Numerical analysis of Augmented Lagrangian algorithms in complementary elastoplasticity

The main subject of the paper is the investigation of Augmented Lagrangian algorithms and update formulas in the solution of elastoplastic problems. A stress rate formulation for elastoplastic models with internal variables and its finite increment form is employed to state the mechanical problem. In this formulation the Augmented Lagrangian is used to enforce the constraint of plastic admissibility directly on the stresses and thermodynamic forces. This is not a limitation of the Augmented Lagrangian approach, and the same framework can be built on more classical displacement formulations as well. The meaning and the derivation of various first and second order Lagrangian multipliers update formulas and iterative schemes is shown. A new diagonal iteration algorithm and the introduction of a scale factor for the Augmented Lagrangian term are proposed. Numerical examples compare the efficiency of several forms of Augmented Lagrangian algorithms and illustrate the influence of the scale factor and of the penalty parameter. Copyright © 2004 John Wiley & Sons, Ltd.     

Evaluation of Textile Bobbins Drying Processes: Experimental and Modelling Studies

Drying of textile bobbins was carried out on two kinds of laboratory-scale dryers: in a cross circulation drying rig, atmospheric pressure heating air is used, while in the through circulation drying apparatus, the heating air is pressurized before the drying process takes place. In both cases, a bobbin was fitted with seven thermocouples, placed in a plane normal to the bobbin axis but at different radial distances, in order to provide temperature data during the drying process. In cross circulation drying, plots of temperature against time indicated the presence of two moving fronts, one starting from inside the bobbin, another from the outside, while in through circulation drying, similar plots showed the presence of an evaporation front, moving from inside the bobbin to the outside. In both canes, numerical models were developed to simulate the drying processes.     

Early-Life Development of the Bifidobacterial Community in the Infant Gut

The establishment of the gut microbiota poses implications for short and long-term health. Bifidobacterium is an important taxon in early life, being one of the most abundant genera in the infant intestinal microbiota and carrying out key functions for maintaining host-homeostasis. Recent metagenomic studies have shown that different factors, such as gestational age, delivery mode, or feeding habits, affect the gut microbiota establishment at high phylogenetic levels. However, their impact on the specific bifidobacterial populations is not yet well understood. Here we studied the impact of these factors on the different Bifidobacterium species and subspecies at both the quantitative and qualitative levels. Fecal samples were taken from 85 neonates at 2, 10, 30, 90 days of life, and the relative proportions of the different bifidobacterial populations were assessed by 16S rRNA–23S rRNA internal transcribed spacer (ITS) region sequencing. Absolute levels of the main species were determined by q-PCR. Our results showed that the bifidobacterial population establishment is affected by gestational age, delivery mode, and infant feeding, as it is evidenced by qualitative and quantitative changes. These data underline the need for understanding the impact of perinatal factors on the gut microbiota also at low taxonomic levels, especially in the case of relevant microbial populations such as Bifidobacterium. The data obtained provide indications for the selection of the species best suited for the development of bifidobacteria-based products for different groups of neonates and will help to develop rational strategies for favoring a healthy early microbiota development when this process is challenged.     

In Vitro Characterization of Human CD24hiCD38hi Regulatory B Cells Shows CD9 Is Not a Stable Breg Cell Marker

Regulatory B (Breg) cells are endowed with immune suppressive functions. Various human and murine Breg subtypes have been reported. While interleukin (IL)-10 intracellular staining remains the most reliable way to identify Breg cells, this technique hinders further essential functional studies. Recent findings suggest that CD9 is an effective surface marker of murine IL-10 competent Breg cells. However, the stability of CD9 and its relevance as a unique marker for human Breg cells, which have been widely characterized as CD24^hiCD38^hi, have not been investigated. Here, we demonstrate that CD9 expression is sensitive to in vitro B cell stimulations. CD9 expression could either be re-expressed or downregulated in purified CD9-negative B cells and CD9-positive B cells, respectively. We found no significant differences in the Breg differentiation capacity of the CD9-negative and CD9-positive B cells. Furthermore, CD9-positive B cells co-express CD40 and CD86, suggesting their nature as B cell activation or co-stimulatory molecules, rather than regulatory ones. Therefore, we report the relatively unstable CD9 as a distinct surface molecule, indicating the need for further research for a more reliable marker to purify human Breg cells.     

Spike train decoding without spike sorting

We propose a novel paradigm for spike train decoding, which avoids entirely spike sorting based on waveform measurements. This paradigm directly uses the spike train collected at recording electrodes from thresholding the bandpassed voltage signal. Our approach is a paradigm, not an algorithm, since it can be used with any of the current decoding algorithms, such as population vector or likelihood-based algorithms. Based on analytical results and an extensive simulation study, we show that our paradigm is comparable to, and sometimes more efficient than, the traditional approach based on well-isolated neurons and that it remains efficient even when all electrodes are severely corrupted by noise, a situation that would render spike sorting particularly difficult. Our paradigm will also save time and computational effort, both of which are crucially important for successful operation of real-time brain-machine interfaces. Indeed, in place of the lengthy spike-sorting task of the traditional approach, it involves an exact expectation EM algorithm that is fast enough that it could also be left to run during decoding to capture potential slow changes in the states of the neurons.     

Effects of a startling acoustic stimulus on reaction time in different parkinsonian syndromes

BACKGROUND: The functional assessment of the startle circuit is usually done by analyzing the acoustic startle response (ASR). However, a startling acoustic stimulus (SAS) also induces changes in the excitability of neural structures that can be demonstrated by studying the SAS-induced change in the behavior of certain neurophysiologic responses. OBJECTIVE: To examine the effects induced by an SAS on voluntary reaction time in patients with parkinsonian syndromes (StartReact effect) and to compare the results with those obtained in a group of age-matched healthy volunteers. METHODS: Twelve patients with idiopathic PD (IPD), seven patients with progressive supranuclear palsy (PSP), seven patients with multisystem atrophy (MSA), and seven healthy age-matched control volunteers performed a simple visual reaction time task and received SAS together with the "go" signal in random trials. RESULTS: Baseline reaction time was significantly slower in PSP patients than in control subjects and MSA patients. The SAS induced a significant shortening of the reaction time in control subjects and in patients with IPD and MSA, but not in patients with PSP. The percentage of reaction time shortening with regard to the baseline values also differed significantly between PSP patients and the other groups of subjects. The StartReact effect was consistent throughout the experiment and showed reduced habituation with repeated testing. CONCLUSIONS: The results are consistent with an abnormal function of the startle circuit in patients with PSP and agree with previous studies using the ASR. The reduced habituation of the StartReact effect favors its clinical applicability in the assessment of differences between patients with parkinsonian syndromes.     

Aligned texture map creation for pose invariant face recognition

In last years, Face recognition based on 3D techniques is an emergent technology which has demonstrated better results than conventional 2D approaches. Using texture (180° multi-view image) and depth maps is supposed to increase the robustness towards the two main challenges in Face Recognition: Pose and illumination. Nevertheless, 3D data should be acquired under highly controlled conditions and in most cases depends on the collaboration of the subject to be recognized. Thus, in applications such as surveillance or control access points, this kind of 3D data may not be available during the recognition process. This leads to a new paradigm using some mixed 2D-3D face recognition systems where 3D data is used in the training but either 2D or 3D information can be used in the recognition depending on the scenario. Following this concept, where only part of the information (partial concept) is used in the recognition, a novel method is presented in this work. This has been called Partial Principal Component Analysis (P²CA) since they fuse the Partial concept with the fundamentals of the well known PCA algorithm. This strategy has been proven to be very robust in pose variation scenarios showing that the 3D training process retains all the spatial information of the face while the 2D picture effectively recovers the face information from the available data. Furthermore, in this work, a novel approach for the automatic creation of 180° aligned cylindrical projected face images using nine different views is presented. These face images are created by using a cylindrical approximation for the real object surface. The alignment is done by applying first a global 2D affine transformation of the image, and afterward a local transformation of the desired face features using a triangle mesh. This local alignment allows a closer look to the feature properties and not the differences. Finally, these aligned face images are used for training a pose invariant face recognition approach (P²CA).     

Multiple instance learning for classifying students in learning management systems

In this paper, a new approach based on multiple instance learning is proposed to predict student’s performance and to improve the obtained results using a classical single instance learning. Multiple instance learning provides a more suitable and optimized representation that is adapted to available information of each student and course eliminating the missing values that make difficult to find efficient solutions when traditional supervised learning is used. To check the efficiency of the new proposed representation, the most popular techniques of traditional supervised learning based on single instances are compared to those based on multiple instance learning. Computational experiments show that when the problem is regarded as a multiple instance one, performance is significantly better and the weaknesses of single-instance representation are overcome.