Low rank subspace clustering (LRSC)

We consider the problem of fitting a union of subspaces to a collection of data points drawn from one or more subspaces and corrupted by noise and/or gross errors. We pose this problem as a non-convex optimization problem, where the goal is to decompose the corrupted data matrix as the sum of a clean and self-expressive dictionary plus a matrix of noise and/or gross errors. By self-expressive we mean a dictionary whose atoms can be expressed as linear combinations of themselves with low-rank coefficients. In the case of noisy data, our key contribution is to show that this non-convex matrix decomposition problem can be solved in closed form from the SVD of the noisy data matrix. The solution involves a novel polynomial thresholding operator on the singular values of the data matrix, which requires minimal shrinkage. For one subspace, a particular case of our framework leads to classical PCA, which requires no shrinkage. For multiple subspaces, the low-rank coefficients obtained by our framework can be used to construct a data affinity matrix from which the clustering of the data according to the subspaces can be obtained by spectral clustering. In the case of data corrupted by gross errors, we solve the problem using an alternating minimization approach, which combines our polynomial thresholding operator with the more traditional shrinkage-thresholding operator. Experiments on motion segmentation and face clustering show that our framework performs on par with state-of-the-art techniques at a reduced computational cost.

Searching for the majority: algorithms of voluntary control.

Voluntary control of information processing is crucial to allocate resources and prioritize the processes that are most important under a given situation; the algorithms underlying such control, however, are often not clear. We investigated possible algorithms of control for the performance of the majority function, in which participants searched for and identified one of two alternative categories (left or right pointing arrows) as composing the majority in each stimulus set. We manipulated the amount (set size of 1, 3, and 5) and content (ratio of left and right pointing arrows within a set) of the inputs to test competing hypotheses regarding mental operations for information processing. Using a novel measure based on computational load, we found that reaction time was best predicted by a grouping search algorithm as compared to alternative algorithms (i.e., exhaustive or self-terminating search). The grouping search algorithm involves sampling and resampling of the inputs before a decision is reached. These findings highlight the importance of investigating the implications of voluntary control via algorithms of mental operations.

Spatial modulation of primate inferotemporal responses by eye position.

BACKGROUND: A key aspect of representations for object recognition and scene analysis in the ventral visual stream is the spatial frame of reference, be it a viewer-centered, object-centered, or scene-based coordinate system. Coordinate transforms from retinocentric space to other reference frames involve combining neural visual responses with extraretinal postural information. METHODOLOGY/PRINCIPAL FINDINGS: We examined whether such spatial information is available to anterior inferotemporal (AIT) neurons in the macaque monkey by measuring the effect of eye position on responses to a set of simple 2D shapes. We report, for the first time, a significant eye position effect in over 40% of recorded neurons with small gaze angle shifts from central fixation. Although eye position modulates responses, it does not change shape selectivity. CONCLUSIONS/SIGNIFICANCE: These data demonstrate that spatial information is available in AIT for the representation of objects and scenes within a non-retinocentric frame of reference. More generally, the availability of spatial information in AIT calls into questions the classic dichotomy in visual processing that associates object shape processing with ventral structures such as AIT but places spatial processing in a separate anatomical stream projecting to dorsal structures.

Generalized fuzzy clustering for segmentation of multi-spectral magnetic resonance images.

An integrated approach for multi-spectral segmentation of MR images is presented. This method is based on the fuzzy c-means (FCM) and includes bias field correction and contextual constraints over spatial intensity distribution and accounts for the non-spherical cluster's shape in the feature space. The bias field is modeled as a linear combination of smooth polynomial basis functions for fast computation in the clustering iterations. Regularization terms for the neighborhood continuity of intensity are added into the FCM cost functions. To reduce the computational complexity, the contextual regularizations are separated from the clustering iterations. Since the feature space is not isotropic, distance measure adopted in Gustafson-Kessel (G-K) algorithm is used instead of the Euclidean distance, to account for the non-spherical shape of the clusters in the feature space. These algorithms are quantitatively evaluated on MR brain images using the similarity measures.

Improved cerebellar tissue classification on magnetic resonance images of brain.

PURPOSE: To develop and implement a method for improved cerebellar tissue classification on the MRI of brain by automatically isolating the cerebellum prior to segmentation. MATERIALS AND METHODS: Dual fast spin echo (FSE) and fluid attenuation inversion recovery (FLAIR) images were acquired on 18 normal volunteers on a 3 T Philips scanner. The cerebellum was isolated from the rest of the brain using a symmetric inverse consistent nonlinear registration of individual brain with the parcellated template. The cerebellum was then separated by masking the anatomical image with individual FLAIR images. Tissues in both the cerebellum and rest of the brain were separately classified using hidden Markov random field (HMRF), a parametric method, and then combined to obtain tissue classification of the whole brain. The proposed method for tissue classification on real MR brain images was evaluated subjectively by two experts. The segmentation results on Brainweb images with varying noise and intensity nonuniformity levels were quantitatively compared with the ground truth by computing the Dice similarity indices. RESULTS: The proposed method significantly improved the cerebellar tissue classification on all normal volunteers included in this study without compromising the classification in remaining part of the brain. The average similarity indices for gray matter (GM) and white matter (WM) in the cerebellum are 89.81 (+/-2.34) and 93.04 (+/-2.41), demonstrating excellent performance of the proposed methodology. CONCLUSION: The proposed method significantly improved tissue classification in the cerebellum. The GM was overestimated when segmentation was performed on the whole brain as a single object.

Relationship between threshold and suprathreshold perception of position and stereoscopic depth.

We seek to determine the relationship between threshold and suprathreshold perception for position offset and stereoscopic depth perception under conditions that elevate their respective thresholds. Two threshold-elevating conditions were used: (1) increasing the interline gap and (2) dioptric blur. Although increasing the interline gap increases position (Vernier) offset and stereoscopic disparity thresholds substantially, the perception of suprathreshold position offset and stereoscopic depth remains unchanged. Perception of suprathreshold position offset also remains unchanged when the Vernier threshold is elevated by dioptric blur. We show that such normalization of suprathreshold position offset can be attributed to the topographical-map-based encoding of position. On the other hand, dioptric blur increases the stereoscopic disparity thresholds and reduces the perceived suprathreshold stereoscopic depth, which can be accounted for by a disparity-computation model in which the activities of absolute disparity encoders are multiplied by a Gaussian weighting function that is centered on the horopter. Overall, the statement "equal suprathreshold perception occurs in threshold-elevated and unelevated conditions when the stimuli are equally above their corresponding thresholds" describes the results better than the statement "suprathreshold stimuli are perceived as equal when they are equal multiples of their respective threshold values."

Symmetric inverse consistent nonlinear registration driven by mutual information.

A nonlinear viscoelastic image registration algorithm based on the demons paradigm and incorporating inverse consistent constraint (ICC) is implemented. An inverse consistent and symmetric cost function using mutual information (MI) as a similarity measure is employed. The cost function also includes regularization of transformation and inverse consistent error (ICE). The uncertainties in balancing various terms in the cost function are avoided by alternatively minimizing the similarity measure, the regularization of the transformation, and the ICE terms. The diffeomorphism of registration for preventing folding and/or tearing in the deformation is achieved by the composition scheme. The quality of image registration is first demonstrated by constructing brain atlas from 20 adult brains (age range 30-60). It is shown that with this registration technique: (1) the Jacobian determinant is positive for all voxels and (2) the average ICE is around 0.004 voxels with a maximum value below 0.1 voxels. Further, the deformation-based segmentation on Internet Brain Segmentation Repository, a publicly available dataset, has yielded high Dice similarity index (DSI) of 94.7% for the cerebellum and 74.7% for the hippocampus, attesting to the quality of our registration method.

Computational identification and functional validation of regulatory motifs in cartilage-expressed genes.

Chondrocyte gene regulation is important for the generation and maintenance of cartilage tissues. Several regulatory factors have been identified that play a role in chondrogenesis, including the positive transacting factors of the SOX family such as SOX9, SOX5, and SOX6, as well as negative transacting factors such as C/EBP and delta EF1. However, a complete understanding of the intricate regulatory network that governs the tissue-specific expression of cartilage genes is not yet available. We have taken a computational approach to identify cis-regulatory, transcription factor (TF) binding motifs in a set of cartilage characteristic genes to better define the transcriptional regulatory networks that regulate chondrogenesis. Our computational methods have identified several TFs, whose binding profiles are available in the TRANSFAC database, as important to chondrogenesis. In addition, a cartilage-specific SOX-binding profile was constructed and used to identify both known, and novel, functional paired SOX-binding motifs in chondrocyte genes. Using DNA pattern-recognition algorithms, we have also identified cis-regulatory elements for unknown TFs. We have validated our computational predictions through mutational analyses in cell transfection experiments. One novel regulatory motif, N1, found at high frequency in the COL2A1 promoter, was found to bind to chondrocyte nuclear proteins. Mutational analyses suggest that this motif binds a repressive factor that regulates basal levels of the COL2A1 promoter.

Systemic levels of the anti-inflammatory decoy receptor soluble RAGE (receptor for advanced glycation end products) are decreased in dogs with inflammatory bowel disease.

Inflammatory bowel disease (IBD) is a common condition in dogs, and a dysregulated innate immunity is believed to play a major role in its pathogenesis. S100A12 is an endogenous damage-associated molecular pattern molecule, which is involved in phagocyte activation and is increased in serum/fecal samples from dogs with IBD. S100A12 binds to the receptor of advanced glycation end products (RAGE), a pattern-recognition receptor, and results of studies in human patients with IBD and other conditions suggest a role of RAGE in chronic inflammation. Soluble RAGE (sRAGE), a decoy receptor for inflammatory proteins (e.g., S100A12) that appears to function as an anti-inflammatory molecule, was shown to be decreased in human IBD patients. This study aimed to evaluate serum sRAGE and serum/fecal S100A12 concentrations in dogs with IBD. Serum and fecal samples were collected from 20 dogs with IBD before and after initiation of medical treatment and from 15 healthy control dogs. Serum sRAGE and serum and fecal S100A12 concentrations were measured by ELISA, and were compared between dogs with IBD and healthy controls, and between dogs with a positive outcome (i.e., clinical remission, n=13) and those that were euthanized (n=6). The relationship of serum sRAGE concentrations with clinical disease activity (using the CIBDAI scoring system), serum and fecal S100A12 concentrations, and histologic disease severity (using a 4-point semi-quantitative grading system) was tested. Serum sRAGE concentrations were significantly lower in dogs with IBD than in healthy controls (p=0.0003), but were not correlated with the severity of histologic lesions (p=0.4241), the CIBDAI score before (p=0.0967) or after treatment (p=0.1067), the serum S100A12 concentration before (p=0.9214) and after treatment (p=0.4411), or with the individual outcome (p=0.4066). Clinical remission and the change in serum sRAGE concentration after treatment were not significantly associated (p=0.5727); however, serum sRAGE concentrations increased only in IBD dogs with complete clinical remission. Also, dogs that were euthanized had significantly higher fecal S100A12 concentrations than dogs that were alive at the end of the study (p=0.0124). This study showed that serum sRAGE concentrations are decreased in dogs diagnosed with IBD compared to healthy dogs, suggesting that sRAGE/RAGE may be involved in the pathogenesis of canine IBD. Lack of correlation between sRAGE and S100A12 concentrations is consistent with sRAGE functioning as a non-specific decoy receptor. Further studies need to evaluate the gastrointestinal mucosal expression of RAGE in healthy and diseased dogs, and also the formation of S100A12-RAGE complexes.

Novel antiviral properties of azithromycin in cystic fibrosis airway epithelial cells.

Virus-associated pulmonary exacerbations, often associated with rhinoviruses (RVs), contribute to cystic fibrosis (CF) morbidity. Currently, there are only a few therapeutic options to treat virus-induced CF pulmonary exacerbations. The macrolide antibiotic azithromycin has antiviral properties in human bronchial epithelial cells. We investigated the potential of azithromycin to induce antiviral mechanisms in CF bronchial epithelial cells. Primary bronchial epithelial cells from CF and control children were infected with RV after azithromycin pre-treatment. Viral RNA, interferon (IFN), IFN-stimulated gene and pattern recognition receptor expression were measured by real-time quantitative PCR. Live virus shedding was assessed by assaying the 50% tissue culture infective dose. Pro-inflammatory cytokine and IFN-β production were evaluated by ELISA. Cell death was investigated by flow cytometry. RV replication was increased in CF compared with control cells. Azithromycin reduced RV replication seven-fold in CF cells without inducing cell death. Furthermore, azithromycin increased RV-induced pattern recognition receptor, IFN and IFN-stimulated gene mRNA levels. While stimulating antiviral responses, azithromycin did not prevent virus-induced pro-inflammatory responses. Azithromycin pre-treatment reduces RV replication in CF bronchial epithelial cells, possibly through the amplification of the antiviral response mediated by the IFN pathway. Clinical studies are needed to elucidate the potential of azithromycin in the management and prevention of RV-induced CF pulmonary exacerbations.

Prolonged activity of the pestiviral RNase Erns as an interferon antagonist after uptake by clathrin-mediated endocytosis

The RNase activity of the envelope glycoprotein E(rns) of the pestivirus bovine viral diarrhea virus (BVDV) is required to block type I interferon (IFN) synthesis induced by single-stranded RNA (ssRNA) and double-stranded RNA (dsRNA) in bovine cells. Due to the presence of an unusual membrane anchor at its C terminus, a significant portion of E(rns) is also secreted. In addition, a binding site for cell surface glycosaminoglycans is located within the C-terminal region of E(rns). Here, we show that the activity of soluble E(rns) as an IFN antagonist is not restricted to bovine cells. Extracellularly applied E(rns) protein bound to cell surface glycosaminoglycans and was internalized into the cells within 1 h of incubation by an energy-dependent mechanism that could be blocked by inhibitors of clathrin-dependent endocytosis. E(rns) mutants that lacked the C-terminal membrane anchor retained RNase activity but lost most of their intracellular activity as an IFN antagonist. Surprisingly, once taken up into the cells, E(rns) remained active and blocked dsRNA-induced IFN synthesis for several days. Thus, we propose that E(rns) acts as an enzymatically active decoy receptor that degrades extracellularly added viral RNA mainly in endolysosomal compartments that might otherwise activate intracellular pattern recognition receptors (PRRs) in order to maintain a state of innate immunotolerance. IMPORTANCE The pestiviral RNase E(rns) was previously shown to inhibit viral ssRNA- and dsRNA-induced interferon (IFN) synthesis. However, the localization of E(rns) at or inside the cells, its species specificity, and its mechanism of interaction with cell membranes in order to block the host's innate immune response are still largely unknown. Here, we provide strong evidence that the pestiviral RNase E(rns) is taken up within minutes by clathrin-mediated endocytosis and that this uptake is mostly dependent on the glycosaminoglycan binding site located within the C-terminal end of the protein. Remarkably, the inhibitory activity of E(rns) remains for several days, indicating the very potent and prolonged effect of a viral IFN antagonist. This novel mechanism of an enzymatically active decoy receptor that degrades a major viral pathogen-associated molecular pattern (PAMP) might be required to efficiently maintain innate and, thus, also adaptive immunotolerance, and it might well be relevant beyond the bovine species.

Total Variation Blind Deconvolution: The Devil Is in the Details

In this paper we study the problem of blind deconvolution. Our analysis is based on the algorithm of Chan and Wong [2] which popularized the use of sparse gradient priors via total variation. We use this algorithm because many methods in the literature are essentially adaptations of this framework. Such algorithm is an iterative alternating energy minimization where at each step either the sharp image or the blur function are reconstructed. Recent work of Levin et al. [14] showed that any algorithm that tries to minimize that same energy would fail, as the desired solution has a higher energy than the no-blur solution, where the sharp image is the blurry input and the blur is a Dirac delta. However, experimentally one can observe that Chan and Wong's algorithm converges to the desired solution even when initialized with the no-blur one. We provide both analysis and experiments to resolve this paradoxical conundrum. We find that both claims are right. The key to understanding how this is possible lies in the details of Chan and Wong's implementation and in how seemingly harmless choices result in dramatic effects. Our analysis reveals that the delayed scaling (normalization) in the iterative step of the blur kernel is fundamental to the convergence of the algorithm. This then results in a procedure that eludes the no-blur solution, despite it being a global minimum of the original energy. We introduce an adaptation of this algorithm and show that, in spite of its extreme simplicity, it is very robust and achieves a performance comparable to the state of the art.

Optimized aperture shapes for depth estimation

The finite depth of field of a real camera can be used to estimate the depth structure of a scene. The distance of an object from the plane in focus determines the defocus blur size. The shape of the blur depends on the shape of the aperture. The blur shape can be designed by masking the main lens aperture. In fact, aperture shapes different from the standard circular aperture give improved accuracy of depth estimation from defocus blur. We introduce an intuitive criterion to design aperture patterns for depth from defocus. The criterion is independent of a specific depth estimation algorithm. We formulate our design criterion by imposing constraints directly in the data domain and optimize the amount of depth information carried by blurred images. Our criterion is a quadratic function of the aperture transmission values. As such, it can be numerically evaluated to estimate optimized aperture patterns quickly. The proposed mask optimization procedure is applicable to different depth estimation scenarios. We use it for depth estimation from two images with different focus settings, for depth estimation from two images with different aperture shapes as well as for depth estimation from a single coded aperture image. In this work we show masks obtained with this new evaluation criterion and test their depth discrimination capability using a state-of-the-art depth estimation algorithm.