Diffusion-weighted MRI (dMRI) is increasingly used to study the normal and abnormal development of fetal brain in-utero. It offers invaluable insights into the neurodevelopmental processes in the fetal stage. However, reliable analysis of fetal dMRI data requires dedicated computational methods that are currently unavailable. The lack of automated methods for fast, accurate, and reproducible data analysis has seriously limited our ability to tap the potential of fetal brain dMRI for medical and scientific applications. In this work, we developed and validated a unified computational framework to:1) segment the brain tissue into white matter, cortical/subcortical gray matter, and cerebrospinal fluid,:2) segment 31 distinct white matter tracts, and:3) parcellate the brain's cortex, deep gray nuclei, and white matter structures into 96 anatomically meaningful regions. We utilized a set of manual, semi-automatic, and automatic approaches to annotate 97 fetal brains. Using these labels, we developed and validated a multi-task deep learning method to perform the three computations. Evaluations show that the new method can accurately carry out all three tasks, achieving a mean Dice similarity coefficient of 0.865 on tissue segmentation, 0.825 on white matter tract segmentation, and 0.819 on parcellation. Further validation on independent external data shows generalizability of the proposed method. The new method can help advance the field of fetal neuroimaging as it can lead to substantial improvements in fetal brain tractography, tract-specific analysis, and structural connectivity assessment.

Functional MRI (fMRI) is widely used to study the functional organization of normal and pathological brains. However, the fMRI signal may be contaminated by subject motion artifacts that are only partially mitigated by motion correction strategies. These artifacts lead to distance-dependent biases in the inferred signal correlations. To mitigate these spurious effects, motion-corrupted volumes are censored from fMRI time series. Censoring can result in discontinuities in the fMRI signal, which may lead to substantial alterations in functional connectivity analysis. We propose a new approach to recover the missing entries from censoring based on structured low rank matrix completion. We formulated the artifact-reduction problem as the recovery of a super-resolved matrix from unprocessed fMRI measurements. We enforced a low rank prior on a large structured matrix, formed from the samples of the time series, to recover the missing entries. The recovered time series, in addition to being motion compensated, are also slice-time corrected at a fine temporal resolution. To achieve a fast and memory-efficient solution for our proposed optimization problem, we employed a variable splitting strategy. We validated the algorithm with simulations, data acquired under different motion conditions, and datasets from the ABCD study. Functional connectivity analysis showed that the proposed reconstruction resulted in connectivity matrices with lower errors in pair-wise correlation than non-censored and censored time series based on a standard processing pipeline. In addition, seed-based correlation analyses showed improved delineation of the default mode network. These demonstrate that the method can effectively reduce the adverse effects of motion in fMRI analysis.

The interpretation and analysis of Magnetic resonance imaging (MRI) benefit from high spatial resolution. Unfortunately, direct acquisition of high spatial resolution MRI is time-consuming and costly, which increases the potential for motion artifact, and suffers from reduced signal-to-noise ratio (SNR). Super-resolution reconstruction (SRR) is one of the most widely used methods in MRI since it allows for the trade-off between high spatial resolution, high SNR, and reduced scan times. Deep learning has emerged for improved SRR as compared to conventional methods. However, current deep learning-based SRR methods require large-scale training datasets of high-resolution images, which are practically difficult to obtain at a suitable SNR. We sought to develop a methodology that allows for dataset-free deep learning-based SRR, through which to construct images with higher spatial resolution and of higher SNR than can be practically obtained by direct Fourier encoding. We developed a dataset-free learning method that leverages a generative neural network trained for each specific scan or set of scans, which in turn, allows for SRR tailored to the individual patient. With the SRR from three short duration scans, we achieved high quality brain MRI at an isotropic spatial resolution of 0.125 cubic mm with six minutes of imaging time for T2 contrast and an average increase of 7.2 dB (34.2%) in SNR to these short duration scans. Motion compensation was achieved by aligning the three short duration scans together. We assessed our technique on simulated MRI data and clinical data acquired from 15 subjects. Extensive experimental results demonstrate that our approach achieved superior results to state-of-the-art methods, while in parallel, performed at reduced cost as scans delivered with direct high-resolution acquisition.

Computed tomography (CT) image reconstruction is a crucial technique for many imaging applications. Among various reconstruction methods, Model-Based Iterative Reconstruction (MBIR) enables super-resolution with superior image quality. MBIR, however, has a high memory requirement that limits the achievable image resolution, and the parallelization for MBIR suffers from limited scalability. In this paper, we propose Asynchronous Consensus MBIR (AC-MBIR) that uses Consensus Equilibrium (CE) to provide a super-resolution algorithm with a small memory footprint, low communication overhead and a high scalability. Super-resolution experiments show that AC-MBIR has a 6.8 times smaller memory footprint and 16 times more scalability, compared with the state-of-the-art MBIR implementation, and maintains a 100% strong scaling efficiency at 146880 cores. In addition, AC-MBIR achieves an average bandwidth of 3.5 petabytes per second at 587520 cores.

Deep convolutional neural networks (CNN) have recently achieved superior performance at the task of medical image segmentation compared to classic models. However, training a generalizable CNN requires a large amount of training data, which is difficult, expensive, and time-consuming to obtain in medical settings. Active Learning (AL) algorithms can facilitate training CNN models by proposing a small number of the most informative data samples to be annotated to achieve a rapid increase in performance. We proposed a new active learning method based on Fisher information (FI) for CNNs for the first time. Using efficient backpropagation methods for computing gradients together with a novel low-dimensional approximation of FI enabled us to compute FI for CNNs with a large number of parameters. We evaluated the proposed method for brain extraction with a patch-wise segmentation CNN model in two different learning scenarios: universal active learning and active semi-automatic segmentation. In both scenarios, an initial model was obtained using labeled training subjects of a source data set and the goal was to annotate a small subset of new samples to build a model that performs well on the target subject(s). The target data sets included images that differed from the source data by either age group (e.g. newborns with different image contrast) or underlying pathology that was not available in the source data. In comparison to several recently proposed AL methods and brain extraction baselines, the results showed that FI-based AL outperformed the competing methods in improving the performance of the model after labeling a very small portion of target data set (<0.25%).

Let us consider a case where all of the elements in some continuous slices are missing in tensor data. In this case, the nuclear-norm and total variation regularization methods usually fail to recover the missing elements. The key problem is capturing some delay/shift-invariant structure. In this study, we consider a low-rank model in an embedded space of a tensor. For this purpose, we extend a delay embedding for a time series to a "multi-way delay-embedding transform" for a tensor, which takes a given incomplete tensor as the input and outputs a higher-order incomplete Hankel tensor. The higher-order tensor is then recovered by Tucker-based low-rank tensor factorization. Finally, an estimated tensor can be obtained by using the inverse multiway delay embedding transform of the recovered higher-order tensor. Our experiments showed that the proposed method successfully recovered missing slices for some color images and functional magnetic resonance images.

We propose an algorithm for electrical source imaging of epileptic discharges that takes a data-driven approach to regularizing the dynamics of solutions. The method is based on linear system identification on short time segments, combined with a classical inverse solution approach. Whereas ensemble averaging of segments or epochs discards inter-segment variations by averaging across them, our approach explicitly models them. Indeed, it may even be possible to avoid the need for the time-consuming process of marking epochs containing discharges altogether. We demonstrate that this approach can produce both stable and accurate inverse solutions in experiments using simulated data and real data from epilepsy patients. In an illustrative example, we show that we are able to image propagation using this approach. We show that when applied to imaging seizure data, our approach reproducibly localized frequent seizure activity to within the margins of surgeries that led to patients' seizure freedom. The same approach could be used in the planning of epilepsy surgeries, as a way to localize potentially epileptogenic tissue that should be resected.

By shortening the acquisition time of MRI, Echo Planar Imaging (EPI) enables the acquisition of a large number of images in a short time, compatible with clinical constraints as required for diffusion or functional MRI. However such images are subject to large, local distortions disrupting their correspondence with the underlying anatomy. The correction of those distortions is an open problem, especially in regions where large deformations occur. We propose a new block-matching registration method to perform EPI distortion correction based on the acquisition of two EPI with opposite phase encoding directions (PED). It relies on new transformations between blocks adapted to the EPI distortion model, and on an adapted optimization scheme to ensure an opposite symmetric transformation. We present qualitative and quantitative results of the block-matching correction using different metrics on a phantom dataset and on in-vivo data. We show the ability of the block-matching to robustly correct EPI distortion even in strongly affected areas.

In magnetic resonance (MR), hardware limitations, scan time constraints, and patient movement often result in the acquisition of anisotropic 3-D MR images with limited spatial resolution in the out-of-plane views. Our goal is to construct an isotropic high-resolution (HR) 3-D MR image through upsampling and fusion of orthogonal anisotropic input scans. We propose a multiframe super-resolution (SR) reconstruction technique based on sparse representation of MR images. Our proposed algorithm exploits the correspondence between the HR slices and the low-resolution (LR) sections of the orthogonal input scans as well as the self-similarity of each input scan to train pairs of overcomplete dictionaries that are used in a sparse-land local model to upsample the input scans. The upsampled images are then combined using wavelet fusion and error backprojection to reconstruct an image. Features are learned from the data and no extra training set is needed. Qualitative and quantitative analyses were conducted to evaluate the proposed algorithm using simulated and clinical MR scans. Experimental results show that the proposed algorithm achieves promising results in terms of peak signal-to-noise ratio, structural similarity image index, intensity profiles, and visualization of small structures obscured in the LR imaging process due to partial volume effects. Our novel SR algorithm outperforms the nonlocal means (NLM) method using self-similarity, NLM method using self-similarity and image prior, self-training dictionary learning-based SR method, averaging of upsampled scans, and the wavelet fusion method. Our SR algorithm can reduce through-plane partial volume artifact by combining multiple orthogonal MR scans, and thus can potentially improve medical image analysis, research, and clinical diagnosis.

Simultaneous multi-slice (SMS) echo-planar imaging has had a huge impact on the acceleration and routine use of diffusion-weighted MRI (DWI) in neuroimaging studies in particular the human connectome project; but also holds the potential to facilitate DWI of moving subjects, as proposed by the new technique developed in this paper. We present a novel registration-based motion tracking technique that takes advantage of the multi-plane coverage of the anatomy by simultaneously acquired slices to enable robust reconstruction of neural microstructure from SMS DWI of moving subjects. Our technique constitutes three main components: 1) motion tracking and estimation using SMS registration, 2) detection and rejection of intra-slice motion, and 3) robust reconstruction. Quantitative results from 14 volunteer subject experiments and the analysis of motion-corrupted SMS DWI of 6 children indicate robust reconstruction in the presence of continuous motion and the potential to extend the use of SMS DWI in very challenging populations.

Diffusion MRI is a key in-vivo non invasive imaging capability that can probe the microstructure of the brain. However, its limited resolution requires complex voxelwise generative models of the diffusion. Diffusion Compartment (DC) models divide the voxel into smaller compartments in which diffusion is homogeneous. We present a comprehensive framework for maximum likelihood estimation (MLE) of such models that jointly features ML estimators of (i) the baseline MR signal, (ii) the noise variance, (iii) compartment proportions, and (iv) diffusion-related parameters. ML estimators are key to providing reliable mapping of brain microstructure as they are asymptotically unbiased and of minimal variance. We compare our algorithm (which efficiently exploits analytical properties of MLE) to alternative implementations and a state-of-the-art strategy. Simulation results show that our approach offers the best reduction in computational burden while guaranteeing convergence of numerical estimators to the MLE. In-vivo results also reveal remarkably reliable microstructure mapping in areas as complex as the centrum semi-ovale. Our ML framework accommodates any DC model and is available freely for multi-tensor models as part of the ANIMA software ( https://github.com/Inria-Visages/Anima-Public/wiki ).

In magnetic resonance (MR), hardware limitation, scanning time, and patient comfort often result in the acquisition of anisotropic 3-D MR images. Enhancing image resolution is desired but has been very challenging in medical image processing. Super resolution reconstruction based on sparse representation and overcomplete dictionary has been lately employed to address this problem; however, these methods require extra training sets, which may not be always available. This paper proposes a novel single anisotropic 3-D MR image upsampling method via sparse representation and overcomplete dictionary that is trained from in-plane high resolution slices to upsample in the out-of-plane dimensions. The proposed method, therefore, does not require extra training sets. Abundant experiments, conducted on simulated and clinical brain MR images, show that the proposed method is more accurate than classical interpolation. When compared to a recent upsampling method based on the nonlocal means approach, the proposed method did not show improved results at low upsampling factors with simulated images, but generated comparable results with much better computational efficiency in clinical cases. Therefore, the proposed approach can be efficiently implemented and routinely used to upsample MR images in the out-of-planes views for radiologic assessment and postacquisition processing.

This work proposes a novel approach for motion-robust diffusion-weighted (DW) brain MRI reconstruction through tracking temporal head motion using slice-to-volume registration. The slice-level motion is estimated through a filtering approach that allows tracking the head motion during the scan and correcting for out-of-plane inconsistency in the acquired images. Diffusion-sensitized image slices are registered to a base volume sequentially over time in the acquisition order where an outlier-robust Kalman filter, coupled with slice-to-volume registration, estimates head motion parameters. Diffusion gradient directions are corrected for the aligned DWI slices based on the computed rotation parameters and the diffusion tensors are directly estimated from the corrected data at each voxel using weighted linear least squares. The method was evaluated in DWI scans of adult volunteers who deliberately moved during scans as well as clinical DWI of 28 neonates and children with different types of motion. Experimental results showed marked improvements in DWI reconstruction using the proposed method compared to the state-of-the-art DWI analysis based on volume-to-volume registration. This approach can be readily used to retrieve information from motion-corrupted DW imaging data.

In recent years, fusing segmentation results obtained based on multiple template images has become a standard practice in many medical imaging applications. Such multiple-templates-based methods are found to provide more reliable and accurate segmentations than the single-template-based methods. In this paper, we present a new approach for learning prior knowledge about the performance parameters of template images using the local intensity similarity information; we also propose a methodology to incorporate that prior knowledge through the estimation of the optimal MAP parameters. The proposed method is evaluated in the context of segmentation of structures in the brain magnetic resonance images by comparing our results with some of the state-of-the-art segmentation methods. These experiments have clearly demonstrated the advantages of learning and incorporating prior knowledge about the performance parameters using the proposed method.

White matter (WM) lesions are thought to play an important role in multiple sclerosis (MS) disease burden. Recent work in the automated segmentation of white matter lesions from magnetic resonance imaging has utilized a model in which lesions are outliers in the distribution of tissue signal intensities across the entire brain of each patient. However, the sensitivity and specificity of lesion detection and segmentation with these approaches have been inadequate. In our analysis, we determined this is due to the substantial overlap between the whole brain signal intensity distribution of lesions and normal tissue. Inspired by the ability of experts to detect lesions based on their local signal intensity characteristics, we propose a new algorithm that achieves lesion and brain tissue segmentation through simultaneous estimation of a spatially global within-the-subject intensity distribution and a spatially local intensity distribution derived from a healthy reference population. We demonstrate that MS lesions can be segmented as outliers from this intensity model of population and subject. We carried out extensive experiments with both synthetic and clinical data, and compared the performance of our new algorithm to those of state-of-the art techniques. We found this new approach leads to a substantial improvement in the sensitivity and specificity of lesion detection and segmentation.

Pelvic floor dysfunction is common in women after childbirth and precise segmentation of magnetic resonance images (MRI) of the pelvic floor may facilitate diagnosis and treatment of patients. However, because of the complexity of its structures, manual segmentation of the pelvic floor is challenging and suffers from high inter and intra-rater variability of expert raters. Multiple template fusion algorithms are promising segmentation techniques for these types of applications, but they have been limited by imperfections in the alignment of templates to the target, and by template segmentation errors. A number of algorithms sought to improve segmentation performance by combining image intensities and template labels as two independent sources of information, carrying out fusion through local intensity weighted voting schemes. This class of approach is a form of linear opinion pooling, and achieves unsatisfactory performance for this application. We hypothesized that better decision fusion could be achieved by assessing the contribution of each template in comparison to a reference standard segmentation of the target image and developed a novel segmentation algorithm to enable automatic segmentation of MRI of the female pelvic floor. The algorithm achieves high performance by estimating and compensating for both imperfect registration of the templates to the target image and template segmentation inaccuracies. A local image similarity measure is used to infer a local reliability weight, which contributes to the fusion through a novel logarithmic opinion pooling. We evaluated our new algorithm in comparison to nine state-of-the-art segmentation methods and demonstrated our algorithm achieves the highest performance.

Diffusion tensor imaging (DTI) is unable to represent the diffusion signal arising from multiple crossing fascicles and freely diffusing water molecules. Generative models of the diffusion signal, such as multi-fascicle models, overcome this limitation by providing a parametric representation for the signal contribution of each population of water molecules. These models are of great interest in population studies to characterize and compare the brain microstructural properties. Central to population studies is the construction of an atlas and the registration of all subjects to it. However, the appropriate definition of registration and atlasing methods for multi-fascicle models have proven challenging. This paper proposes a mathematical framework to register and analyze multi-fascicle models. Specifically, we define novel operators to achieve interpolation, smoothing and averaging of multi-fascicle models. We also define a novel similarity metric to spatially align multi-fascicle models. Our framework enables simultaneous comparisons of different microstructural properties that are confounded in conventional DTI. The framework is validated on multi-fascicle models from 24 healthy subjects and 38 patients with tuberous sclerosis complex, 10 of whom have autism. We demonstrate the use of the multi-fascicle models registration and analysis framework in a population study of autism spectrum disorder.

Recent research has demonstrated that improved image segmentation can be achieved by multiple template fusion utilizing both label and intensity information. However, intensity weighted fusion approaches use local intensity similarity as a surrogate measure of local template quality for predicting target segmentation and do not seek to characterize template performance. This limits both the usefulness and accuracy of these techniques. Our work here was motivated by the observation that the local intensity similarity is a poor surrogate measure for direct comparison of the template image with the true image target segmentation. Although the true image target segmentation is not available, a high quality estimate can be inferred, and this in turn allows a principled estimate to be made of the local quality of each template at contributing to the target segmentation. We developed a fusion algorithm that uses probabilistic segmentations of the target image to simultaneously infer a reference standard segmentation of the target image and the local quality of each probabilistic segmentation. The concept of comparing templates to a hidden reference standard segmentation enables accurate assessments of the contribution of each template to inferring the target image segmentation to be made, and in practice leads to excellent target image segmentation. We have used the new algorithm for the multiple-template-based segmentation and parcellation of magnetic resonance images of the brain. Intensity and label map images of each one of the aligned templates are used to train a local Gaussian mixture model based classifier. Then, each classifier is used to compute the probabilistic segmentations of the target image. Finally, the generated probabilistic segmentations are fused together using the new fusion algorithm to obtain the segmentation of the target image. We evaluated our method in comparison to other state-of-the-art segmentation methods. We demonstrated that our new fusion algorithm has higher segmentation performance than these methods.

The electroencephalography source estimation problem consists of inferring cortical activation from measurements of electrical potential taken on the scalp surface. This inverse problem is intrinsically ill-posed. In particular the dimensionality of cortical sources greatly exceeds the number of electrode measurements, and source estimation requires regularization to obtain a unique solution. In this work, we introduce a novel regularization function called cortical graph smoothing, which exploits knowledge of anatomical connectivity available from diffusion-weighted imaging. Given a weighted graph description of the anatomical connectivity of the brain, cortical graph smoothing penalizes the weighted sum of squares of differences of cortical activity across the graph edges, thus encouraging solutions with consistent activation across anatomically connected regions. We explore the performance of the cortical graph smoothing source estimates for analysis of the event related potential for simple motor tasks, and compare against the commonly used minimum norm, weighted minimum norm, LORETA and sLORETA source estimation methods. Evaluated over a series of 18 subjects, the proposed cortical graph smoothing method shows superior localization accuracy compared to the minimum norm method, and greater relative peak intensity than the other comparison methods.

We present a new algorithm, called local MAP STAPLE, to estimate from a set of multi-label segmentations both a reference standard segmentation and spatially varying performance parameters. It is based on a sliding window technique to estimate the segmentation and the segmentation performance parameters for each input segmentation. In order to allow for optimal fusion from the small amount of data in each local region, and to account for the possibility of labels not being observed in a local region of some (or all) input segmentations, we introduce prior probabilities for the local performance parameters through a new maximum a posteriori formulation of STAPLE. Further, we propose an expression to compute confidence intervals in the estimated local performance parameters. We carried out several experiments with local MAP STAPLE to characterize its performance and value for local segmentation evaluation. First, with simulated segmentations with known reference standard segmentation and spatially varying performance, we show that local MAP STAPLE performs better than both STAPLE and majority voting. Then we present evaluations with data sets from clinical applications. These experiments demonstrate that spatial adaptivity in segmentation performance is an important property to capture. We compared the local MAP STAPLE segmentations to STAPLE, and to previously published fusion techniques and demonstrate the superiority of local MAP STAPLE over other state-of-the-art algorithms.

We present a novel algorithm to accelerate feature based registration, and demonstrate the utility of the algorithm for the alignment of large transmission electron microscopy (TEM) images to create 3-D images of neural ultrastructure. In contrast to the most similar algorithms, which achieve small computation times by truncated search, our algorithm uses a novel randomized projection to accelerate feature comparison and to enable global search. Further, we demonstrate robust estimation of nonrigid transformations with a novel probabilistic correspondence framework, that enables large TEM images to be rapidly brought into alignment, removing characteristic distortions of the tissue fixation and imaging process. We analyze the impact of randomized projections upon correspondence detection, and upon transformation accuracy, and demonstrate that accuracy is maintained. We provide experimental results that demonstrate significant reduction in computation time and successful alignment of TEM images.

The evaluation of the quality of segmentations of an image, and the assessment of intra- and inter-expert variability in segmentation performance, has long been recognized as a difficult task. For a segmentation validation task, it may be effective to compare the results of an automatic segmentation algorithm to multiple expert segmentations. Recently an expectation-maximization (EM) algorithm for simultaneous truth and performance level estimation (STAPLE) was developed to this end to compute both an estimate of the reference standard segmentation and performance parameters from a set of segmentations of an image. The performance is characterized by the rate of detection of each segmentation label by each expert in comparison to the estimated reference standard. This previous work provides estimates of performance parameters,but does not provide any information regarding the uncertainty of the estimated values. An estimate of this inferential uncertainty, if available, would allow the estimation of confidence intervals for the values of the parameters. This would facilitate the interpretation of the performance of segmentation generators and help determine if sufficient data size and number of segmentations have been obtained to precisely characterize the performance parameters. We present a new algorithm to estimate the inferential uncertainty of the performance parameters for binary and multi-category segmentations. It is derived for the special case of the STAPLE algorithm based on established theory for general purpose covariance matrix estimation for EM algorithms. The bounds on the performance parameters are estimated by the computation of the observed information matrix.We use this algorithm to study the bounds on performance parameters estimates from simulated images with specified performance parameters, and from interactive segmentations of neonatal brain MRIs. We demonstrate that confidence intervals for expert segmentation performance parameters can be estimated with our algorithm. We investigate the influence of the number of experts and of the segmented data size on these bounds, showing that it is possible to determine the number of image segmentations and the size of images necessary to achieve a chosen level of accuracy in segmentation performance assessment.

In this paper, we present a segmentation scheme that automatically and accurately segments all the cartilages from magnetic resonance (MR) images of nonpathological knees. Our scheme involves the automatic segmentation of the bones using a three-dimensional active shape model, the extraction of the expected bone-cartilage interface (BCI), and cartilage segmentation from the BCI using a deformable model that utilizes localization, patient specific tissue estimation and a model of the thickness variation. The accuracy of this scheme was experimentally validated using leave one out experiments on a database of fat suppressed spoiled gradient recall MR images. The scheme was compared to three state of the art approaches, tissue classification, a modified semi-automatic watershed algorithm and nonrigid registration (B-spline based free form deformation). Our scheme obtained an average Dice similarity coefficient (DSC) of (0.83, 0.83, 0.85) for the (patellar, tibial, femoral) cartilages, while (0.82, 0.81, 0.86) was obtained with a tissue classifier and (0.73, 0.79, 0.76) was obtained with nonrigid registration. The average DSC obtained for all the cartilages using a semi-automatic watershed algorithm (0.90) was slightly higher than our approach (0.89), however unlike this approach we segment each cartilage as a separate object. The effectiveness of our approach for quantitative analysis was evaluated using volume and thickness measures with a median volume difference error of (5.92, 4.65, 5.69) and absolute Laplacian thickness difference of (0.13, 0.24, 0.12) mm.

Fast magnetic resonance imaging slice acquisition techniques such as single shot fast spin echo are routinely used in the presence of uncontrollable motion. These techniques are widely used for fetal magnetic resonance imaging (MRI) and MRI of moving subjects and organs. Although high-quality slices are frequently acquired by these techniques, inter-slice motion leads to severe motion artifacts that are apparent in out-of-plane views. Slice sequential acquisitions do not enable 3-D volume representation. In this study, we have developed a novel technique based on a slice acquisition model, which enables the reconstruction of a volumetric image from multiple-scan slice acquisitions. The super-resolution volume reconstruction is formulated as an inverse problem of finding the underlying structure generating the acquired slices. We have developed a robust M-estimation solution which minimizes a robust error norm function between the model-generated slices and the acquired slices. The accuracy and robustness of this novel technique has been quantitatively assessed through simulations with digital brain phantom images as well as high-resolution newborn images. We also report here successful application of our new technique for the reconstruction of volumetric fetal brain MRI from clinically acquired data.

The comparison of images of a patient to a reference standard may enable the identification of structural brain changes. These comparisons may involve the use of vector or tensor images (i.e., 3-D images for which each voxel can be represented as an RN vector) such as diffusion tensor images (DTI) or transformations. The recent introduction of the Log-Euclidean framework for diffeomorphisms and tensors has greatly simplified the use of these images by allowing all the computations to be performed on a vector-space. However, many sources can result in a bias in the images, including disease or imaging artifacts. In order to estimate and compensate for these sources of variability, we developed a new algorithm, called continuous STAPLE, that estimates the reference standard underlying a set of vector images. This method, based on an expectation-maximization method similar in principle to the validation method STAPLE, also estimates for each image a set of parameters characterizing their bias and variance with respect to the reference standard. We demonstrate how to use these parameters for the detection of atypical images or outliers in the population under study. We identified significant differences between the tensors of diffusion images of multiple sclerosis patients and those of control subjects in the vicinity of lesions.

Neurosurgical resection is a therapeutic intervention in the treatment of brain tumors. Precision of the resection can be improved by utilizing Magnetic Resonance Imaging (MRI) as an aid in decision making during Image Guided Neurosurgery (IGNS). Image registration adjusts pre-operative data according to intra-operative tissue deformation. Some of the approaches increase the registration accuracy by tracking image landmarks through the whole brain volume. High computational cost used to render these techniques inappropriate for clinical applications. In this paper we present a parallel implementation of a state of the art registration method, and a number of needed incremental improvements. Overall, we reduced the response time for registration of an average dataset from about an hour and for some cases more than an hour to less than seven minutes, which is within the time constraints imposed by neurosurgeons. For the first time in clinical practice we demonstrated, that with the help of distributed computing non-rigid MRI registration based on volume tracking can be computed intra-operatively.

This paper introduces a new metric to gather a large collection of segmented images into a same reference system. Different positions for each subject (pose parameters) as well as high energy shape variations need to be compensated before performing statistical analysis (like principal components analysis) on the database. The atlas is obtained as the hidden variable of an expectation-maximization (EM), looking for the right signal intensity at each voxel in the collection of subjects. Each subject is aligned on the current probabilistic atlas by maximizing mutual information. A fast stochastic optimization algorithm is used for estimating pose and scale parameters and a variational approach have been designed to estimate non-rigid transformations. We illustrate the effectiveness of this method for the alignment of 31 brain segmented in 4 labels: background, white and gray matter and ventricles. Our approach has the advantage of keeping a reasonably low complexity even for large databases.

This paper introduces a new registration method estimating the displacement field of bodies which deformations are constrained by an articulated rigid body. We propose an articulated transformation model embedded in a general registration scheme. A fast stochastic gradient descent optimization strategy suitable for noisy cost functions has been chosen to maximize the mutual information metric. Once registered, we propose to propagate the deformation by a linear elastic model through the use of a tetrahedral mesh. We demonstrate this method on 3D CT of neck images where bony structures between different patient images, as vertebrae, may rigidly move while other tissues may deform.

We present a new algorithm to register 3-D preoperative magnetic resonance (MR) images to intraoperative MR images of the brain which have undergone brain shift. This algorithm relies on a robust estimation of the deformation from a sparse noisy set of measured displacements. We propose a new framework to compute the displacement field in an iterative process, allowing the solution to gradually move from an approximation formulation (minimizing the sum of a regularization term and a data error term) to an interpolation formulation (least square minimization of the data error term). An outlier rejection step is introduced in this gradual registration process using a weighted least trimmed squares approach, aiming at improving the robustness of the algorithm. We use a patient-specific model discretized with the finite element method in order to ensure a realistic mechanical behavior of the brain tissue. To meet the clinical time constraint, we parallelized the slowest step of the algorithm so that we can perform a full 3-D image registration in 35 s (including the image update time) on a heterogeneous cluster of 15 personal computers. The algorithm has been tested on six cases of brain tumor resection, presenting a brain shift of up to 14 mm. The results show a good ability to recover large displacements, and a limited decrease of accuracy near the tumor resection cavity.

We propose a new model to simulate the three-dimensional (3-D) growth of glioblastomas multiforma (GBMs), the most aggressive glial tumors. The GBM speed of growth depends on the invaded tissue: faster in white than in gray matter, it is stopped by the dura or the ventricles. These different structures are introduced into the model using an atlas matching technique. The atlas includes both the segmentations of anatomical structures and diffusion information in white matter fibers. We use the finite element method (FEM) to simulate the invasion of the GBM in the brain parenchyma and its mechanical interaction with the invaded structures (mass effect). Depending on the considered tissue, the former effect is modeled with a reaction-diffusion or a Gompertz equation, while the latter is based on a linear elastic brain constitutive equation. In addition, we propose a new coupling equation taking into account the mechanical influence of the tumor cells on the invaded tissues. The tumor growth simulation is assessed by comparing the in-silico GBM growth with the real growth observed on two magnetic resonance images (MRIs) of a patient acquired with 6 mo difference. Results show the feasibility of this new conceptual approach and justifies its further evaluation.

The watershed transform has interesting properties that make it useful for many different image segmentation applications: it is simple and intuitive, can be parallelized, and always produces a complete division of the image. However, when applied to medical image analysis, it has important drawbacks (oversegmentation, sensitivity to noise, poor detection of thin or low signal to noise ratio structures). We present an improvement to the watershed transform that enables the introduction of prior information in its calculation. We propose to introduce this information via the use of a previous probability calculation. Furthermore, we introduce a method to combine the watershed transform and atlas registration, through the use of markers. We have applied our new algorithm to two challenging applications: knee cartilage and gray matter/white matter segmentation in MR images. Numerical validation of the results is provided, demonstrating the strength of the algorithm for medical image segmentation.

Characterizing the performance of image segmentation approaches has been a persistent challenge. Performance analysis is important since segmentation algorithms often have limited accuracy and precision. Interactive drawing of the desired segmentation by human raters has often been the only acceptable approach, and yet suffers from intra-rater and inter-rater variability. Automated algorithms have been sought in order to remove the variability introduced by raters, but such algorithms must be assessed to ensure they are suitable for the task. The performance of raters (human or algorithmic) generating segmentations of medical images has been difficult to quantify because of the difficulty of obtaining or estimating a known true segmentation for clinical data. Although physical and digital phantoms can be constructed for which ground truth is known or readily estimated, such phantoms do not fully reflect clinical images due to the difficulty of constructing phantoms which reproduce the full range of imaging characteristics and normal and pathological anatomical variability observed in clinical data. Comparison to a collection of segmentations by raters is an attractive alternative since it can be carried out directly on the relevant clinical imaging data. However, the most appropriate measure or set of measures with which to compare such segmentations has not been clarified and several measures are used in practice. We present here an expectation-maximization algorithm for simultaneous truth and performance level estimation (STAPLE). The algorithm considers a collection of segmentations and computes a probabilistic estimate of the true segmentation and a measure of the performance level represented by each segmentation. The source of each segmentation in the collection may be an appropriately trained human rater or raters, or may be an automated segmentation algorithm. The probabilistic estimate of the true segmentation is formed by estimating an optimal combination of the segmentations, weighting each segmentation depending upon the estimated performance level, and incorporating a prior model for the spatial distribution of structures being segmented as well as spatial homogeneity constraints. STAPLE is straightforward to apply to clinical imaging data, it readily enables assessment of the performance of an automated image segmentation algorithm, and enables direct comparison of human rater and algorithm performance.

In this paper, we propose a new approach to compute geodesic distance transformations in arbitrary 2D and 3D domains. The distance transformation proposed here is robust and has proved to have a computational complexity linear in the domain size. Our scheme is based on a new technique which we call occlusion points propagation, and with a higher accuracy than other geodesic distance transformations proposed before. We validate the algorithm with a set of synthetic domains, and we also make comparisons with two similar algorithms called B/sub d/-geodesic distance transformation and B/sub d/-geodesic distance transformation with circular propagation.

A novel algorithm to compute Voronoi diagrams of order k in arbitrary 2D and 3D domains is proposed. The algorithm is based on a fast ordered propagation distance transformation called occlusion points propagation geodesic distance transformation (OPPGDT) which is robust and linear in the domain size, and has higher accuracy than other geodesic distance transformations published before. Our approach has proved to have a computational complexity of order O(k.m) with m the domain size and k the order of the diagram. Voronoi diagrams have been extensively used in many areas and we show here that Voronoi diagrams computed in nonconvex domains, are extremely useful for the segmentation of medical images. We validated our algorithm with a set of 2D and 3D synthetic nonconvex domains, and with the segmentation of a medical dataset showing its robustness and performance.

We present a new algorithm for the nonrigid registration of three-dimensional magnetic resonance (MR) intraoperative image sequences showing brain shift. The algorithm tracks key surfaces of objects (cortical surface and the lateral ventricles) in the image sequence using a deformable surface matching algorithm. The volumetric deformation field of the objects is then inferred from the displacements at the boundary surfaces using a linear elastic biomechanical finite-element model. Two experiments on synthetic image sequences are presented, as well as an initial experiment on intraoperative MR images showing brain shift. The results of the registration algorithm show a good correlation of the internal brain structures after deformation, and a good capability of measuring surface as well as subsurface shift. We measured distances between landmarks in the deformed initial image and the corresponding landmarks in the target scan. Cortical surface shifts of up to 10 mm and subsurface shifts of up to 6 mm were recovered with an accuracy of 1 mm or less and 3 mm or less respectively.

We aimed to study the performance of a parallel implementation of an intraoperative nonrigid registration algorithm that accurately simulates the biomechanical properties of the brain and its deformations during surgery. The algorithm was designed to allow for improved surgical navigation and quantitative monitoring of treatment progress in order to improve the surgical outcome and to reduce the time required in the operating room. We have applied the algorithm to two neurosurgery cases with promising results. High performance computing is a key enabling technology that allows the biomechanical simulation to be executed quickly enough for the algorithm to be practical. Our parallel implementation was evaluated on a symmetric multi-processor and two clusters and exhibited similar performance characteristics on each. The implementation was sufficiently fast to be used in the operating room during a neurosurgery procedure. It allowed a three-dimensional volumetric deformation to be simulated in less than ten seconds.

Image-guided surgery is an application for which high performance computing is increasingly becoming a critical technology. Advances in image-guided surgery techniques have made it possible to acquire images of a patient whilst the surgery is taking place, to align these images with high resolution 3D scans of the patient acquired preoperatively and to merge intraoperative images from multiple imaging modalities. The application of these technologies has now become a routine clinical procedure in some hospitals. However, as the type of procedures undertaken is expanded, it is becoming clear that the use of image fusion and linear registration technology alone has some limitations. We have developed a novel image segmentation algorithm that makes use of an individualized template of normal patient anatomy in order to compute the segmentation of intraoperative imaging data. Intraoperative image segmentation is highly data and compute intensive. In order to achieve accurate segmentation in a time frame compatible with surgical intervention, we have developed a parallel version of our segmentation algorithm, and implemented the algorithm on a symmetric multiprocessor architecture. We have studied the accuracy of the segmentation algorithm, and the scalability and bandwidth requirements of our parallel implementation.