Denoising Diffusion MRI via Graph Total Variance in Spatioangular Domain.

Diffusion MRI (DMRI) plays an essential role in diagnosing brain disorders related to white matter abnormalities. However, it suffers from heavy noise, which restricts its quantitative analysis. The total variance (TV) regularization is an effective noise reduction technique that penalizes noise-induced variances. However, existing TV-based denoising methods only focus on the spatial domain, overlooking that DMRI data lives in a combined spatioangular domain. It eventually results in an unsatisfactory noise reduction effect. To resolve this issue, we propose to remove the noise in DMRI using graph total variance (GTV) in the spatioangular domain. Expressly, we first represent the DMRI data using a graph, which encodes the geometric information of sampling points in the spatioangular domain. We then perform effective noise reduction using the powerful GTV regularization, which penalizes the noise-induced variances on the graph. GTV effectively resolves the limitation in existing methods, which only rely on spatial information for removing the noise. Extensive experiments on synthetic and real DMRI data demonstrate that GTV can remove the noise effectively and outperforms state-of-the-art methods.

Texture Analysis in the Assessment of Rectal Cancer: Comparison of T2WI and Diffusion-Weighted Imaging.

Texture analysis (TA) techniques derived from T2-weighted imaging (T2WI) and apparent diffusion coefficient (ADC) maps of rectal cancer can both achieve good diagnosis performance. This study was to compare TA from T2WI and ADC maps between different pathological T and N stages to confirm which TA analysis is better in diagnosis performance. 146 patients were enrolled in this study. Tumor TA was performed on every patient's T2WI and ADC maps, respectively; then, skewness, kurtosis, uniformity, entropy, energy, inertia, and correlation were calculated. Our results demonstrated that those significant different parameters derived from T2WI had better diagnostic performance than those from ADC maps in differentiating pT3b-4 and pN1-2 stage tumors. In particular, the energy derived from T2WI was an optimal parameter for diagnostic efficiency. High-resolution T2WI plays a key point in the local stage of rectal cancer; thus, TA derived from T2WI may be a more useful tool to aid radiologists and surgeons in selecting treatment.

Outcomes assessment in intrahepatic cholangiocarcinoma using qualitative and quantitative imaging features.

BACKGROUND: To assess the performance of imaging features, including radiomics texture features, in predicting histopathologic tumor grade, AJCC stage, and outcomes [time to recurrence (TTR) and overall survival (OS)] in patients with intrahepatic cholangiocarcinoma (ICC). METHODS: Seventy-three patients (26 M/47F, mean age 63y) with pre-operative imaging (CT, n = 37; MRI, n = 21; CT and MRI, n = 15] within 6 months of resection were included in this retrospective study. Qualitative imaging traits were assessed by 2 observers. A 3rd observer measured tumor apparent diffusion coefficient (ADC), enhancement ratios (ERs), and Haralick texture features. Blood biomarkers and imaging features were compared with histopathology (tumor grade and AJCC stage) and outcomes (TTR and OS) using log-rank, generalized Wilcoxon, Cox proportional hazards regression, and Fisher exact tests. RESULTS: Median TTR and OS were 53.9 and 79.7 months. ICC recurred in 64.4% (47/73) of patients and 46.6% (34/73) of patients died. There was fair accuracy for some qualitative imaging features in the prediction of worse tumor grade (maximal AUC of 0.68 for biliary obstruction on MRI, p = 0.032, observer 1) and higher AJCC stage (maximal AUC of 0.73 for biliary obstruction on CT, p = 0.002, observer 2; and AUC of 0.73 for vascular involvement on MRI, p = 0.01, observer 2). Cox proportional hazards regression analysis showed that CA 19-9 [hazard ratio (HR) 2.44/95% confidence interval (CI) 1.31-4.57/p = 0.005)] and tumor size on imaging (HR 1.13/95% CI 1.04-1.22/p = 0.003) were significant predictors of TTR, while CA 19-9 (HR 4.08/95% CI 1.75-9.56, p = 0.001) and presence of metastatic lymph nodes at histopathology (HR 2.86/95% CI 1.35-6.07/p = 0.006) were significant predictors of OS. On multivariable analysis, satellite lesions on CT (HR 2.79/95%CI 1.01-7.15/p = 0.032, observer 2), vascular involvement on MRI (HR 0.10/95% CI 0.01-0.85/p = 0.032, observer 1), and texture feature MRI variance (HR 0.55/95% CI 0.31-0.97, p = 0.040) predicted TTR once adjusted for the independent predictors CA 19-9 and tumor size on imaging. Several qualitative and quantitative features demonstrated associations with TTR, OS, and AJCC stage at univariable analysis (range: HR 0.35-19; p < 0.001-0.045), however none were predictive of OS at multivariable analysis when adjusted for CA 19-9 and metastatic lymph nodes (p > 0.088). CONCLUSIONS: There was reasonable accuracy in predicting tumor grade and higher AJCC stage in ICC utilizing certain qualitative and quantitative imaging traits. Serum CA 19-9, tumor size, presence of metastatic lymph nodes, and qualitative imaging traits of satellite lesions and vascular involvement are predictors of patient outcomes, along with a promising predictive ability of certain quantitative texture features.

An Assessment of the Relationship between Structural and Functional Imaging of Cerebrovascular Disease and Cognition-Related Fibers.

In order to assess the relationship between structural and functional imaging of cerebrovascular disease and cognition-related fibers, this paper chooses a total of 120 patients who underwent cerebral small vessel disease (CSVD) treatment at a designated hospital by this study from June 2013 to June 2018 and divides them into 3 groups according to the random number table method: vascular dementia (VaD) group, vascular cognitive impairment no dementia (VCIND) group, and noncognition impairment (NCI) group with 40 cases of patients in each group. Cognitive function measurement and imaging examination were performed for these 3 groups of patients, and the observation indicators of cognitive state examination (CSE), mental assessment scale (MAS), clock drawing test (CDT), adult intelligence scale (AIS), frontal assessment battery (FAB), verbal fluency test (VFT), trail making test (TMT), cognitive index (CI), white matter lesions (WML), third ventricle width (TVW), and frontal horn index (FHI) were tested, respectively. The results shows that the average scores of CSE, MAS, AIS, and VFT in the VaD and VCIND group are lower than those of the NCI group and the differences are statistically significant (P < 0.05); the average scores of FAB, TMT, and CI in the VaD group are higher than those of the VCIND group and the differences are also statistically significant (P < 0.05); the average scores of FHI and TVW in the VaD group are lower than those of the VCIND and NCI group with statistically significant differences (P < 0.05); the average scores of WML, CDT, and AIS in the VaD group are higher than those of the VCIND and NCI group with statistically significant differences (P < 0.05). Therefore, it is believed that the structural and functional imaging features of cerebrovascular disease are closely related to cognition-related fibers, and the incidence of white matter lesions is closely related to the degree of lesions and cognitive dysfunction of cerebral small vessel disease, in which a major risk factor for cognitive dysfunction in patients with small blood vessels is the severity of white matter lesions; brain imaging and neuropsychiatric function assessment can better understand the relationship between cerebrovascular disease and cognitive impairment. The results of this study provide a reference for the further research studies on the relationship between structural and functional imaging of cerebrovascular disease and cognition-related fibers.

Diagnostic performance of intravoxel incoherent motion diffusion-weighted imaging and dynamic contrast-enhanced MRI for assessment of anal fistula activity.

OBJECTIVE: To evaluate intravoxel incoherent motion (IVIM) diffusion-weighted imaging (DWI) and dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) sequences for quantitative characterization of anal fistula activity. METHODS: This retrospective study was approved by the institutional review board. One hundred and two patients underwent MRI for clinical suspicion of anal fistula. Forty-three patients with demonstrable anal fistulas met the inclusion criteria. Quantitative analysis included measurement of DCE and IVIM parameters. The reference standard was clinical activity based on medical records. Statistical analyses included Bayesian analysis with Markov Chain Monte Carlo, multivariable logistic regression, and receiver operating characteristic analyses. RESULTS: Brevity of enhancement, defined as the time difference between the wash-in and wash-out, was longer in active than inactive fistulas (p = 0.02). Regression coefficients of multivariable logistic regression analysis revealed that brevity of enhancement increased and normalized perfusion area under curve decreased with presence of active fistulas (p = 0.03 and p = 0.04, respectively). By cross-validation, a logistic regression model that included quantitative perfusion parameters (DCE and IVIM) performed significantly better than IVIM only (p < 0.001). Area under the curves for distinguishing patients with active from those with inactive fistulas were 0.669 (95% confidence interval [CI]: 0.500, 0.838) for a model with IVIM only, 0.860 (95% CI: 0.742, 0.977) for a model with IVIM and brevity of enhancement, and 0.921 (95% CI: 0.846, 0.997) for a model with IVIM and all DCE parameters. CONCLUSION: The inclusion of brevity of enhancement measured by DCE-MRI improved assessment of anal fistula activity over IVIM-DWI only.

Measurements of Hepatic Metastasis on MR Imaging:: Assessment of Interobserver and Intersequence Variability.

RATIONALE AND OBJECTIVES: The purpose of the study was to investigate interobserver and intersequence variability in the measurement of hepatic metastasis on magnetic resonance imaging (MRI). MATERIALS AND METHODS: This retrospective study was conducted with an institutional review board-approved waiver of informed consent and was in compliance with the Health Insurance Portability and Accountability Act. We searched medical records at our institution for patients with histologically proven metastases to the liver who had undergone MRI from January 2008 to June 2010. We identified 20 patients with 30 measurable liver lesions. The liver lesions were measured on five different MRI sequences. A presenter radiologist selected and localized all metastatic lesions considered to be measurable according to the Response Evaluation Criteria in Solid Tumors, and these lesions were measured (Eisenhauer et al., 2009) by three radiologists independently. We calculated lesion-wise intraclass correlation coefficients (ICCs) to estimate interobserver and intersequence agreement in lesion diameter measurement. A Bland-Altman plot was used to estimate the limits of agreement between radiologists and MRI sequences. RESULTS: There were 30 metastases, and almost all of which had regular and well-defined margins. Interobserver ICCs were greater than 0.95 for different MRI sequences except for the measurements in apparent diffusion coefficient images. Intersequence ICCs were greater than 0.92. Bland-Altman plots between physicians confirmed that reader measurements were closely tied together, with small differences in means. CONCLUSIONS: MRI can reproducibly measure hepatic metastatic lesions without significant variability among interpreting radiologists or among MRI sequences, and is thus a reliable method for assessing the size of hepatic metastasis.

Evaluation of the European Society of Urogenital Radiology (ESUR) PI-RADS scoring system for assessment of extra-prostatic extension in prostatic carcinoma.

INTRODUCTION: To evaluate extra-prostatic extension (EPE) comparing PI-RADS to non-standardized reporting. MATERIALS AND METHODS: With IRB approval, 145 consecutive patients underwent radical prostatectomy (RP) and multi-parametric (T2W+DWI+DCE) MRI between 2012 and 2013. Eighty patients (66.3% with EPE) were staged without PI-RADS and 65 patients (64.6% with EPE) were staged using a 5-point PI-RADS scoring system. Studies were reported by fellowship-trained radiologists in routine clinical practice. Individual PIRADS scores were assessed using ROC to determine the score which optimized sensitivity/specificity. Diagnostic accuracy for EPE was compared with/without PI-RADS using the McNemar test. Subgroup analysis by radiologist experience was performed using Spearman correlation and chi-square. RESULTS: Area under ROC curve for EPE using PI-RADS was 0.62 and optimal sensitivity/specificity was achieved with PI-RADS score ≥ 3. Compared to non-standardized reporting, sensitivity for EPE improved with PI-RADS (59.5% [49.1-68.2] vs. 24.5% [16.7-31.2]), p=0.01; with no difference in specificity (68.0% [50.5-82.6]) vs. (75.0% [60.1-87.6]), p=0.06. Overall accuracy improved with PI-RADS (62.7% [49.6-73.6] vs. 42.0% [31.7-50.7%]), p=0.006. Diagnostic accuracy was better among experienced radiologists without PI-RADS (p=0.005); however, there was no difference in accuracy by reader experience using PI-RADS (p=0.24). CONCLUSION: The PI-RADS criteria for EPE improves sensitivity without reducing specificity. PI-RADS may reduce differences in accuracy by reader experience.

Assessment of non-Gaussian diffusion with singly and doubly stretched biexponential models of diffusion-weighted MRI (DWI) signal attenuation in prostate tissue.

Non-Gaussian diffusion dynamics was investigated in the two distinct water populations identified by a biexponential model of diffusion in prostate tissue. Diffusion-weighted MRI (DWI) signal attenuation was measured ex vivo in two formalin-fixed prostates at 9.4 T with diffusion times Δ = 10, 20 and 40 ms, and b values in the range 0.017-8.2 ms/µm(2) . A conventional biexponential model was compared with models in which either the lower diffusivity component or both of the components of the biexponential were stretched. Models were compared using Akaike's Information Criterion (AIC) and a leave-one-out (LOO) test of model prediction accuracy. The doubly stretched (SS) model had the highest LOO prediction accuracy and lowest AIC (highest information content) in the majority of voxels at Δ = 10 and 20 ms. The lower diffusivity stretching factor (α2 ) of the SS model was consistently lower (range ~0.3-0.9) than the higher diffusivity stretching factor (α1 , range ~0.7-1.1), indicating a high degree of diffusion heterogeneity in the lower diffusivity environment, and nearly Gaussian diffusion in the higher diffusivity environment. Stretched biexponential models demonstrate that, in prostate tissue, the two distinct water populations identified by the simple biexponential model individually exhibit non-Gaussian diffusion dynamics.

Quantitative assessment of diffusional kurtosis anisotropy.

Diffusional kurtosis imaging (DKI) measures the diffusion and kurtosis tensors to quantify restricted, non-Gaussian diffusion that occurs in biological tissue. By estimating the kurtosis tensor, DKI accounts for higher order diffusion dynamics, when compared with diffusion tensor imaging (DTI), and consequently can describe more complex diffusion profiles. Here, we compare several measures of diffusional anisotropy which incorporate information from the kurtosis tensor, including kurtosis fractional anisotropy (KFA) and generalized fractional anisotropy (GFA), with the diffusion tensor-derived fractional anisotropy (FA). KFA and GFA demonstrate a net enhancement relative to FA when multiple white matter fiber bundle orientations are present in both simulated and human data. In addition, KFA shows net enhancement in deep brain structures, such as the thalamus and the lenticular nucleus, where FA indicates low anisotropy. Thus, KFA and GFA provide additional information relative to FA with regard to diffusional anisotropy, and may be particularly advantageous for the assessment of diffusion in complex tissue environments.

Visual and region of interest-based inter-rater agreement in the assessment of the diffusion-weighted imaging- fluid-attenuated inversion recovery mismatch.

BACKGROUND AND PURPOSE: WAKE-UP is a randomized, placebo-controlled MRI-based trial of thrombolysis in wake-up stroke using the mismatch between a lesion's visibility in diffusion-weighted imaging and fluid-attenuated inversion recovery (FLAIR) sequences as its main imaging inclusion criterion. Visual judgment of lesion conspicuity on FLAIR is however methodically limited by moderate inter-rater agreement. We therefore sought to improve rating homogeneity by incorporating quantitative signal intensity measurements. METHODS: One hundred forty-three data sets of patients with acute ischemic stroke were visually rated by 8 raters with respect to WAKE-UP study inclusion and exclusion criteria, and inter-rater agreement was calculated. A subanalysis was performed on 45 cases to determine a threshold value of relative signal intensity (rSI) between the ischemic lesion and contralateral healthy tissue which best corresponded to a visually established verdict of FLAIR positivity. The usefulness of this threshold in improving inter-rater agreement was evaluated in an additional sample of 50 patients. RESULTS: Inter-rater agreement for inclusion into the WAKE-UP trial was 73% with a free-marginal κ of 0.46. A threshold of rSI which best correlated with the visual rating of lesions as FLAIR positive was 1.20. The addition of rSI measurements to visual evaluation did not change the inter-rater agreement. CONCLUSIONS: Introducing a semiquantitative measure for FLAIR rSI did not improve the agreement between individual raters. However, enhancing visual assessment with rSI measurements can provide reassurance to local investigators in cases of uncertainty.

Accelerating fibre orientation estimation from diffusion weighted magnetic resonance imaging using GPUs.

With the performance of central processing units (CPUs) having effectively reached a limit, parallel processing offers an alternative for applications with high computational demands. Modern graphics processing units (GPUs) are massively parallel processors that can execute simultaneously thousands of light-weight processes. In this study, we propose and implement a parallel GPU-based design of a popular method that is used for the analysis of brain magnetic resonance imaging (MRI). More specifically, we are concerned with a model-based approach for extracting tissue structural information from diffusion-weighted (DW) MRI data. DW-MRI offers, through tractography approaches, the only way to study brain structural connectivity, non-invasively and in-vivo. We parallelise the Bayesian inference framework for the ball & stick model, as it is implemented in the tractography toolbox of the popular FSL software package (University of Oxford). For our implementation, we utilise the Compute Unified Device Architecture (CUDA) programming model. We show that the parameter estimation, performed through Markov Chain Monte Carlo (MCMC), is accelerated by at least two orders of magnitude, when comparing a single GPU with the respective sequential single-core CPU version. We also illustrate similar speed-up factors (up to 120x) when comparing a multi-GPU with a multi-CPU implementation.

Non-parametric Bayesian estimation of apparent diffusion coefficient from diffusion-weighted magnetic resonance imaging data.

A promising approach to prostate cancer diagnosis is multi-parametric MRI. One of the key modalities used in multi-parametric MRI is diffusion weighted MRI. Using multiple diffusion weighted MR acquisitions taken with different magnetic gradient strengths, the apparent diffusion coefficient (ADC) is calculated and can be used to identify tumors in the prostate. Current algorithms used to calculate ADC assume a parametric measurement model, but this assumption is not true due to the presence of additional phenomena during the acquisition process. A novel Non-parametric Estimated ADC (NEstA) algorithm is proposed which uses a Monte Carlo strategy to learn the inherent measurement distribution model based on the underlying statistical behavior of the DWI measurements to estimate the ADC values. The proposed algorithm is compared to the results of the commonly used least-squares (LS) estimation algorithm for computing ADC values. Nine test patient cases with visible tumors in the prostate gland were processed using both algorithms and compared visually. It was found that NEstA produced ADC data with reduced artifacts while preserving structure. Quantitatively, Fisher's criterion measuring the separability of the healthy prostate and tumor tissues was computed for the nine patient cases, comparing the NEstA and LS methods. It was found that Fisher's criterion increased with the NEstA method, meaning the separation of classes was more pronounced.

MR imaging assessment of lumbar intervertebral disk degeneration and age-related changes: apparent diffusion coefficient versus T2 quantitation.

BACKGROUND AND PURPOSE: T2 and ADC mappings are 2 quantitative MR imaging tools for assessing IVDD. This study aimed to compare these 2 measures in detecting IVDD and its age-related changes. MATERIALS AND METHODS: Thirty-seven asymptomatic volunteers and 28 patients with back pain or sciatica were examined, and their lumbar disk T2 and ADC maps were quantified via sagittal imaging protocols at 1.5T. For all participants, the Pfirrmann system was used by 2 radiologists for grading disks. T2 and ADC values in the inner portion of disks were measured, and their variances in different grades were analyzed by 1-way ANOVA testing. The ability of T2 and ADC measures to differentiate IVDD grades was compared on the basis of their ROC curves. For asymptomatic subjects, the correlations between age and the 2 MR imaging measures were assessed by the Pearson correlation test. RESULTS: Both T2 and ADC values were found to decrease with the increasing Pfirrmann grades except T2 in grade V. Significant T2 differences were seen among grades I-IV, but not between grades IV and V. There were no significant ADC differences among grades I-III. Moreover, the areas under the ROC curves differed significantly (0.95 and 0.67 for T2 and ADC, respectively). Linear regression analysis revealed that T2 yielded more significant correlation with age (r = -0.77) than ADC did (r = -0.37). CONCLUSIONS: T2 quantitation provides a more sensitive and robust approach for detecting and characterizing the early stage of IVDD and age-related disk changes.

³He lung morphometry technique: accuracy analysis and pulse sequence optimization.

The (3)He lung morphometry technique (Yablonskiy et al., JAP, 2009), based on MRI measurements of hyperpolarized gas diffusion in lung airspaces, provides unique information on the lung microstructure at the alveolar level. 3D tomographic images of standard morphological parameters (mean airspace chord length, lung parenchyma surface-to-volume ratio, and the number of alveoli per unit lung volume) can be created from a rather short (several seconds) MRI scan. These parameters are most commonly used to characterize lung morphometry but were not previously available from in vivo studies. A background of the (3)He lung morphometry technique is based on a previously proposed model of lung acinar airways, treated as cylindrical passages of external radius R covered by alveolar sleeves of depth h, and on a theory of gas diffusion in these airways. The initial works approximated the acinar airways as very long cylinders, all with the same R and h. The present work aims at analyzing effects of realistic acinar airway structures, incorporating airway branching, physiological airway lengths, a physiological ratio of airway ducts and sacs, and distributions of R and h. By means of Monte-Carlo computer simulations, we demonstrate that our technique allows rather accurate measurements of geometrical and morphological parameters of acinar airways. In particular, the accuracy of determining one of the most important physiological parameter of lung parenchyma - surface-to-volume ratio - does not exceed several percent. Second, we analyze the effect of the susceptibility induced inhomogeneous magnetic field on the parameter estimate and demonstrate that this effect is rather negligible at B(0) ≤ 3T and becomes substantial only at higher B(0) Third, we theoretically derive an optimal choice of MR pulse sequence parameters, which should be used to acquire a series of diffusion-attenuated MR signals, allowing a substantial decrease in the acquisition time and improvement in accuracy of the results. It is demonstrated that the optimal choice represents three not equidistant b-values: b(1)=0, b(2)∼2 s/cm(2), b(3)∼8 s/cm(2).

Evaluating the accuracy and precision of a two-compartment Kärger model using Monte Carlo simulations.

Specific parameters of the neuronal tissue microstructure, such as axonal diameters, membrane permeability and intracellular water fractions are assessable using diffusion MRI. These parameters are commonly estimated using analytical models, which may introduce bias in the estimated parameters due to the approximations made when deriving the models. As an alternative to using analytical models, a database of signal curves generated by fast Monte Carlo simulations can be employed. Simulated diffusion MRI measurements were generated and evaluated using the two-compartment Kärger model as well as the simulation model based on a database containing signal curves from approximately 60000 simulations performed with different combinations of microstructural parameters. A protocol based on a pulsed gradient spin echo sequence with diffusion times of 30 and 60 ms and with gradient amplitudes obtainable with a clinical MRI scanner was employed for the investigations. When using the analytical model, a major negative bias (up to approximately 25%) in the estimated intracellular volume fraction was observed for short exchange times, while almost no bias was seen for the simulation model. In general, the simulation model improved the accuracy of the estimated parameters as compared to the analytical model, except for the exchange time parameter.

Double-wave-vector diffusion-weighting experiments with multiple concatenations at long mixing times.

MR sequences where two diffusion-weighting periods are applied successively in a single acquisition seem to be a promising tool for the investigation of tissue structure on a microscopic level such as the characterization of the compartment size or eccentricity measures of pores. However, the application of such double-wave-vector (DWV) experiments on whole-body MR systems is hampered by the long gradient pulses required that have been shown to reduce the signal modulation. In this work, it is demonstrated that involving multiple concatenations of the two diffusion-weighting periods can ameliorate this problem in experiments with long mixing times between the two wave vectors. The recently presented tensor equation is extended to multiple concatenations. As confirmed by Monte-Carlo simulations, this model shows a good approximation of the signals observed for typical whole-body gradient pulse durations and the derived anisotropy measures are obtained with good accuracy. Most importantly, the signal modulation is increased with multiple concatenations because shorter gradient pulses can be used to achieve the desired diffusion-weighting. Thus, the multiple concatenation approach may help to improve the applicability and reliability of DWV measurements with long mixing times on standard whole-body MR systems.

An optimized wild bootstrap method for evaluation of measurement uncertainties of DTI-derived parameters in human brain.

Evaluation of measurement uncertainties (or errors) in diffusion tensor-derived parameters is essential to quantify the sensitivity and specificity of these quantities as potential surrogate biomarkers for pathophysiological processes with diffusion tensor imaging (DTI). Computational methods such as the Monte Carlo simulation have provided insights into characterization of the measurement uncertainty in DTI. However, due to the complexity of real brain data as well as different sources of variations during the image acquisition, a robust estimator for DTI measurement uncertainty in human brain is required. Recent studies have shown that wild bootstrap, a novel nonparametric statistical method, can potentially provide effective estimations of DTI measurement uncertainties in human brain DTI data. In this study, we further optimized the DTI application of the wild bootstrap method for typical clinical applications. We evaluated the validity of wild bootstrap utilizing numerical simulations with different combinations of DTI protocol parameters and wild bootstrap experimental designs, and quantitatively compared estimates of uncertainties from wild bootstrapping with those from Monte Carlo simulations. Our results demonstrate that a wild bootstrap implementation using at least 1000 wild bootstrap iterations with a type II or type III heteroskedasticity consistent covariance matrix estimator provides robust evaluations of most DTI protocols.

Whole brain voxel-wise analysis of single-subject serial DTI by permutation testing.

Diffusion tensor MRI (DTI) has been widely used to investigate brain microstructural changes in pathological conditions as well as for normal development and aging. In particular, longitudinal changes are vital to the understanding of progression but these studies are typically designed for specific regions of interest. To analyze changes in these regions traditional statistical methods are often employed to elucidate group differences which are measured against the variability found in a control cohort. However, in some cases, rather than collecting multiple subjects into two groups, it is necessary and more informative to analyze the data for individual subjects. There is also a need for understanding changes in a single subject without prior information regarding the spatial distribution of the pathology, but no formal statistical framework exists for these voxel-wise analyses of DTI. In this study, we present PERVADE (permutation voxel-wise analysis of diffusion estimates), a whole brain analysis method for detecting localized FA changes between two separate points in time of any given subject, without any prior hypothesis about where changes might occur. Exploiting the nature of DTI that it is calculated from multiple diffusion-weighted images of each region, permutation testing, a non-parametric hypothesis testing technique, was modified for the analysis of serial DTI data and implemented for voxel-wise hypothesis tests of diffusion metric changes, as well as for suprathreshold cluster analysis to correct for multiple comparisons. We describe PERVADE in detail and present results from Monte Carlo simulation supporting the validity of the technique as well as illustrative examples from a healthy subject and patients in the early stages of multiple sclerosis.

A method for calibrating diffusion gradients in diffusion tensor imaging.

OBJECTIVE: To calibrate and correct the gradient errors including gradient amplitude scaling errors, background/imaging gradients, and residual gradients in diffusion tensor imaging (DTI). METHODS: A calibration protocol using an isotropic phantom was proposed. Gradient errors were estimated by using linear regression analyses on quadratic functions of diffusion gradients along 3 orthogonal directions. A 6-element total effective scaling vector is generated from the calibration protocol to retrospectively correct gradient errors in DTI experiments. RESULTS: The accuracy of the calibration protocol was within 1% or less in estimating gradient scaling errors. On both the brain study and the computer simulations, the retrospective correction minimized undesirable estimate biases of DTI measurements due to gradient errors. CONCLUSION: The protocol and retrospective correction are shown to be effective. The method may be used for prospective correction if actual diffusion-gradient waveforms are available. The methodology is expandable to general diffusion imaging schemes.

Bayesian regularization of diffusion tensor images.

Diffusion tensor imaging (DTI) is a powerful tool in the study of the course of nerve fiber bundles in the human brain. Using DTI, the local fiber orientation in each image voxel can be described by a diffusion tensor which is constructed from local measurements of diffusion coefficients along several directions. The measured diffusion coefficients and thereby the diffusion tensors are subject to noise, leading to possibly flawed representations of the 3-dimensional (3D) fiber bundles. In this paper, we develop a Bayesian procedure for regularizing the diffusion tensor field, fully utilizing the available 3D information of fiber orientation. The use of the procedure is exemplified on synthetic and in vivo data.