Image Analysis

Functional properties of a neuron are coupled with its morphology, particularly the morphology of dendritic spines. Spine volume has been used as the primary morphological parameter in order the characterize the structure and function coupling. However, this reductionist approach neglects the rich shape repertoire of dendritic spines. First step to incorporate spine shape information into functional coupling is classifying main spine shapes that were proposed in the literature. Due to the lack of reliable and fully automatic tools to analyze the morphology of the spines, such analysis is often performed manually, which is a laborious and time intensive task and prone to subjectivity. In this paper we present an automated approach to extract features using basic image processing techniques, and classify spines into mushroom or stubby by applying machine learning algorithms. Out of 50 manually segmented mushroom and stubby spines, Support Vector Machine was able to classify 98% of the spines correctly.

The use of appearance and shape priors in image segmentation is known to improve accuracy; however, existing techniques have several drawbacks. Active shape and appearance models require landmark points and assume unimodal shape and appearance distributions. Level set based shape priors are limited to global shape similarity. In this paper, we present a novel shape and appearance priors for image segmentation based on an implicit parametric shape representation called disjunctive normal shape model (DNSM). DNSM is formed by disjunction of conjunctions of half-spaces defined by discriminants. We learn shape and appearance statistics at varying spatial scales using nonparametric density estimation. Our method can generate a rich set of shape variations by locally combining training shapes. Additionally, by studying the intensity and texture statistics around each discriminant of our shape model, we construct a local appearance probability map. Experiments carried out on both medical and natural image datasets show the potential of the proposed method.

To stimulate progress in automating the reconstruction of neural circuits, we organized the first international challenge on 2D segmentation of electron microscopic (EM) images of the brain. Participants submitted boundary maps predicted for a test set of images, and were scored based on their agreement with a consensus of human expert annotations. The winning team had no prior experience with EM images, and employed a convolutional network. This "deep learning" approach has since become accepted as a standard for segmentation of EM images. The challenge has continued to accept submissions, and the best so far has resulted from cooperation between two teams. The challenge has probably saturated, as algorithms cannot progress beyond limits set by ambiguities inherent in 2D scoring and the size of the test dataset. Retrospective evaluation of the challenge scoring system reveals that it was not sufficiently robust to variations in the widths of neurite borders. We propose a solution to this problem, which should be useful for a future 3D segmentation challenge.

Background
In the area of connectomics, there is a significant gap between the time required for data acquisition and dense reconstruction of the neural processes contained in the same dataset. Automatic methods are able to eliminate this timing gap, but the state-of-the-art accuracy so far is insufficient for use without user corrections. If completed naively, this process of correction can be tedious and time consuming.

New method
We present a new semi-automatic method that can be used to perform 3D segmentation of neurites in EM image stacks. It utilizes an automatic method that creates a hierarchical structure for recommended merges of superpixels. The user is then guided through each predicted region to quickly identify errors and establish correct links.

Results
We tested our method on three datasets with both novice and expert users. Accuracy and timing were compared with published automatic, semi-automatic, and manual results.

Comparison with existing methods
Post-automatic correction methods have also been used in Mishchenko et al. (2010) and Haehn et al. (2014). These methods do not provide navigation or suggestions in the manner we present. Other semi-automatic methods require user input prior to the automatic segmentation such as Jeong et al. (2009) and Cardona et al. (2010) and are inherently different than our method.

Conclusion
Using this method on the three datasets, novice users achieved accuracy exceeding state-of-the-art automatic results, and expert users achieved accuracy on par with full manual labeling but with a 70% time improvement when compared with other examples in publication.

The local field potential (LFP) is of growing importance in neurophysiology as a metric of network activity and as a readout signal for use in brain-machine interfaces. However, there are uncertainties regarding the kind and visual field extent of information carried by LFP signals, as well as the specific features of the LFP signal conveying such information, especially under naturalistic conditions. To address these questions, we recorded LFP responses to natural images in V1 of awake and anesthetized macaques using Utah multielectrode arrays. First, we have shown that it is possible to identify presented natural images from the LFP responses they evoke using trained Gabor wavelet (GW) models. Because GW models were devised to explain the spiking responses of V1 cells, this finding suggests that local spiking activity and LFPs (thought to reflect primarily local synaptic activity) carry similar visual information. Second, models trained on scalar metrics, such as the evoked LFP response range, provide robust image identification, supporting the informative nature of even simple LFP features. Third, image identification is robust only for the first 300 ms following image presentation, and image information is not restricted to any of the spectral bands. This suggests that the short-latency broadband LFP response carries most information during natural scene viewing. Finally, best image identification was achieved by GW models incorporating information at the scale of ∼0.5° in size and trained using four different orientations. This suggests that during natural image viewing, LFPs carry stimulus-specific information at spatial scales corresponding to few orientation columns in macaque V1.

We introduce a novel general regression model that is based on a linear combination of a new set of non-local basis functions that forms an effective feature space. We propose a training algorithm that learns all the model parameters simultaneously and offer an initialization scheme for parameters of the basis functions. We show through several experiments that the proposed method offers better coverage for high-dimensional space compared to local Gaussian basis functions and provides competitive performance in comparison to other state-of-the-art regression methods.

We develop a novel supervised learning/classification method, called disjunctive normal random forest (DNRF). A DNRF is an ensemble of randomly trained disjunctive normal decision trees (DNDT). To construct a DNDT, we formulate each decision tree in the random forest as a disjunction of rules, which are conjunctions of Boolean functions. We then approximate this disjunction of conjunctions with a differentiable function and approach the learning process as a risk minimization problem that incorporates the classification error into a single global objective function. The minimization problem is solved using gradient descent. DNRFs are able to learn complex decision boundaries and achieve low generalization error. We present experimental results demonstrating the improved performance of DNDTs and DNRFs over conventional decision trees and random forests. We also show the superior performance of DNRFs over state-of-the-art classification methods on benchmark datasets.

The association between developmental trajectories of language-related white matter fiber pathways from 6 to 24 months of age and individual differences in language production at 24 months of age was investigated. The splenium of the corpus callosum, a fiber pathway projecting through the posterior hub of the default mode network to occipital visual areas, was examined as well as pathways implicated in language function in the mature brain, including the arcuate fasciculi, uncinate fasciculi, and inferior longitudinal fasciculi. The hypothesis that the development of neural circuitry supporting domain-general orienting skills would relate to later language performance was tested in a large sample of typically developing infants. The present study included 77 infants with diffusion weighted MRI scans at 6, 12 and 24 months and language assessment at 24 months. The rate of change in splenium development varied significantly as a function of language production, such that children with greater change in fractional anisotropy (FA) from 6 to 24 months produced more words at 24 months. Contrary to findings from older children and adults, significant associations between language production and FA in the arcuate, uncinate, or left inferior longitudinal fasciculi were not observed. The current study highlights the importance of tracing brain development trajectories from infancy to fully elucidate emerging brain–behavior associations while also emphasizing the role of the splenium as a key node in the structural network that supports the acquisition of spoken language.

The visualization of variability in 3D shapes or surfaces, which is a type of ensemble uncertainty visualization for volume data, provides a means of understanding the underlying distribution for a collection or ensemble of surfaces. While ensemble visualization for surfaces is already described in the literature, we conduct an expert-based evaluation in a particular medical imaging application: the construction of atlases or templates from a population of images. In this work, we extend contour boxplots to 3D, allowing us to evaluate it against an enumeration-style visualization of the ensemble members and also other conventional visualizations used by atlas builders, namely examining the atlas image and the corresponding images/data provided as part of the construction process. We present feedback from domain experts on the efficacy of contour boxplots compared to other modalities when used as part of the atlas construction and analysis stages of their work.

Purpose
Statistical shape analysis of anatomical structures plays an important role in many medical image analysis applications such as understanding the structural changes in anatomy in various stages of growth or disease. Establishing accurate correspondence across object populations is essential for such statistical shape analysis studies.

Methods
In this paper, we present an entropy-based correspondence framework for computing point-based correspondence among populations of surfaces in a groupwise manner. This robust framework is parameterization-free and computationally efficient. We review the core principles of this method as well as various extensions to deal effectively with surfaces of complex geometry and application-driven correspondence metrics.

Results
We apply our method to synthetic and biological datasets to illustrate the concepts proposed and compare the performance of our framework to existing techniques.

Conclusions
Through the numerous extensions and variations presented here, we create a very flexible framework that can effectively handle objects of various topologies, multi-object complexes, open surfaces, and objects of complex geometry such as high-curvature regions or extremely thin features.

Computing image atlases that are representative of a dataset
is an important first step for statistical analysis of images. Most current approaches estimate a single atlas to represent the average of a large population of images, however, a single atlas is not sufficiently expressive to capture distributions of images with multiple modes. In this paper, we present a mixture model for building diffeomorphic multi-atlases that can represent sub-populations without knowing the category of each observed data point. In our probabilistic model, we treat diffeomorphic image transformations as latent variables, and integrate them out using a Monte Carlo Expectation Maximization (MCEM) algorithm via Hamiltonian Monte Carlo (HMC) sampling. A key benefit of our model is that the mixture modeling inference procedure results in an automatic clustering of the dataset. Using 2D synthetic data generated from known parameters, we demonstrate the ability of our model to successfully recover the multi-atlas and automatically cluster the dataset. We also show the effectiveness of the proposed method in a multi-atlas estimation problem for 3D brain images.

BACKGROUND AND PURPOSE
Diffusion tensor imaging (DTI) tractography reconstruction of white matter pathways can help guide brain tumor resection. However, DTI tracts are complex mathematical objects and the validity of tractography-derived information in clinical settings has yet to be fully established. To address this issue, we initiated the DTI Challenge, an international working group of clinicians and scientists whose goal was to provide standardized evaluation of tractography methods for neurosurgery. The purpose of this empirical study was to evaluate different tractography techniques in the first DTI Challenge workshop.

METHODS
Eight international teams from leading institutions reconstructed the pyramidal tract in four neurosurgical cases presenting with a glioma near the motor cortex. Tractography methods included deterministic, probabilistic, filtered, and global approaches. Standardized evaluation of the tracts consisted in the qualitative review of the pyramidal pathways by a panel of neurosurgeons and DTI experts and the quantitative evaluation of the degree of agreement among methods.

RESULTS
The evaluation of tractography reconstructions showed a great interalgorithm variability. Although most methods found projections of the pyramidal tract from the medial portion of the motor strip, only a few algorithms could trace the lateral projections from the hand, face, and tongue area. In addition, the structure of disagreement among methods was similar across hemispheres despite the anatomical distortions caused by pathological tissues.

CONCLUSIONS
The DTI Challenge provides a benchmark for the standardized evaluation of tractography methods on neurosurgical data. This study suggests that there are still limitations to the clinical use of tractography for neurosurgical decision making.

PURPOSE:
To evaluate the performance of an edge-based registration technique in correcting for respiratory motion artifacts in magnetic resonance renographic (MRR) data and to examine the efficiency of a semiautomatic software package in processing renographic data from a cohort of clinical patients.

MATERIALS AND METHODS:
The developed software incorporates an image-registration algorithm based on the generalized Hough transform of edge maps. It was used to estimate glomerular filtration rate (GFR), renal plasma flow (RPF), and mean transit time (MTT) from 36 patients who underwent free-breathing MRR at 3T using saturation-recovery turbo-FLASH. The processing time required for each patient was recorded. Renal parameter estimates and model-fitting residues from the software were compared to those from a previously reported technique. Interreader variability in the software was quantified by the standard deviation of parameter estimates among three readers. GFR estimates from our software were also compared to a reference standard from nuclear medicine.

RESULTS:
The time taken to process one patient's data with the software averaged 12 ± 4 minutes. The applied image registration effectively reduced motion artifacts in dynamic images by providing renal tracer-retention curves with significantly smaller fitting residues (P < 0.01) than unregistered data or data registered by the previously reported technique. Interreader variability was less than 10% for all parameters. GFR estimates from the proposed method showed greater concordance with reference values (P < 0.05).

CONCLUSION:
These results suggest that the proposed software can process MRR data efficiently and accurately. Its incorporated registration technique based on the generalized Hough transform effectively reduces respiratory motion artifacts in free-breathing renographic acquisitions. J. Magn. Reson. Imaging 2015.

In multiatlas segmentation, one typically registers several atlases to the novel image, and their respective segmented label images are transformed and fused to form the final segmentation. In this work, we provide a new dynamical system perspective for multiatlas segmentation, inspired by the following fact: The transformation that aligns the current atlas to the novel image can be not only computed by direct registration but also inferred from the transformation that aligns the previous atlas to the image together with the transformation between the two atlases. This process is similar to the global positioning system on a vehicle, which gets position by inquiring from the satellite and by employing the previous location and velocity—neither answer in isolation being perfect. To solve this problem, a dynamical system scheme is crucial to combine the two pieces of information; for example, a Kalman filtering scheme is used. Accordingly, in this work, a Kalman multiatlas segmentation is proposed to stabilize the global/affine registration step. The contributions of this work are twofold. First, it provides a new dynamical systematic perspective for standard independent multiatlas registrations, and it is solved by Kalman filtering. Second, with very little extra computation, it can be combined with most existing multiatlas segmentation schemes for better registration/segmentation accuracy.

Early brain development is characterized by rapid organization and structuring. Magnetic resonance–diffusion tensor imaging (MR-DTI) provides the possibility of capturing these changes noninvasively by following individuals longitudinally to better understand departures from normal brain development in subjects at risk for mental illness. This article illustrates the modeling of neurodevelopmental trajectories using a recently developed framework. Descriptions include the estimation of normative models for healthy singletons and twins and a statistical framework to predict development at 2 years of age only based on neonatal image data – a capability with excellent potential for preclinical diagnosis and eventual early therapeutic intervention.

Numerous brain imaging studies indicate that the corpus callosum is smaller in older children and adults with autism spectrum disorder. However, there are no published studies examining the morphological development of this connective pathway in infants at-risk for the disorder. Magnetic resonance imaging data were collected from 270 infants at high familial risk for autism spectrum disorder and 108 low-risk controls at 6, 12 and 24 months of age, with 83% of infants contributing two or more data points. Fifty-seven children met criteria for ASD based on clinical-best estimate diagnosis at age 2 years. Corpora callosa were measured for area, length and thickness by automated segmentation. We found significantly increased corpus callosum area and thickness in children with autism spectrum disorder starting at 6 months of age. These differences were particularly robust in the anterior corpus callosum at the 6 and 12 month time points. Regression analysis indicated that radial diffusivity in this region, measured by diffusion tensor imaging, inversely predicted thickness. Measures of area and thickness in the first year of life were correlated with repetitive behaviours at age 2 years. In contrast to work from older children and adults, our findings suggest that the corpus callosum may be larger in infants who go on to develop autism spectrum disorder. This result was apparent with or without adjustment for total brain volume. Although we did not see a significant interaction between group and age, cross-sectional data indicated that area and thickness differences diminish by age 2 years. Regression data incorporating diffusion tensor imaging suggest that microstructural properties of callosal white matter, which includes myelination and axon composition, may explain group differences in morphology.

Substance use during pregnancy and the postpartum period may have significant implications for both mother and the developing child. However, the neurobiological basis of the impact of substance use on parenting is less well understood. Here we examined the impact of maternal substance use on cortical gray matter (GM) and white matter volumes, and whether this was associated with individual differences in motivational systems of behavioral activation and inhibition. Mothers were included in the substance-using group if any addictive substance was used during pregnancy and/or in the immediate postpartum period (within 3 months of delivery). GM volume was reduced in substance-using mothers compared to non-substance-using mothers, particularly in frontal brain regions. In substance-using mothers, we also found that frontal GM was negatively correlated with levels of behavioral activation (i.e., the motivation to approach rewarding stimuli). This effect was absent in non-substance-using mothers. Taken together, these findings indicate a reduction in GM volume is associated with substance use, and that frontal GM volumetric differences may be related to approach motivation in substance-using mothers.

In this paper, we present a generative Bayesian approach for estimating the low-dimensional latent space of diffeomorphic shape variability in a population of images. We develop a latent variable model for principal geodesic analysis (PGA) that provides a probabilistic framework for factor analysis in the space of diffeomorphisms. A sparsity prior in the model results in automatic selection of the number of relevant dimensions by driving unnecessary principal geodesics to zero. To infer model parameters, including the image atlas, principal geodesic deformations, and the effective dimensionality, we introduce an expectation maximization (EM) algorithm. We evaluate our proposed model on 2D synthetic data and the 3D OASIS brain database of magnetic resonance images, and show that the automatically selected latent dimensions from our model are able to reconstruct unobserved testing images with lower error than both linear principal component analysis (LPCA) in the image space and tangent space principal component analysis (TPCA) in the diffeomorphism space.

Prenatal drug exposure, particularly prenatal cocaine exposure (PCE), incurs great public and scientific interest because of its associated neurodevelopmental consequences. However, the neural underpinnings of PCE remain essentially uncharted, and existing studies in school-aged children and adolescents are confounded greatly by postnatal environmental factors. In this study, leveraging a large neonate sample (N = 152) and non-invasive resting-state functional magnetic resonance imaging, we compared human infants with PCE comorbid with other drugs (such as nicotine, alcohol, marijuana, and antidepressant) with infants with similar non-cocaine poly drug exposure and drug-free controls. We aimed to characterize the neural correlates of PCE based on functional connectivity measurements of the amygdala and insula at the earliest stage of development. Our results revealed common drug exposure-related connectivity disruptions within the amygdala–frontal, insula–frontal, and insula–sensorimotor circuits. Moreover, a cocaine-specific effect was detected within a subregion of the amygdala–frontal network. This pathway is thought to play an important role in arousal regulation, which has been shown to be irregular in PCE infants and adolescents. These novel results provide the earliest human-based functional delineations of the neural-developmental consequences of prenatal drug exposure and thus open a new window for the advancement of effective strategies aimed at early risk identification and intervention.

This paper presents a fast geodesic shooting algorithm for diffeomorphic image registration. We first introduce a novel finite-dimensional Lie algebra structure on the space of bandlimited velocity fields. We then show that this space can effectively represent initial velocities for diffeomorphic image registration at much lower dimensions than typically used, with little to no loss in registration accuracy. We then leverage the fact that the geodesic evolution equations, as well as the adjoint Jacobi field equations needed for gradient descent methods, can be computed entirely in this finite-dimensional Lie algebra. The result is a geodesic shooting method for large deformation metric mapping (LDDMM) that is dramatically faster and less memory intensive than state-of-the-art methods. We demonstrate the effectiveness of our model to register 3D brain images and compare its registration accuracy, runtime, and memory consumption with leading LDDMM methods. We also show how our algorithm breaks through the prohibitive time and memory requirements of diffeomorphic atlas building.