Embedded restricted Boltzmann machines for fusion of mixed data types and applications in social measurements analysis

Analysis and fusion of social measurements is important to understand what shapes the public’s opinion and the sustainability of the global development. However, modeling data collected from social responses is challenging as the data is typically complex and heterogeneous, which might take the form of stated facts, subjective assessment, choices, preferences or any combination thereof. Model-wise, these responses are a mixture of data types including binary, categorical, multicategorical, continuous, ordinal, count and rank data. The challenge is therefore to effectively handle mixed data in the a unified fusion framework in order to perform inference and analysis. To that end, this paper introduces eRBM (Embedded Restricted Boltzmann Machine) – a probabilistic latent variable model that can represent mixed data using a layer of hidden variables transparent across different types of data. The proposed model can comfortably support largescale data analysis tasks, including distribution modelling, data completion, prediction and visualisation. We demonstrate these versatile features on several moderate and large-scale publicly available social survey datasets.

Metabolite profiles in the anterior cingulate cortex of depressed patients differentiate those taking N-acetyl-cysteine versus placebo

Background: Increased oxidative stress is thought to contribute to the pathophysiology of major depressive disorder (MDD), which is in part due to diminished levels of glutathione, the primary anti-oxidant of the brain. Oral administration of N-acetyl-cysteine (NAC) replenishes glutathione and has therefore been shown to reduce depressive symptoms. Proton magnetic spectroscopy (1H-MRS) that allows quantification of brain metabolites pertinent to both MDD and oxidative biology may provide some novel insights into the neurobiological effects of NAC, and in particular metabolite concentrations within the anterior cingulate cortex (ACC) are likely to be important given the key role of this region in the regulation of affect.

Objective: The aim of this study was to determine whether the metabolite profile of the ACC in MDD patients predicts treatment with adjunctive NAC versus placebo.

Methods: This study was nested within a multicentre, randomized, double-blind, placebo-controlled study of MDD participants treated with adjunctive NAC. Participants (n = 76) from one site completed the spectroscopy component at the end of treatment (12 weeks). Spectra from a single-voxel in the ACC were acquired and absolute concentrations of glutamate (Glu), glutamate-glutamine (Glx), N-acetyl-aspartate (NAA) and myo-inositol (mI) were obtained. Binary logistic regression analysis was performed to determine whether metabolite profiles could predict NAC versus placebo group membership.

Results: When predicting group outcome (NAC or placebo), Glx, NAA and mI were a significant model, and had 75% accuracy, while controlling for depression severity and sex. However, the Glu, NAA and mI profile was only predictive at a trend level, with 68.3% accuracy. For both models, the log of the odds of a participant being in the NAC group was positively related to NAA, Glx and Glu levels and negatively related to mI levels.

Conclusion: The finding of higher Glx and NAA levels being predictive of the NAC group provides preliminary support for the putative anti-oxidative role of NAC in MDD.

Measuring the capacity of image fusion

Multisource image fusion is usually achieved by repeatedly fusing source images in pairs. However, there is no guarantee on the delivered quality considering the amount of information to be squeezed into the same spatial dimension. This paper presents a fusion capacity measure and examines the limits at which fusing more images will not add further information. The fusion capacity index employs Mutual Information (MI) to measure how far the histogram of the examined image is from a uniformly distributed histogram of a saturated image.

Combined joint fusion for index and middle carpometacarpal instability in elite boxers

Isolated instability of the index and middle finger carpometacarpal joints is uncommon. An unpublished injury pattern in a consecutive series of 13 elite boxers is described, with mid-term radiological and functional results of two novel surgical treatments. All the boxers (from Australia and the UK) were unable to compete owing to pain and weakness. Four boxers were managed initially by open reduction and temporary K-wire fixation. Nine boxers underwent primary carpometacarpal arthrodesis. All were able to return to their previous level of competition. One boxer who had undergone a soft tissue reconstruction competed at international level but required an arthrodesis because of recurrent symptoms. In elite boxers, simple reduction and wiring may be appropriate for an acute injury causing index or middle finger carpometacarpal joint instability, however, arthrodesis is the treatment of choice when instability and degenerative changes are present.

Image fusion metrics : evolution in a nutshell

Image fusion process merges two images into a single more informative image. Objective image fusion per- formance metrics rely primarily on measuring the amount of information transferred from each source image into the fused image. Objective image fusion metrics have evolved from image processing dissimilarity metrics. Additionally, researchers have developed many additions to image dissimilarity metrics in order to better value the local fusion worthy features in source images. This paper studies the evolution of objective image fusion performance metrics and their subjective and objective validation. It describes how a fusion performance metric evolves starting with image dissimilarity metrics, its realization into image fusion contexts, its localized weighting factors and the validation process.

Fusion

 Solo exhibition of art works

Liza McCosh: fusion

Exhibition catalogue with introduction, essay, biography and images

Multi-sensory face biometric fusion (for personal identification)

The objective of this work is to recognize faces using sets of images in visual and thermal spectra. This is challenging because the former is greatly affected by illumination changes, while the latter frequently contains occlusions due to eye-wear and is inherently less discriminative. Our method is based on a fusion of the two modalities. Specifically: we examine (i) the effects of preprocessing of data in each domain, (ii) the fusion of holistic and local facial appearance, and (iii) propose an algorithm for combining the similarity scores in visual and thermal spectra in the presence of prescription glasses and significant pose variations, using a small number of training images (5-7). Our system achieved a high correct identification rate of 97% on a freely available test set of 29 individuals and extreme illumination changes.

Versican processing by a disintegrin-like and metalloproteinase domain with thrombospondin-1 repeats proteinases-5 and -15 facilitates myoblast fusion

Skeletal muscle development and regeneration requires the fusion of myoblasts into multinucleated myotubes. Because the enzymatic proteolysis of a hyaluronan and versican-rich matrix by ADAMTS versicanases is required for developmental morphogenesis, we hypothesized that the clearance of versican may facilitate the fusion of myoblasts during myogenesis. Here, we used transgenic mice and an in vitro model of myoblast fusion, C2C12 cells, to determine a potential role for ADAMTS versicanases. Versican processing was observed during in vivo myogenesis at the time when myoblasts were fusing to form multinucleated myotubes. Relevant ADAMTS genes, chief among them Adamts5 and Adamts15, were expressed both in developing embryonic muscle and differentiating C2C12 cells. Reducing the levels of Adamts5 mRNA in vitro impaired myoblast fusion, which could be rescued with catalytically active but not the inactive forms of ADAMTS5 or ADAMTS15. The addition of inactive ADAMTS5, ADAMTS15, or full-length V1 versican effectively impaired myoblast fusion. Finally, the expansion of a hyaluronan and versican-rich matrix was observed upon reducing the levels of Adamts5 mRNA in myoblasts. These data indicate that these ADAMTS proteinases contribute to the formation of multinucleated myotubes such as is necessary for both skeletal muscle development and during regeneration, by remodeling a versican-rich pericellular matrix of myoblasts. Our study identifies a possible pathway to target for the improvement of myogenesis in a plethora of diseases including cancer cachexia, sarcopenia, and muscular dystrophy.