One key issue that must be addressed during the development of image segmentation algorithms is the accuracy of the results they produce. Algorithm developers require this so they can see where methods need to be improved and see how new developments compare with existing ones. Users of algorithms also need to understand the characteristics of algorithms when they select and apply them to their neuroimaging analysis applications. Many metrics have been proposed to characterize error and success rates in segmentation, and several datasets have also been made public for evaluation. Still, the methodologies used in analyzing and reporting these results vary from study to study, so even when studies use the same metrics their numerical results may not necessarily be directly comparable. To address this problem, we developed a web-based resource for evaluating the performance of skull-stripping in T1-weighted MRI. The resource provides both the data to be segmented and an online application that performs a validation study on the data. Users may download the test dataset, segment it using whichever method they wish to assess, and upload their segmentation results to the server. The server computes a series of metrics, displays a detailed report of the validation results, and archives these for future browsing and analysis. We applied this framework to the evaluation of 3 popular skull-stripping algorithms--the Brain Extraction Tool [Smith, S.M., 2002. Fast robust automated brain extraction. Hum. Brain Mapp. 17 (3),143-155 (Nov)], the Hybrid Watershed Algorithm [Ségonne, F., Dale, A.M., Busa, E., Glessner, M., Salat, D., Hahn, H.K., Fischl, B., 2004. A hybrid approach to the skull stripping problem in MRI. NeuroImage 22 (3), 1060-1075 (Jul)], and the Brain Surface Extractor [Shattuck, D.W., Sandor-Leahy, S.R., Schaper, K.A., Rottenberg, D.A., Leahy, R.M., 2001. Magnetic resonance image tissue classification using a partial volume model. NeuroImage 13 (5), 856-876 (May) under several different program settings. Our results show that with proper parameter selection, all 3 algorithms can achieve satisfactory skull-stripping on the test data.