R2D2: A scalable deep learning toolkit for medical imaging segmentation

Deep learning has gained a significant popularity in recent years thanks to its tremendous success across a wide range of relevant fields of applications, including medical image analysis domain in particular. Although convolutional neural networks (CNNs) based medical applications have been providing powerful solutions and revolutionizing medicine, efficiently training of CNNs models is a tedious and challenging task. It is a computationally intensive process taking long time and rare system resources, which represents a significant hindrance to scientific research progress. In order to address this challenge, we propose in this article, R2D2, a scalable intuitive deep learning toolkit for medical imaging semantic segmentation. To the best of our knowledge, the present work is the first that aims to tackle this issue by offering a novel distributed versions of two well-known and widely used CNN segmentation architectures [ie, fully convolutional network (FCN) and U-Net]. We introduce the design and the core building blocks of R2D2. We further present and analyze its experimental evaluation results on two different concrete medical imaging segmentation use cases. R2D2 achieves up to 17.5× and 10.4× speedup than single-node based training of U-Net and FCN, respectively, with a negligible, though still unexpected segmentation accuracy loss. R2D2 offers not only an empirical evidence and investigates in-depth the latest published works but also it facilitates and significantly reduces the effort required by researchers to quickly prototype and easily discover cutting-edge CNN configurations and architectures.     

Assessing the health research’s social impact: a systematic review

Scientometrics - In recent year, a growing attention is dedicated to the assessment of research’s social impact. While prior research has often dealt with results of research, the last decade...     

Chalcogenide Thermoelectrics Empowered by an Unconventional Bonding Mechanism

Thermoelectric materials have attracted significant research interest in recent decades due to their promising application potential in interconverting heat and electricity. Unfortunately, the strong coupling between the material parameters that determine thermoelectric efficiency, i.e., the Seebeck coefficient, electrical conductivity, and thermal conductivity, complicates the optimization of thermoelectric energy converters. Main‐group chalcogenides provide a rich playground to alleviate the interdependence of these parameters. Interestingly, only a subgroup of octahedrally coordinated chalcogenides possesses good thermoelectric properties. This subgroup is also characterized by other outstanding characteristics suggestive of an exceptional bonding mechanism, which has been coined metavalent bonding. This conclusion is further supported by a map that separates different bonding mechanisms. In this map, all octahedrally coordinated chalcogenides with good performance as thermoelectrics are located in a well‐defined region, implying that the map can be utilized to identify novel thermoelectrics. To unravel the correlation between chemical bonding mechanism and good thermoelectric properties, the consequences of this unusual bonding mechanism on the band structure are analyzed. It is shown that features such as band degeneracy and band anisotropy are typical for this bonding mechanism, as is the low lattice thermal conductivity. This fundamental understanding, in turn, guides the rational materials design for improved thermoelectric performance by tailoring the chemical bonding mechanism.     

Weissella cibaria short-fermented liquid sourdoughs based on quinoa or amaranth flours as fat replacer in focaccia bread formulation

Production of the traditional Italian Focaccia bread (FcBread) requires reformulation to meet the new rules for healthier foods. We tested the applicability of a short-fermented (15 h) liquid sourdough (LS) (dough yield, DY, 250), obtained using quinoa (Q) or amaranth (Am) flour fermented with a Weissella cibaria strain (C43-11) producing exopolysaccharides (EPS) as fat replacer in yeast-leavened FcBread. The LSs were applied during the FcBread-making process at industrial pilot plant, reducing the added fat amount by 20% (FcBread-QLS and FcBread-AmLS) and products were compared with reference samples not containing EPS. All the products were analysed for physico-chemical properties (pH, TTA, organic acids, protein content and profile, instrumental colour), textural properties, sensory quality and glycemic index. The application of both pseudo-cereal-based W. cibaria LSs in FcBread allowed obtaining products with a reduced amount of fat, increased protein amount, mainly due to the albumin and glutenin/glutelin fractions, reduced glycemic index, improved texture and preserved in the traditional sensory profile of ‘focaccia’.     

Design and manufacturing of anti-intrusion bars made of aluminium foam filled tubes

The role of an anti-intrusion bar for automotive use is to absorb the kinetic energy of the colliding vehicles that is partially converted into internal work of the members involved in the crash. The aim of this paper is to investigate the performances of anti-intrusion bars, made by tubes filled with aluminium foams. The reason for using a cellular material as a filler deals with its capacity to absorb energy during plastic deformation, while being lightweight. The study is mainly conducted by evaluating some key technical issues of the manufacturing problem and by running experimental and numerical analyses. The evaluation of materials and shapes of the closed sections to be filled is made in the perspective of a car manufacturer (production costs, weight reduction, space availability in a car door, etc.). Experimentally, foams are produced starting from an industrial aluminium precursor with a TiH₂ blowing agent. Empty and foam filled tubes are tested in three point bending, in order to evaluate their performances in terms of several performance parameters. Different manufacturing conditions, geometries and tube materials are investigated. The option of using hydroformed tubes, with non constant cross section, for the production of foam filled side structures id also discussed.     

Batch and Fed‐Batch Degradation of Enrofloxacin by the Fenton Process

Enrofloxacin is a synthetic second‐generation fluoroquinolone used as an antimicrobial agent exclusively in veterinary medicine. To simulate the treatment of wastewater contaminated by enrofloxacin, four‐day long fed‐batch runs were carried out according to the Fenton process with an enrofloxacin solution as model, to which hydrogen peroxide and ferrous ion were added twice a day. The residual enrofloxacin concentration was practically coincident to that detected at the end of the batch tests. Hydrogen peroxide was almost completely consumed after each feeding period, while the total organic carbon (TOC) concentration decreased gradually within three days, corresponding to a reduction > 58 %. From the third day on, the TOC falling rate was quite low. A yellow sludge settled due to the precipitation of both Fe(OH)₃ and a complex formed by ferric ion with adsorbed enrofloxacin and/or its oxidation products.