Image segmentation towards new image representation methods

Very low bit-rate video coding has recently become one of the most important areas of image communication and a large variety of applications have already been identified. Since conventional approaches are reaching a saturation point, in terms of coding efficiency, a new generation of video coding techniques, aiming at a deeper “understanding” of the image, is being studied. In this context, image analysis, particularly the identification of objects or regions in images (segmentation), is a very important step. This paper describes a segmentation algorithm based on split and merge. Images are first simplified using mathematical morphology operators, which eliminate perceptually less relevant details. The simplified image is then split according to a quad tree structure and the resulting regions are finally merged in three steps: merge, elimination of small regions and control of the number of regions.     

Experience classification for transfer learning in traffic signal control

In recent years, due to the drastic rise in the number of vehicles and the lack of sufficient infrastructure, traffic jams, air pollution, and fuel consumption have increased in cities. The optimization of timing for traffic lights is one of the solutions for the mentioned problems. Many methods have been introduced to deal with these problems, including reinforcement learning. Although a great number of learning-based methods have been used in traffic signal control, they suffer from poor performance and slow learning convergence. In this paper, a transfer learning-based method for traffic signal control has been proposed. Multi-agent system has also been used for modelling the traffic network and transfer learning has been used to make reinforcement learning agents transfer their experience to each other. Furthermore, a classifier has been utilized to classify the transferred experiences. The results show that using the proposed method leads to a significant improvement on average delay time and convergence time of the learning process.

     

Optimization of the grapes reception process

Neural Computing and Applications - Grapes reception is a key process in wine production. The harvest days are extremely challenging days in managing the reception of the grapes, as the winery...     

Modeling of the condyle elements within a biomechanical knee model

The development of a computational multibody knee model able to capture some of the fundamental properties of the human knee articulation is presented. This desideratum is reached by including the kinetics of the real knee articulation. The research question is whether an accurate modeling of the condyle contact in the knee will lead to reproduction of the complex combination of flexion/extension, abduction/adduction, and tibial rotation observed in the real knee. The model is composed by two anatomic segments, the tibia and the femur, whose characteristics are functions of the geometric and anatomic properties of the real bones. The biomechanical model characterization is developed under the framework of multibody systems methodologies using Cartesian coordinates. The type of approach used in the proposed knee model is the joint surface contact conditions between ellipsoids, representing the two femoral condyles, and points, representing the tibial plateau and the menisci. These elements are closely fitted to the actual knee geometry. This task is undertaken by considering a parameter optimization process to replicate experimental data published in the literature, namely that by Lafortune and his coworkers in 1992. Then kinematic data in the form of flexion/extension patterns are imposed on the model corresponding to the stance phase of the human gait. From the results obtained, by performing several computational simulations, it can be observed that the knee model approximates the average secondary motion patterns observed in the literature. Because the literature reports considerable inter-individual differences in the secondary motion patterns, the knee model presented here is also used to check whether it is possible to reproduce the observed differences with reasonable variations of bone shape parameters. This task is accomplished by a parameter study, in which the main variables that define the geometry of condyles are taken into account. It was observed that the data reveal a difference in secondary kinematics of the knee in flexion versus extension. The likely explanation for this fact is the elastic component of the secondary motions created by the combination of joint forces and soft tissue deformations. The proposed knee model is, therefore, used to investigate whether this observed behavior can be explained by reasonable elastic deformations of the points representing the menisci in the model.     

Vapor pressure measurements of para-hydrogen adsorbed on exfoliated graphite

Monolayer isotherms were measured for 15, 18, and 20 K for para-hydrogen adsorbed on exfoliated graphite foam, and for the second, third, and fourth layers in the temperature interval of 8 to 20 K, using the standard volumetric method. From the data, the isosteric heat of adsorption, molar entropies and internal energies were obtained as a function of coverage. The values for coverages above the third layer were compared to those at the p-H₂ bulk triple point, showing that, within our experimental uncertainty, the results are closer to the values of bulk solid para-hydrogen. Critical temperatures for the second and third layers and the triple point temperature for the second layer, were determined, yielding values of T_c2 = (10.0±0.1) K, T_c3 = (11.0±0.5) K and T_t2 = (6.5±0.1) K respectively. Features occurring along the monolayer coverages are compared to transitions which occur in the para-hydrogen phase diagrams adsorbed on graphite, obtained from heat capacity measurements by other authors. From the isotherms, compressibilities were calculated as a function of coverage for several temperatures. Whenever possible the obtained results were compared to existing data in the literature.     

The Chick Chorioallantoic Membrane Model: A New In Vivo Tool to Evaluate Breast Cancer Stem Cell Activity

The high plasticity of cancer stem-like cells (CSCs) allows them to differentiate and proliferate, specifically when xenotransplanted subcutaneously into immunocompromised mice. CSCs are highly tumorigenic, even when inoculated in small numbers. Thus, in vivo limiting dilution assays (LDA) in mice are the current gold standard method to evaluate CSC enrichment and activity. The chick embryo chorioallantoic membrane (CAM) is a low cost, naturally immune-incompetent and reproducible model widely used to evaluate the spontaneous growth of human tumor cells. Here, we established a CAM-LDA assay able to rapidly reproduce tumor specificities—in particular, the ability of the small population of CSCs to form tumors. We used a panel of organotropic metastatic breast cancer cells, which show an enrichment in a stem cell gene signature, enhanced CD44⁺/CD24^−/low cell surface expression and increased mammosphere-forming efficiency (MFE). The size of CAM-xenografted tumors correlate with the number of inoculated cancer cells, following mice xenograft growth pattern. CAM and mice tumors are histologically comparable, displaying both breast CSC markers CD44 and CD49f. Therefore, we propose a new tool for studying CSC prevalence and function—the chick CAM-LDA—a model with easy handling, accessibility, rapid growth and the absence of ethical and regulatory constraints.     

Dehydropeptide Supramolecular Hydrogels and Nanostructures as Potential Peptidomimetic Biomedical Materials

Supramolecular peptide hydrogels are gaining increased attention, owing to their potential in a variety of biomedical applications. Their physical properties are similar to those of the extracellular matrix (ECM), which is key to their applications in the cell culture of specialized cells, tissue engineering, skin regeneration, and wound healing. The structure of these hydrogels usually consists of a di- or tripeptide capped on the N-terminus with a hydrophobic aromatic group, such as Fmoc or naphthalene. Although these peptide conjugates can offer advantages over other types of gelators such as cross-linked polymers, they usually possess the limitation of being particularly sensitive to proteolysis by endogenous proteases. One of the strategies reported that can overcome this barrier is to use a peptidomimetic strategy, in which natural amino acids are switched for non-proteinogenic analogues, such as D-amino acids, β-amino acids, or dehydroamino acids. Such peptides usually possess much greater resistance to enzymatic hydrolysis. Peptides containing dehydroamino acids, i.e., dehydropeptides, are particularly interesting, as the presence of the double bond also introduces a conformational restraint to the peptide backbone, resulting in (often predictable) changes to the secondary structure of the peptide. This review focuses on peptide hydrogels and related nanostructures, where α,β-didehydro-α-amino acids have been successfully incorporated into the structure of peptide hydrogelators, and the resulting properties are discussed in terms of their potential biomedical applications. Where appropriate, their properties are compared with those of the corresponding peptide hydrogelator composed of canonical amino acids. In a wider context, we consider the presence of dehydroamino acids in natural compounds and medicinally important compounds as well as their limitations, and we consider some of the synthetic strategies for obtaining dehydropeptides. Finally, we consider the future direction for this research area.     

P16 and HPV Genotype Significance in HPV-Associated Cervical Cancer—A Large Cohort of Two Tertiary Referral Centers

The expression of p16 is a good surrogate of human papillomavirus (HPV) infection in HPV-associated cancers. The significance of p16 expression, HPV genotype and genera in the outcome of patients with HPV-associated cervical cancer (CC) is unclear. Our aim is to ascertain the prognostic significance of these factors. Data from 348 patients (median age: 47.5 years old) with CC, diagnosed in two referral centers, were retrospectively collected. Advanced disease (FIGO2018 IB2-IV) was present in 68% of patients. A single HPV genotype was identified in 82.8% of patients. The most common HPVs were HPV16 (69%) and HPV18 (14%). HPV genera reflected this distribution. HPV16 tumors presented at an earlier stage. P16 was negative in 18 cases (5.2%), 83.3% of which were squamous cell carcinomas. These cases occurred in older patients who tended to have advanced disease. In the univariate analysis, HPV16 (HR: 0.58; p = 0.0198), α-9 genera (HR: 0.37; p = 0.0106) and p16 overexpression (HR: 0.54; p = 0.032) were associated with better survival. HPV16 (HR: 0.63; p = 0.0174) and α-9 genera (HR: 0.57; p = 0.0286) were associated with less relapse. In the multivariate analysis, only the International Federation of Gynecology and Obstetrics (FIGO) stage retained an independent prognostic value. HPV16, α-9 genera and p16 overexpression were associated with better survival, although not as independent prognostic factors. Patients with p16-negative HPV-associated CC were older, presented with advanced disease and had worse prognosis.     

Improved Controlled Release and Brain Penetration of the Small Molecule S14 Using PLGA Nanoparticles

Phosphodiesterase 7 (PDE7) is an enzyme responsible for the degradation of cyclic adenosine monophosphate (cAMP), an important cellular messenger. PDE7’s role in neurotransmission, expression profile in the brain and the druggability of other phosphodiesterases have motivated the search for potent inhibitors to treat neurodegenerative and inflammatory diseases. Different heterocyclic compounds have been described over the years; among them, phenyl-2-thioxo-(1H)-quinazolin-4-one, called S14, has shown very promising results in different in vitro and in vivo studies. Recently, polymeric nanoparticles have been used as new formulations to target specific organs and produce controlled release of certain drugs. In this work, we describe poly(lactic-co-glycolic acid) (PLGA)-based polymeric nanoparticles loaded with S14. Their preparation, optimization, characterization and in vivo drug release profile are here presented as an effort to improve pharmacokinetic properties of this interesting PDE7 inhibitor.