Deep Learning for Semantic Segmentation of Defects in Advanced STEM Images of Steels.

Crystalline materials exhibit long-range ordered lattice unit, within which resides nonperiodic structural features called defects. These crystallographic defects play a vital role in determining the physical and mechanical properties of a wide range of material systems. While computer vision has demonstrated success in recognizing feature patterns in images with well-defined contrast, automated identification of nanometer scale crystallographic defects in electron micrographs governed by complex contrast mechanisms is still a challenging task. Here, building upon an advanced defect imaging mode that offers high feature clarity, we introduce DefectSegNet - a new convolutional neural network (CNN) architecture that performs semantic segmentation of three common crystallographic defects in structural alloys: dislocation lines, precipitates and voids. Results from supervised training on a small set of high-quality defect images of steels show high pixel-wise accuracy across all three types of defects: 91.60 ± 1.77% on dislocations, 93.39 ± 1.00% on precipitates, and 98.85 ± 0.56% on voids. We discuss the sources of uncertainties in CNN prediction and the training data in terms of feature density, representation and homogeneity and their effects on deep learning performance. Further defect quantification using DefectSegNet prediction outperforms human expert average, presenting a promising new workflow for fast and statistically meaningful quantification of materials defects.

Crystal structure and chemistry of tricadmium digermanium tetra-arsenide, Cd3Ge2As4.

A cadmium germanium arsenide compound, Cd3Ge2As4, was synthesized using a double-containment fused quartz ampoule method within a rocking furnace and a melt-quench technique. The crystal structure was determined from single-crystal X-ray diffraction (SC-XRD), scanning and transmission electron microscopies (i.e. SEM, STEM, and TEM), and selected area diffraction (SAD) and confirmed with electron backscatter diffraction (EBSD). The chemistry was verified with electron energy loss spectroscopy (EELS).

Developing the latest framework to measure and incentivise pharmaceutical industry contributions to health research and development.

Major pharmaceutical companies contribute important expertise to health research and development (R&D), particularly in their ability to develop and bring pharmaceuticals to market. The Access to Medicine Index evaluates how 20 of the world's largest pharmaceutical companies are directing R&D efforts towards the needs of people living in low- and middle-income countries. In dissemination of its findings, the Index stimulates pharmaceutical companies to expand R&D activities in this direction. The Index methodology is reviewed every 2 years, most recently for the 2018 Index, to ensure their R&D activity is benchmarked against current health R&D priorities as defined by the global health community. The review is based on consensus-building processes involving global health stakeholders. In the latest review, two main changes to the methodology were made that will further deepen the Index's analysis of (1) how far companies' R&D activity aligns with global health priorities; and (2) whether companies make plans to ensure resulting innovations reach populations in need globally. These changes will be applied in the 2018 Access to Medicine Index. Importantly, the methodology review process highlighted the need for further prioritisation from the global health community, in particular to emphasise to innovators which product innovations are needed most critically to address the burden of non-communicable diseases in low- and middle-income countries. Should such prioritisations be developed, the Index can play an important role in tracking and stimulating company contributions towards them.

Towards bend-contour-free dislocation imaging via diffraction contrast STEM.

Dislocation imaging using transmission electron microscopy (TEM) has been an invaluable tool for characterizing crystallographic defects in metals. Compared to conventional TEM imaging, diffraction contrast imaging scanning transmission electron microscopy (DCI STEM) with appropriate setting can provide better defect contrast with almost negligible bend contour artifacts, enabling more effective analysis of dislocation structures. Here, we investigated why STEM can suppresses bend contour, and how dislocation contrast behaves along with different STEM imaging parameters. Using a body-centered cubic HT-9 ferritic/martensitic alloy as an example, a simple procedure and operational theory are described at the beginning to help set up DCI STEM experiments. Comparing with conventional TEM and the STEM strictly complying with the principle of reciprocity, we found that a pair of STEM convergence and collection semi-angles, αS and ßS, a few milliradians in size is essential for bend-contour-free defect imaging. It works in concert such that the convergence STEM probe opens up the reciprocal space, and then a comparable collection region evens out the rocking-curve oscillation and alleviates bend contours from the reciprocal space. This fundamental advantage is unique in DCI STEM. Practical guidelines regarding STEM parameters and specimen orientation and thickness are then provided for DCI STEM dislocation imaging. Lastly, we show that coupling DCI STEM with spectrum images of electron energy loss spectroscopy and of energy-dispersive X-ray Spectroscopy offers a comprehensive characterization of crystallographic defects and chemical information of complex microstructures.

Carbon Contamination During Ion Irradiation - Accurate Detection and Characterization of its Effect on Microstructure of Ferritic/Martensitic Steels.

Accelerator-based ion beam irradiation techniques have been used to study radiation effects in materials for decades. Although carbon contamination induced by ion beams in target materials is a well-known issue in some material systems, it has not been fully characterized nor quantified for studies in ferritic/martensitic (F/M) steels that are candidate materials for applications such as core structural components in advanced nuclear reactors. It is an especially important issue for this class of material because of the strong effect of carbon level on precipitate formation. In this paper, the ability to quantify carbon contamination using three common techniques, namely time-of-flight secondary ion mass spectroscopy (ToF-SIMS), atom probe tomography (APT), and transmission electron microscopy (TEM) is compared. Their effectiveness and shortcomings in determining carbon contamination are presented and discussed. The corresponding microstructural changes related to carbon contamination in ion irradiated F/M steels are also presented and briefly discussed.

Access to hepatitis C medicines.

Hepatitis C is a global epidemic. Worldwide, 185 million people are estimated to be infected, most of whom live in low- and middle-income countries. Recent advances in the development of antiviral drugs have produced therapies that are more effective, safer and better tolerated than existing treatments for the disease. These therapies present an opportunity to curb the epidemic, provided that they are affordable, that generic production of these medicines is scaled up and that awareness and screening programmes are strengthened. Pharmaceutical companies have a central role to play. We examined the marketed products, pipelines and access to medicine strategies of 20 of the world's largest pharmaceutical companies. Six of these companies are developing medicines for hepatitis C: AbbVie, Bristol-Myers Squibb, Gilead, Johnson & Johnson, Merck & Co. and Roche. These companies employ a range of approaches to supporting hepatitis C treatment, including pricing strategies, voluntary licensing, capacity building and drug donations. We give an overview of the engagement of these companies in addressing access to hepatitis C products. We suggest actions companies can take to play a greater role in curbing this epidemic: (i) prioritizing affordability assessments; (ii) developing access strategies early in the product lifecycle; and (iii) licensing to manufacturers of generic medicines.

L'hépatite C est une épidémie mondiale. On estime à 185 millions le nombre d'individus infectés par ce virus dans le monde, la plupart vivant dans des pays à revenu faible ou intermédiaire. Les récents progrès dans la mise au point de médicaments antiviraux ont conduit à des traitements plus efficaces, plus surs et mieux tolérés que les traitements existants pour soigner cette maladie. Ces traitements permettraient de freiner l'épidémie, à condition que leur coût soit abordable, que la production de médicaments génériques soit intensifiée et que les programmes de sensibilisation et de dépistage soient renforcés. Les sociétés pharmaceutiques ont, à cet égard, un rôle central à jouer. Nous avons examiné les produits commercialisés, les produits à l'étude et les stratégies d'accès aux médicaments de 20 des plus importantes sociétés pharmaceutiques mondiales. Six de ces sociétés développent des médicaments pour soigner l'hépatite C: AbbVie, Bristol-Myers Squibb, Gilead, Johnson & Johnson, Merck & Co. et Roche. Ces sociétés adoptent diverses approches pour faciliter le traitement de l'hépatite C, qui reposent notamment sur des stratégies de fixation des prix, l'octroi volontaire de licences, un renforcement des capacités et des dons de médicaments. Nous donnons un aperçu des efforts déployés par ces sociétés pour faciliter l'accès aux médicaments permettant de soigner l'hépatite C, et proposons des actions que peuvent mener ces sociétés afin de jouer un plus grand rôle dans l'enraiement de cette épidémie: (i) donner un degré de priorité élevé aux évaluations de l'accessibilité économique; (ii) développer des stratégies d'accès au début du cycle de vie du produit; et (iii) octroyer des licences aux fabricants de médicaments génériques.

La hepatitis C es una epidemia global. Se estima que, en todo el mundo, hay 185 millones de personas infectadas, la mayoría de las cuales viven en países de ingresos bajos y medios. Los recientes avances en el desarrollo de antivirales han producido terapias más efectivas, seguras y de mejor tolerancia que los tratamientos para la enfermedad existentes. Estas terapias presentan una oportunidad para poner freno a la epidemia, siempre y cuando sean asequibles, aumentar la producción genérica de dichos medicamentos y reforzar los programas de sensibilización y detección. Las empresas farmacéuticas juegan un papel central. Se examinaron los productos comercializados, tuberías y estrategias de acceso a los medicamentos de veinte de las mayores empresas farmacéuticas del mundo. Seis de estas empresas están desarrollando medicamentos para tratar la hepatitis C: AbbVie, Bristol-Myers Squibb, Gilead, Johnson & Johnson, Merck & Co. y Roche. Estas empresas emplean una gama de enfoques para apoyar el tratamiento para la hepatitis C, incluyendo las estrategias de fijación de precios, la concesión voluntaria de licencias, la creación de capacidad y las donaciones de medicamentos. Ofrecemos una visión general del compromiso de estas empresas a la hora de ofrecer acceso a los productos para tratar la hepatitis C. Sugerimos acciones que las empresas pueden llevar a cabo para tener un papel más importante a la hora de frenar esta epidemia: (i) dar prioridad a los criterios de asequibilidad; (ii) desarrollar estrategias de acceso al principio del ciclo de vida del producto; y (iii) ofrecer licencias a los fabricantes de medicamentos genéricos.

Multislice simulation of transmission electron microscopy imaging of helium bubbles in Fe.

Formation of nanoscale helium (He) bubbles in reduced activation ferritic/martensitic steels may lead to degradation of mechanical properties of materials. Transmission electron microscopy (TEM) has commonly been used to image the Fresnel contrast of He bubbles, using an underfocus of 0.5-1 µm. This paper presents our study of multislice simulation of the size correlation between imaged Fresnel rings and the actual He bubbles. It was found that for bubbles equal to or >3 nm in diameter, the imaged bubble size, represented by its inner diameter of the first dark Fresnel ring (D(in)) in underfocused imaging conditions, increases with increasing electron-beam incoherency, but decreases with increasing underfocus. The electron-beam accelerating voltage, bubble size, bubble position and TEM sample thickness were found to have no significant influence on the deviation of D(in) from the actual bubble size (D(0)). However, for bubbles equal to or <2 nm, D(in)/D(0) increases dramatically with increasing underfocus when it is above a threshold limit (e.g. Δf = -1 µm for a 2-nm bubble). The results of this study also suggested that He bubbles can be differentiated from argon (Ar) bubbles by contrast differences.