Parameter-insensitive kernel in extreme learning for non-linear support vector regression

Support vector regression (SVR) is a state-of-the-art method for regression which uses the ε‐sensitive loss and produces sparse models. However, non-linear SVRs are difficult to tune because of the additional kernel parameter. In this paper, a new parameter-insensitive kernel inspired from extreme learning is used for non-linear SVR. Hence, the practitioner has only two meta-parameters to optimise. The proposed approach reduces significantly the computational complexity yet experiments show that it yields performances that are very close from the state-of-the-art. Unlike previous works which rely on Monte-Carlo approximation to estimate the kernel, this work also shows that the proposed kernel has an analytic form which is computationally easier to evaluate.     

Software transactional memories: an approach for multicore programming

The recent advance of multicore architectures and the deployment of multiprocessors as the mainstream computing platforms have given rise to a new concurrent programming impetus. Software transactional memories (STM) are one of the most promising approaches to take up this challenge. The aim of a STM system is to discharge the application programmer from the management of synchronization when he/she has to write multiprocess programs. His/her task is to decompose his/her program into a set of sequential tasks that access shared objects, and to decompose each task in atomic units of computation. The management of the required synchronization is ensured by the associated STM system. This paper presents two existing STM systems, and a new one based on time-window mechanism. The paper, which focuses mainly on STM principles, has an introductory and survey flavor.     

Discovery of Heterocyclic Nonacetamide Synaptic Vesicle Protein 2A (SV2A) Ligands with Single‐Digit Nanomolar Potency: Opening Avenues towards the First SV2A Positron Emission Tomography (PET) Ligands

The role of the synaptic vesicle protein 2A (SV2A) protein, target of the antiepileptic drug levetiracetam, is still mostly unknown. Considering its potential to provide in vivo functional insights into the role of SV2A in epileptic patients, the development of an SV2A positron emission tomography (PET) tracer has been undertaken. Using a 3D pharmacophore model based on close analogues of levetiracetam, we report the rationale design of three heterocyclic non‐acetamide lead compounds, UCB‐A, UCB‐H and UCB‐J, the first single‐digit nanomolar SV2A ligands with suitable properties for development as PET tracers.     

Novel Heterobimetallic Radiotheranostic: Preparation,Activity, and Biodistribution

A novel Ru^II(arene) theranostic complex is presented. It is based on a 1,4,7,10‐tetraazacyclododecane‐1,4,7,10‐tetraacetic acid macrocycle bearing a triarylphosphine and can be tracked in vivo by using the γ emission of ¹⁵³Sm atoms. Notably, the heteroditopic ligand can be selectively metalated with ruthenium at the phosphorus atom despite the presence of other functionalities that are prone to metal coordination. Subsequent labeling with radionuclides such as ¹⁵³Sm can then be performed easily. The resulting heterobimetallic complex exhibits favorable solubility and stability properties in biologically relevant media. It also shows in vitro cytotoxicity in line with that expected for this type of metallodrug, and is nontoxic to the organism as a whole. As a proof of concept, initial studies in healthy mice were performed to obtain information about the uptake, biodistribution, and excretion of the radiolabeled complex.     

Dispersive mixing efficiency of an elongational flow mixer on PP/EPDM blends: Morphological analysis and correlation with viscoelastic properties

Polypropylene and ethylene‐propylene‐diene terpolymer (PP/EPDM) blends were melt compounded in a new mixing device, designed in our laboratory under the trademark of RMX®, which predominantly generates elongational flows. Dispersion of the EPDM minor phase in PP was carried out in both RMX® and in an internal mixer (Haake Rheomix 600) at equivalent specific mixing energies and the resultant morphologies obtained by SEM were analyzed and compared. A better dispersive mixing efficiency of the RMX® mixer, i.e., lower D_n and D_v of the dispersed EPDM phase was observed. The impact of elongational flow was more pronounced for blends having a high viscosity ratio p, indicating an enhanced droplet break‐up mechanism, which was attributed to the combination of high shear rates inside the mixing element and important elongational flows in the convergent/divergent zones. The morphology of the blends was correlated with their linear viscoelastic properties by using the Palierne model. Very good agreement was found for the PP/EPDM 80/20 blends but for higher EPDM content, the Palierne model failed to describe the rheological behavior, which was attributed to percolation of the minor phase with increasing the concentration. Higher elasticity at low frequencies was observed for blends processed in the RMX®, which was attributed to a higher generated interfacial area. POLYM. ENG. SCI., 54:1444–1457, 2014. © 2013 Society of Plastics Engineers     

Slip Length Measurements Using µPIV and TIRF-Based Velocimetry

The goal of this paper is to review progress (mostly recent) made in micro and nanovelocimetry, focusing on two techniques: µPIV (microparticle image velocimetry) and nanoPTV (nanoparticle tracking velocimetry). The paper focuses on the measurement of slippage (taken as a benchmark for these techniques), concentrating on work done in our group. We review the developments of µPIV that led, in the last ten years, to the achievement of 100 nm accuracy in the measurement of slip lengths. Later, this approach was complemented by nanoPTV, which recently obtained ±5 nm precision. Here, we also mention recent application of these techniques toward better characterization of microgel and polymer flows. As a whole, the two techniques have conveyed valuable information on flow behavior within and close to the boundaries of microchannels, on the importance of wetting, and on the role of surface heterogeneities. µPIV is commercially available but nanoPTV is not mature. Interesting instrumental developments are expected in the future for the latter technique.     

Quantitative and reliable in vitro method combining scanning electron microscopy and image analysis for the screening of osteotropic modulators

The increased generation and up-regulated activity of bone resorbing cells (osteoclasts) play a part in the impairment of bone remodeling in many bone diseases. Numerous drugs (bisphosphonates, calcitonin, selective estrogen receptor modulators) have been proposed to inhibit this increased osteoclastic activity. In this report, we describe a pit resorption assay quantified by scanning electron microscopy coupled with image analysis. Total rabbit bone cells with large numbers of osteoclasts were cultured on dentin slices. The whole surface of the dentin slice was scanned and both the number of resorption pits and the total resorbed surface area were measured. Resorption pits appeared at 48 h and increased gradually up to 96 h. Despite the observation of a strong correlation between the total resorption area and the number of pits, we suggest that area measurement is the most relevant marker for osteoclastic activity. Osteotropic factors stimulating or inhibiting osteoclastic activity were used to test the variations in resorption activity as measured with our method. This reproducible and sensitive quantitative method is a valuable tool for screening for osteoclastic inhibitors and, more generally, for investigating bone modulators.