A practical model-based segmentation approach for improved activation detection in single-subject functional magnetic resonance imaging studies.

Functional magnetic resonance imaging (fMRI) maps cerebral activation in response to stimuli but this activation is often difficult to detect, especially in low-signal contexts and single-subject studies. Accurate activation detection can be guided by the fact that very few voxels are, in reality, truly activated and that these voxels are spatially localized, but it is challenging to incorporate both these facts. We address these twin challenges to single-subject and low-signal fMRI by developing a computationally feasible and methodologically sound model-based approach, implemented in the R package MixfMRI, that bounds the a priori expected proportion of activated voxels while also incorporating spatial context. An added benefit of our methodology is the ability to distinguish voxels and regions having different intensities of activation. Our suggested approach is evaluated in realistic two- and three-dimensional simulation experiments as well as on multiple real-world datasets. Finally, the value of our suggested approach in low-signal and single-subject fMRI studies is illustrated on a sports imagination experiment that is often used to detect awareness and improve treatment in patients in persistent vegetative state (PVS). Our ability to reliably distinguish activation in this experiment potentially opens the door to the adoption of fMRI as a clinical tool for the improved treatment and therapy of PVS survivors and other patients.

Reduced-Rank Tensor-on-Tensor Regression and Tensor-Variate Analysis of Variance.

Fitting regression models with many multivariate responses and covariates can be challenging, but such responses and covariates sometimes have tensor-variate structure. We extend the classical multivariate regression model to exploit such structure in two ways: first, we impose four types of low-rank tensor formats on the regression coefficients. Second, we model the errors using the tensor-variate normal distribution that imposes a Kronecker separable format on the covariance matrix. We obtain maximum likelihood estimators via block-relaxation algorithms and derive their computational complexity and asymptotic distributions. Our regression framework enables us to formulate tensor-variate analysis of variance (TANOVA) methodology. This methodology, when applied in a one-way TANOVA layout, enables us to identify cerebral regions significantly associated with the interaction of suicide attempters or non-attemptor ideators and positive-, negative- or death-connoting words in a functional Magnetic Resonance Imaging study. Another application uses three-way TANOVA on the Labeled Faces in the Wild image dataset to distinguish facial characteristics related to ethnic origin, age group and gender. A R package totr implements the methodology.

Personalized synthetic MR imaging with deep learning enhancements.

PURPOSE: Personalized synthetic MRI (syn-MRI) uses MR images of an individual subject acquired at a few design parameters (echo time, repetition time, flip angle) to obtain underlying parametric ( ρ , T 1 , T 2 ) $$ \left(\rho, {\mathrm{T}}_1,{\mathrm{T}}_2\right) $$ maps, from where MR images of that individual at other design parameter settings are synthesized. However, classical methods that use least-squares (LS) or maximum likelihood estimators (MLE) are unsatisfactory at higher noise levels because the underlying inverse problem is ill-posed. This article provides a pipeline to enhance the synthesis of such images in three-dimensional (3D) using a deep learning (DL) neural network architecture for spatial regularization in a personalized setting where having more than a few training images is impractical. METHODS: Our DL enhancements employ a Deep Image Prior (DIP) with a U-net type denoising architecture that includes situations with minimal training data, such as personalized syn-MRI. We provide a general workflow for syn-MRI from three or more training images. Our workflow, called DIPsyn-MRI, uses DIP to enhance training images, then obtains parametric images using LS or MLE before synthesizing images at desired design parameter settings. DIPsyn-MRI is implemented in our publicly available Python package DeepSynMRI available at: https://github.com/StatPal/DeepSynMRI. RESULTS: We demonstrate feasibility and improved performance of DIPsyn-MRI on 3D datasets acquired using the Brainweb interface for spin-echo and FLASH imaging sequences, at different noise levels. Our DL enhancements improve syn-MRI in the presence of different intensity nonuniformity levels of the magnetic field, for all but very low noise levels. CONCLUSION: This article provides recipes and software to realistically facilitate DL-enhanced personalized syn-MRI.

Quantitative matching of forensic evidence fragments utilizing 3D microscopy analysis of fracture surface replicas.

Silicone casts are widely used by practitioners in the comparative analysis of forensic items. Fractured surfaces carry unique details that can provide accurate quantitative comparisons of forensic fragments. In this study, a statistical analysis comparison protocol was applied to a set of 3D topological images of fractured surface pairs and their replicas to provide confidence in the quantitative statistical comparison between fractured items and their silicone cast replicas. A set of 10 fractured stainless steel samples were fractured from the same metal rod under controlled conditions and were replicated using a standard forensic casting technique. Six 3D topological maps with 50% overlap were acquired for each fractured pair. Spectral analyses were utilized to identify the correlation between topological surface features at different length scales of the surface topology. We selected two frequency bands over the critical wavelength (greater than two-grain diameters) for statistical comparison. Our statistical model utilized a matrix-variate t-distribution that accounts for overlap between images to model match and non-match population densities. A decision rule identified the probability of matched and unmatched pairs of surfaces. The proposed methodology correctly classified the fractured steel surfaces and their replicas with a posterior probability of match exceeding 99.96%. Moreover, the replication technique shows potential in accurately replicating fracture surface topological details with a wavelength greater than 20 µm, which far exceeds the feature comparison range on most metallic alloy surfaces. Our framework establishes the basis and limits for forensic comparison of fractured articles and their replicas while providing a reliable fracture mechanics-based quantitative statistical forensic comparison.

A Matrix-free Likelihood Method for Exploratory Factor Analysis of High-dimensional Gaussian Data.

This paper proposes a novel profile likelihood method for estimating the covariance parameters in exploratory factor analysis of high-dimensional Gaussian datasets with fewer observations than number of variables. An implicitly restarted Lanczos algorithm and a limited-memory quasi-Newton method are implemented to develop a matrix-free framework for likelihood maximization. Simulation results show that our method is substantially faster than the expectation-maximization solution without sacrificing accuracy. Our method is applied to fit factor models on data from suicide attempters, suicide ideators and a control group.

Classification with the matrix-variate-t distribution.

Matrix-variate distributions can intuitively model the dependence structure of matrix-valued observations that arise in applications with multivariate time series, spatio-temporal or repeated measures. This paper develops an Expectation-Maximization algorithm for discriminant analysis and classification with matrix-variate t-distributions. The methodology shows promise on simulated datasets or when applied to the forensic matching of fractured surfaces or the classification of functional Magnetic Resonance, satellite or hand gestures images.

Efficient Bandwidth Estimation in 2D Filtered Backprojection Reconstruction.

A generalized cross-validation approach to estimate the reconstruction filter bandwidth in 2D filtered backprojection is presented. The method writes the reconstruction equation in equivalent backprojected filtering form, derives results on eigendecomposition of symmetric 2D circulant matrices, and applies them to make bandwidth estimation a computationally efficient operation within the context of standard backprojected filtering reconstruction. Performance evaluations on a range of simulated emission tomography experiments give promising results. The superior performance holds at both low and high total expected counts, pointing to the method's applicability even in weak signal-to-noise-ratio situations. The approach also applies to the more general class of elliptically symmetric filters, with the reconstructed estimate's performance often better than even that obtained with the true optimal radially symmetric filter.

Fast Adaptive Smoothing and Thresholding for Improved Activation Detection in Low-Signal fMRI.

Functional magnetic resonance imaging is a noninvasive tool for studying cerebral function. Many factors challenge activation detection, especially in low-signal scenarios that arise in the performance of high-level cognitive tasks. We provide a fully automated fast adaptive smoothing and thresholding (FAST) algorithm that uses smoothing and extreme value theory on correlated statistical parametric maps for thresholding. Performance on experiments spanning a range of low-signal settings is very encouraging. The methodology also performs well in a study to identify the cerebral regions that perceive only-auditory-reliable or only-visual-reliable speech stimuli.

COMPLEX-VALUED TIME SERIES MODELING FOR IMPROVED ACTIVATION DETECTION IN FMRI STUDIES.

A complex-valued data-based model with pth order autoregressive errors and general real/imaginary error covariance structure is proposed as an alternative to the commonly-used magnitude-only data-based autoregressive model for fMRI time series. Likelihood-ratio-test-based activation statistics are derived for both models and compared for experimental and simulated data. For a dataset from a right-hand finger-tapping experiment, the activation map obtained using complex-valued modeling more clearly identifies the primary activation region (left functional central sulcus) than the magnitude-only model. Such improved accuracy in mapping the left functional central sulcus has important implications in neurosurgical planning for tumor and epilepsy patients. Additionally, we develop magnitude and phase detrending procedures for complex-valued time series and examine the effect of spatial smoothing. These methods improve the power of complex-valued data-based activation statistics. Our results advocate for the use of the complex-valued data and the modeling of its dependence structures as a more efficient and reliable tool in fMRI experiments over the current practice of using only magnitude-valued datasets.

Merging K-means with hierarchical clustering for identifying general-shaped groups.

Clustering partitions a dataset such that observations placed together in a group are similar but different from those in other groups. Hierarchical and K-means clustering are two approaches but have different strengths and weaknesses. For instance, hierarchical clustering identifies groups in a tree-like structure but suffers from computational complexity in large datasets while K-means clustering is efficient but designed to identify homogeneous spherically-shaped clusters. We present a hybrid non-parametric clustering approach that amalgamates the two methods to identify general-shaped clusters and that can be applied to larger datasets. Specifically, we first partition the dataset into spherical groups using K-means. We next merge these groups using hierarchical methods with a data-driven distance measure as a stopping criterion. Our proposal has the potential to reveal groups with general shapes and structure in a dataset. We demonstrate good performance on several simulated and real datasets.

Ricean over Gaussian modelling in magnitude fMRI Analysis-Added Complexity with Negligible Practical Benefits.

It is well-known that Gaussian modeling of functional Magnetic Resonance Imaging (fMRI) magnitude time-course data, which are truly Rice-distributed, constitutes an approximation, especially at low signal-to-noise ratios (SNRs). Based on this fact, previous work has argued that Rice-based activation tests show superior performance over their Gaussian-based counterparts at low SNRs and should be preferred in spite of the attendant additional computational and estimation burden. Here, we revisit these past studies and after identifying and removing their underlying limiting assumptions and approximations, provide a more comprehensive comparison. Our experimental evaluations using ROC curve methodology show that tests derived using Ricean modeling are substantially superior over the Gaussian-based activation tests only for SNRs below 0.6, i.e SNR values far lower than those encountered in fMRI as currently practiced.

On the Expectation-Maximization Algorithm for Rice-Rayleigh Mixtures With Application to Noise Parameter Estimation in Magnitude MR Datasets.

Magnitude magnetic resonance (MR) images are noise-contaminated measurements of the true signal, and it is important to assess the noise in many applications. A recently introduced approach models the magnitude MR datum at each voxel in terms of a mixture of upto one Rayleigh and an a priori unspecified number of Rice components, all with a common noise parameter. The Expectation-Maximization (EM) algorithm was developed for parameter estimation, with the mixing component membership of each voxel as the missing observation. This paper revisits the EM algorithm by introducing more missing observations into the estimation problem such that the complete (observed and missing parts) dataset can be modeled in terms of a regular exponential family. Both the EM algorithm and variance estimation are then fairly straightforward without any need for potentially unstable numerical optimization methods. Compared to local neighborhood- and wavelet-based noise-parameter estimation methods, the new EM-based approach is seen to perform well not only on simulation datasets but also on physical phantom and clinical imaging data.

Synthetic magnetic resonance imaging revisited.

Synthetic magnetic resonance (MR) imaging is an approach suggested in the literature to predict MR images at different design parameter settings from at least three observed MR scans. However, performance is poor when no regularization is used in the estimation and otherwise computationally impractical to implement for 3-D imaging methods. We propose a method which accounts for spatial context in MR images by the imposition of a Gaussian Markov random field (MRF) structure on a transformation of the spin-lattice relaxation time, the spin-spin relaxation time and the proton density at each voxel. The MRF structure is specified through a matrix normal distribution. We also model the observed magnitude images using the more accurate but computationally challenging Rice distribution. A one-step-late expectation-maximization approach is adopted to make our approach computationally practical. We evaluate predictive performance in generating synthetic MR images in a clinical setting: our results indicate that our suggested approach is not only computationally feasible to implement but also shows excellent performance.

A re-defined and generalized percent-overlap-of-activation measure for studies of fMRI reproducibility and its use in identifying outlier activation maps.

Functional Magnetic Resonance Imaging (fMRI) is a popular noninvasive modality to investigate activation in the human brain. The end result of most fMRI experiments is an activation map corresponding to the given paradigm. These maps can vary greatly from one study to the next, so quantifying the reliability of identified activation over several fMRI studies is important. The percent overlap of activation (Rombouts et al., 1998; Machielsen et al., 2000) is a global reliability measure between activation maps drawn from any two fMRI studies. A slightly modified but more intuitive measure is provided by the Jaccard (1901) coefficient of similarity, whose use we study in this paper. A generalization of these measures is also proposed to comprehensively summarize the reliability of multiple fMRI studies. Finally, a testing mechanism to flag potentially anomalous studies is developed. The methodology is illustrated on studies involving left- and right-hand motor task paradigms performed by a right-hand dominant male subject several times over a period of two months, with excellent results.

Noise estimation in magnitude MR datasets.

Estimating the noise parameter in magnitude magnetic resonance (MR) images is important in a wide range of applications. We propose an automatic noise estimation method that does not rely on a substantial proportion of voxels being from the background. Specifically, we model the magnitude of the observed signal as a mixture of Rice distributions with common noise parameter. The expectation-maximization (EM) algorithm is used to estimate all parameters, including the common noise parameter. The algorithm needs initializing values for which we provide some strategies that work well. The number of components in the mixture model also needs to be estimated en route to noise estimation and we provide a novel approach to doing so. Our methodology performs very well on a range of simulation experiments and physical phantom data. Finally, the methodology is demonstrated on four clinical datasets.

Assessing certainty of activation or inactivation in test-retest fMRI studies.

Functional Magnetic Resonance Imaging (fMRI) is widely used to study activation in the human brain. In most cases, data are commonly used to construct activation maps corresponding to a given paradigm. Results can be very variable, hence quantifying certainty of identified activation and inactivation over studies is important. This paper provides a model-based approach to certainty estimation from data acquired over several replicates of the same experimental paradigm. Specifically, the p-values derived from the statistical analysis of the data are explicitly modeled as a mixture of their underlying distributions; thus, unlike the methodology currently in use, there is no subjective thresholding required in the estimation process. The parameters governing the mixture model are easily obtained by the principle of maximum likelihood. Further, the estimates can also be used to optimally identify voxel-specific activation regions along with their corresponding certainty measures. The methodology is applied to a study involving a motor paradigm performed on a single subject several times over a period of two months. Simulation experiments used to calibrate performance of the method are promising. The methodology is also seen to be robust in determining areas of activation and their corresponding certainties.

Initializing partition-optimization algorithms.

Clustering datasets is a challenging problem needed in a wide array of applications. Partition-optimization approaches, such as k-means or expectation-maximization (EM) algorithms, are sub-optimal and find solutions in the vicinity of their initialization. This paper proposes a staged approach to specifying initial values by finding a large number of local modes and then obtaining representatives from the most separated ones. Results on test experiments are excellent. We also provide a detailed comparative assessment of the suggested algorithm with many commonly-used initialization approaches in the literature. Finally, the methodology is applied to two datasets on diurnal microarray gene expressions and industrial releases of mercury.

Clustering in the presence of scatter.

SUMMARY: A new methodology is proposed for clustering datasets in the presence of scattered observations. Scattered observations are defined as unlike any other, so traditional approaches that force them into groups can lead to erroneous conclusions. Our suggested approach is a scheme which, under assumption of homogeneous spherical clusters, iteratively builds cores around their centers and groups points within each core while identifying points outside as scatter. In the absence of scatter, the algorithm reduces to k-means. We also provide methodology to initialize the algorithm and to estimate the number of clusters in the dataset. Results in experimental situations show excellent performance, especially when clusters are elliptically symmetric. The methodology is applied to the analysis of the United States Environmental Protection Agency's Toxic Release Inventory reports on industrial releases of mercury for the year 2000.

Wrist circumference is related to patellar tendon thickness in healthy men and women.

The efficacy of dominant wrist circumference measurements to predict dominant lower extremity patellar tendon thickness at regions of interest for bone-patellar tendon-bone (BPTB) autograft harvest was studied among 24 healthy men and women. Dominant wrist circumference displayed good relationships with dominant lower extremity patellar tendon thickness as determined by two-dimensional diagnostic ultrasound. This initial screening method may assist surgeons as they consider graft selection for patients who may be at risk for developing or exacerbating preexisting patellofemoral joint or knee extensor mechanism conditions with BPTB autograft harvest.

Sex differences in the cerebral BOLD signal response to painful heat stimuli.

There are limited data addressing the question of sex differences in pain-related cerebral processing. This study examined whether pain-related blood oxygenation level-dependent (BOLD) signal change measured with functional magnetic resonance imaging (fMRI) demonstrated sex differences, under conditions of equivalent pain perception. Twenty-eight healthy volunteers (17 women, 11 men) were subject to a fMRI scan while noxious heat stimuli were applied to the dorsum of the left foot. Significant BOLD signal modulation was observed in several nociceptive processing regions of interest (ROIs) in all subjects. There were no sex differences in the spatial extent of BOLD signal change for any ROI, but the signal amplitude was lower for women in most ROIs and significantly so for the primary somatosensory cortex (S1), the midanterior cingulate cortex, and the dorsolateral prefrontal cortex (DLPFC). The BOLD signal response could be positive or negative, and frequently, both polarities were observed within a single ROI. In most ROIs, women show proportionately more voxels with negative signal change than men, and this difference was statistically significant for the S1 and the DLPFC. The time course of the negative signal change was very similar to that of the positive signal change, suggesting that the latter was not "driving" the former. The location of negative and positive clusters formed distinct patterns in several of the ROIs, and these patterns suggest something other than a local "steal" phenomenon as an explanation for the negative signal changes. Sex differences in baseline cerebral blood flow may contribute to the BOLD signal differences observed in this study.