Stacking Fault-Enriched MoNi4/MoO2 Enables High-Performance Hydrogen Evolution.

Producing green hydrogen in a cost-competitive manner via water electrolysis will make the long-held dream of hydrogen economy a reality. Although platinum (Pt)-based catalysts show good performance toward hydrogen evolution reaction (HER), the high cost and scarce abundance challenge their economic viability and sustainability. Here, a non-Pt, high-performance electrocatalyst for HER achieved by engineering high fractions of stacking fault (SF) defects for MoNi4/MoO2 nanosheets (d-MoNi) through a combined chemical and thermal reduction strategy is shown. The d-MoNi catalyst offers ultralow overpotentials of 78 and 121 mV for HER at current densities of 500 and 1000 mA cm-2 in 1 M KOH, respectively. The defect-rich d-MoNi exhibits four times higher turnover frequency than the benchmark 20% Pt/C, together with its excellent durability (> 100 h), making it one of the best-performing non-Pt catalysts for HER. The experimental and theoretical results reveal that the abundant SFs in d-MoNi induce a compressive strain, decreasing the proton adsorption energy and promoting the associated combination of *H into hydrogen and molecular hydrogen desorption, enhancing the HER performance. This work provides a new synthetic route to engineer defective metal and metal alloy electrocatalysts for emerging electrochemical energy conversion and storage applications.

Enhancing the Life Cycle of Zinc-Iodine Batteries in Ionic Liquid-Based Electrolyte.

Aqueous zinc-iodine (Zn-I2) batteries are gaining significant attention due to their low-cost, high safety and high theoretical capacity. Nevertheless, their long cycle and durability have been hampered due to the use of aqueous media that overtime lead to Zn dendrite formation, hydrogen evolution reaction, and polyiodide dissolution. Xiao et. al., recently reported the addition of an imidazolium-based ionic liquid (IL), to an aqueous electrolyte, that plays a key role in modifying the solvation of Zn2+ ions in the bulk electrolyte and the inner Helmotlz plane, which eliminates free H2O molecules to be present on the Zn anode surface. UV/Vis and NMR spectroscopy also indicates a strong interaction between imidazolium cation [EMIM]+ and I3-, thereby reducing polyiodide shuttling and enhancing the cycle life of the battery. Overall, a capacity decay rate of only 0.01 % per cycle after over 18,000 cycles at 4 Ag-1, is observed making the use of IL additives in aqueous electrolytes highly promising candidates for Zn-I2 batteries.

Efficacy and safety of artificial tears containing Artemia salina extract with dinucleotides for dry eye.

CLINICAL RELEVANCE: This clinical trial was conducted as part of the marketing procedures for a medical device comprising artificial tears containing Artemia salina extract with dinucleotides. These molecules previously demonstrated secretagogue properties by enhancing the production of aqueous, mucinous, and lipidic components of the tears. BACKGROUND: After confirming the efficacy of artificial tears containing Artemia salina extract in an animal model, this study proceeded to evaluate their efficacy and safety on dry eye participants. METHODS: A randomised controlled clinical trial was performed on 36 dry eye participants (41.6 ± 20.6 years). Half of the participants were treated with saline solution as a placebo for four weeks, while the other half were treated with artificial tears containing Artemia salina, randomly assigned. After a wash-out period of two weeks, the treatments were crossed for another four weeks. Participants were assessed at baseline and after one week, two weeks, and four weeks. Efficacy variables were: eye dryness frequency (primary), eye comfort, visual satisfaction, tear secretion, tear break-up time, corneal staining, conjunctival staining, and conjunctival hyperaemia. Safety variables were: high- and low-contrast visual acuity, intraocular pressure, and eye fundus images analysis. RESULTS: Compared with the baseline, the saline solution showed no significant changes in any of the studied variables after four weeks of treatment (p ≥ 0.05). However, the topical instillation of the artificial tears with Artemia salina for four weeks significantly improved eye dryness frequency (p = 0.014) and corneal staining (p = 0.010). No systemic or ocular adverse events were reported during the clinical trial. CONCLUSION: The topical instillation of artificial tears containing Artemia salina in mild to moderate dry eye participants for four weeks slightly improved their symptoms related to eye dryness frequency and reduced corneal damage, with no undesirable side effects observed.

Innovative Strategy for Developing PEDOT Composite Scaffold for Reversible Oxygen Reduction Reaction.

Metal-air batteries are an emerging technology with great potential to satisfy the demand for energy in high-consumption applications. However, this technology is still in an early stage, facing significant challenges such as a low cycle life that currently limits its practical use. Poly(3,4-ethylenedioxythiophene) (PEDOT) conducting polymer has already demonstrated its efficiency as catalyst for oxygen reduction reaction (ORR) discharge as an alternative to traditional expensive and nonsustainable metal catalysts. Apart from that, in most electrochemical processes, three phenomena are needed: redox activity and electronic and ionic conduction. Material morphology is important to maximize the contact area and optimize the 3 mechanisms to obtain high-performance devices. In this work, porous scaffolds of PEDOT-organic ionic plastic crystal (OIPC) are prepared through vapor phase polymerization to be used as porous self-standing cathodes. The scaffolds, based on abundant elements, showed good thermal stability (200 °C), with potential ORR reversible electrocatalytic activity: 60% of Coulombic efficiency in aqueous medium after 200 cycles.

Biobased Acrylic Latexes/Sodium Carboxymethyl Cellulose Aqueous Binders for Lithium-Ion NMC 811 Cathodes.

The increasing demands for sustainable energy storage technologies have prompted extensive research in the development of eco-friendly materials for lithium-ion batteries (LIBs). This research article presents the design of biobased latexes, which are fluorine-free and rely on renewable resources, based on isobornyl methacrylate (IBOMA) and 2-octyl acrylate (2OA) to be used as binders in batteries. Three different compositions of latexes were investigated, varying the ratio of IBOMA and 2OA: (1) Poly2OA homopolymer, (2) Poly(2OA0,6-co-IBOMA0,4) random copolymer, and (3) PolyIBOMA homopolymer. The combination of the two monomers provided a balance between rigidity from the hard monomer (IBOMA) and flexibility from the soft one (2OA). The study evaluated the performance of the biobased latexes using sodium carboxymethyl cellulose (CMC) as a thickener and cobinder by fabricating LiNi0.8Mn0.1Co0.1O2 (NMC 811) cathodes. Also, to compare with the state of the art, organic processed PVDF electrodes were prepared. Among aqueous slurries, rheological analysis showed that the CMC + Poly(2OA0,6-co-IBOMA0,4) binder system resulted in the most stable and well-dispersed slurries. Also, the electrodes prepared with this latex demonstrated enhanced adhesion (210 ± 9 N m-1) and reduced cracks compared to other aqueous compositions. Electrochemical characterization revealed that the aqueous processed cathodes using the CMC + Poly(2OA0,6-co-IBOMA0,4) biobased latex displayed higher specific capacities than the control with no latex at high C-rates (100.3 ± 2.1 vs 64.5 ± 0.8 mAh g-1 at 5C) and increased capacity retention after 90 cycles at 0.5C (84% vs 81% for CMC with no latex). Overall, the findings of this study suggest that biobased latexes, specifically the CMC + Poly(2OA0,6-co-IBOMA0,4) composition, are promising as environmentally friendly binders for NMC 811 cathodes, contributing to the broader goal of achieving sustainable energy storage systems.

Bio-based ether solvent and ionic liquid electrolyte for sustainable sodium-air batteries.

Sodium-air batteries (SABs) are receiving considerable attention for the development of next generation battery alternatives due to their high theoretical energy density (up to 1105 W h kg-1). However, most of the studies on this technology are still based on organic solvents; in particular, diglyme, which is highly flammable and toxic for the unborn child. To overcome these safety issues, this research investigates the first use of a branched ether solvent 1,2,3-trimethoxypropane (TMP) as an alternative electrolyte to diglyme for SABs. Through this work, the reactivity of the central tertiary carbon in TMP towards bare sodium metal was identified, while the addition of N-butyl-N-methylpyrrolidinium bis(trifluoromethylsulfonyl)imide ([C4mpyr][TFSI]) as a co-solvent proved to be an effective strategy to limit the reactivity. Moreover, a Na-ß-alumina disk was employed for anode protection, to separate the TMP-based electrolyte from the sodium metal. The new cell design resulted in improved cell performance: discharge capacities of up to 1.92 and 2.31 mA h cm-2 were achieved for the 16.6 mol% NaTFSI in TMP and 16.6 mol% NaTFSI in TMP/[C4mpyr][TFSI] compositions, respectively. By means of SEM, Raman and 23Na NMR techniques, NaO2 cubes were identified to be the major discharge product for both electrolyte compositions. Moreover, the hybrid electrolyte was shown to hinder the formation of side-products during discharge - the ratio of NaO2 to side-products in the hybrid electrolyte was 2.4 compared with 0.8 for the TMP-based electrolyte - and a different charge mechanism for the dissolution of NaO2 cubes for each electrolyte was observed. The findings of this work demonstrate the high potential of TMP as a base solvent for SABs, and the importance of careful electrolyte composition design in order to step towards greener and less toxic batteries.

Carrageenans as Sustainable Water-Processable Binders for High-Voltage NMC811 Cathodes.

Poly(vinylidene fluoride) (PVDF) is the most common binder for cathode electrodes in lithium-ion batteries. However, PVDF is a fluorinated compound and requires toxic N-methyl-2-pyrrolidone (NMP) as a solvent during the slurry preparation, making the electrode fabrication process environmentally unfriendly. In this study, we propose the use of carrageenan biopolymers as a sustainable source of water-processable binders for high-voltage NMC811 cathodes. Three types of carrageenan (Carr) biopolymers were investigated, with one, two, or three sulfonate groups (SO3-), namely, kappa, iota, and lambda carrageenans, respectively. In addition to the nature of carrageenans, this article also reports the optimization of the cathode formulations, which were prepared by using between 5 wt % of the binder to a lower amount of 2 wt %. Processing of the aqueous slurries and the nature of the binder, in terms of the morphology and electrochemical performance of the electrodes, were also investigated. The Carr binder with 3SO3- groups (3SO3-Carr) exhibited the highest discharge capacities, delivering 133.1 mAh g-1 at 3C and 105.0 mAh g-1 at 5C, which was similar to the organic-based PVDF electrode (136.1 and 108.7 mAh g-1, respectively). Furthermore, 3SO3-Carr reached an outstanding capacity retention of 91% after 90 cycles at 0.5C, which was attributed to a homogeneous NMC811 and a conductive carbon particle dispersion, superior adhesion strength to the current collector (17.3 ± 0.7 N m-1 vs 0.3 ± 0.1 N m-1 for PVDF), and reduced charge-transfer resistance. Postmortem analysis unveiled good preservation of the NMC811 particles, while the 1SO3-Carr and 2SO3-Carr electrodes showed damaged morphologies.

Enhanced Dissolution of Metal Oxides in Hydroxylated Solvents - Towards Application in Lithium-Ion Battery Leaching.

The recovery of critical metals from spent lithium-ion batteries (LIBs) is rapidly growing. Current methods are energy-intensive and hazardous, while alternative solvent-based strategies require more studies on their 'green' character, metal dissolution mechanism and industrial applicability. Herein, we bridged this gap by studying the effect of dilute HCl solutions in hydroxylated solvents to dissolve Co, Ni and Mn oxides. Ethylene glycol emerged consistently as the most effective solvent, dissolving up to four times more Co and Ni oxides than using aqueous acidic media, attributed to improved chloro-complex formation and solvent effects. These effects had a significant contribution compared to acid type and concentration. The highest Co dissolution (0.27 M) was achieved in 0.5 M HCl in 25 % (v/v) glycerol in water, using less acid and a significant amount of water compared to other solvent systems, as well as mild temperatures (40 °C). This solvent was applied to dissolve battery cathode material, achieving 100 % dissolution of Co and Mn and 94 % dissolution of Ni, following what was concluded to be a mixed mechanism. These results offer a simple alternative to current leaching processes, reducing acid consumption, enhancing atomic efficiency, and paving the way for optimized industrial hydrometallurgical processes leaning to 'greener' strategies.

Removal of polyvinylidene fluoride binder and other organics for enhancing the leaching efficiency of lithium and cobalt from black mass.

The agglomeration and encapsulation of recoverable materials of interest (e.g. metals and graphite) as a result of the presence of polyvinylidene fluoride (PVDF) in spent lithium-ion batteries (LIBs) with mixed chemistries (black mass) lower the extraction efficiency of metals. In this study, organic solvents and alkaline solutions were used as non-toxic reagents to investigate the removal of a PVDF binder from a black mass. The results demonstrated that 33.1%, 31.4%, and 31.4% of the PVDF were removed using dimethylformamide (DMF), dimethylacetamide (DMAc), and dimethyl sulfoxide (DMSO) at 150, 160, and 180 °C, respectively. Under these conditions, the peel-off efficiencies for DMF, DMAc, and DMSO were 92.9%, 85.3%, and approximately 92.9%, respectively. Using tetrabutylammonium bromide (TBAB) as a catalyst and 5 M sodium hydroxide (NaOH) at room temperature (RT- 21 °C-23 °C), 50.3% of PVDF and other organic compounds were eliminated. The removal efficiency was enhanced to approximately 60.5% when the temperature was raised to 80 °C using NaOH. Using 5 M potassium hydroxide at RT in a TBAB-containing solution, ca. 32.8% removal efficiency was obtained; raising the temperature to 80 °C further enhanced the removal efficiency to almost 52.7%. The peel-off efficiency was 100% for both alkaline solutions. Lithium extraction increased from 47.2% to 78.7% following treatment with DMSO and to 90.1% following treatment with NaOH via leaching black mass (2 M sulfuric acid, solid-to-liquid ratio (S/L): 100 g L-1 at 50 °C, for 1 h without a reducing agent) before and after removal of the PVDF binder. Cobalt's recovery went from 28.5% to 61.3% with DMSO treatment to 74.4% with NaOH treatment.

Protocolo de extracción y administración precoz de calostro como inmunoterapia de oro para recién nacidos de peso ≤ 1500 g y/o edad gestacional ≤ 32 semanas / Protocol for the extraction and early administration of colostrum as gold immunotherapy for newborns weighing ≤ 1500g and/or gestational age ≤ 32 weeks

Introducción: El calostro se obtiene durante los primeros días tras el parto y hasta dos semanas posparto en caso de nacimiento prematuro. Este último es rico en inmunoglobulina A y otros inmunocomponentes. Al impregnar la mucosa orofaríngea del prematuro con calostro, las citoquinas activan el tejido linfático orofaríngeo, estimulando así el sistema inmune. Objetivo: Protocolizar la extracción de calostro materno en las dos horas tras el parto o cesárea y su administración orofaríngea tras el nacimiento en las primeras seis horas de vida. Metodología: Se realizó una revisión de la literatura científica en las principales bases de datos Cochrane plus, Pubmed, Scielo, CINAHL, ENFISPO y Cuiden. La mayoría de artículos consultados son de los últimos diez años. Los descriptores utilizados fueron calostro/calostrum, orofaríngeo/oropharyngeal, neonato pretérmino/infant,premature y prevención/prevention. Formación sobre habilidades y técnicas sobre extracción precoz de calostro materno. Resultados: Se elaboró un protocolo multidisciplinar de actuación para ofrecer a los recién nacidos pretérmino de forma precoz el calostro de su propia madre. Conclusión: Este procedimiento es una intervención sencilla, segura y de bajo coste que beneficia a la mujer y al recién nacido pretérmino. Se ha relacionado con la disminución de la morbimortalidad en recién nacidos pretérmino <1500 g y/o <32 s. (AU)

Introduction: Colostrum is released during the first few days after the birth of the newborn and up to two weeks postpartum in case of preterm delivery. Colostrum is a source of inmunoglobulin A and other immune components. When it permeates the oropharyngeal mucosa of the premature baby, cytokines activate the oropharyngeal lymphatic tissue and this stimulates the immune system. Objective: Protocolize the extraction of matern colostrum during the first two hours after the partum or caesarean section and the oropharyngeal administration during the first six hours of life. Methodology: A review of the scientific literature was carried out in the main Cochrane plus, Pubmed, Scielo, CINAHL, ENFISPO and Cuiden databases. Most of the articles consulted are from the last ten years. The descriptors used were colostrum/colostrum, oropharyngeal/oropharyngeal, preterm neonate/infant, premature and prevention/prevention.Training on skills and techniques on early extraction of maternal colostrum. Results: A multidisciplinary action protocol was developed to offer preterm newborns early their own mother's colostrum. Conclusion: This procedure is a simple, secure and low cost intervention that benefits women and preterm newborns. Besides, it has been associated with a decrease in morbidity and mortality in preterm newborns with less than 1500 grams and/or 32 weeks of gestation. (AU)

A Deep Learning Framework for the Detection and Quantification of Reticular Pseudodrusen and Drusen on Optical Coherence Tomography.

Purpose: The purpose of this study was to develop and validate a deep learning (DL) framework for the detection and quantification of reticular pseudodrusen (RPD) and drusen on optical coherence tomography (OCT) scans. Methods: A DL framework was developed consisting of a classification model and an out-of-distribution (OOD) detection model for the identification of ungradable scans; a classification model to identify scans with drusen or RPD; and an image segmentation model to independently segment lesions as RPD or drusen. Data were obtained from 1284 participants in the UK Biobank (UKBB) with a self-reported diagnosis of age-related macular degeneration (AMD) and 250 UKBB controls. Drusen and RPD were manually delineated by five retina specialists. The main outcome measures were sensitivity, specificity, area under the receiver operating characteristic (ROC) curve (AUC), kappa, accuracy, intraclass correlation coefficient (ICC), and free-response receiver operating characteristic (FROC) curves. Results: The classification models performed strongly at their respective tasks (0.95, 0.93, and 0.99 AUC, respectively, for the ungradable scans classifier, the OOD model, and the drusen and RPD classification models). The mean ICC for the drusen and RPD area versus graders was 0.74 and 0.61, respectively, compared with 0.69 and 0.68 for intergrader agreement. FROC curves showed that the model's sensitivity was close to human performance. Conclusions: The models achieved high classification and segmentation performance, similar to human performance. Translational Relevance: Application of this robust framework will further our understanding of RPD as a separate entity from drusen in both research and clinical settings.

Unveiling the Impact of the Cations and Anions in Ionic Liquid/Glyme Hybrid Electrolytes for Na-O2 Batteries.

A series of hybrid electrolytes composed of diglyme and ionic liquids (ILs) have been investigated for Na-O2 batteries, as a strategy to control the growth and purity of the discharge products during battery operation. The dependence of chemical composition of the ILs on the size, purity, and distribution of the discharge products has been evaluated using a wide range of experimental and spectroscopic techniques. The morphology and composition of the discharge products found in the Na-O2 cells have a complex dependence on the physicochemical properties of the electrolyte as well as the speciation of the Na+ and superoxide radical anion. All of these factors control the nucleation and growth phenomena as well as electrolyte stability. Smaller discharge particle sizes and largely homogeneous (2.7 ± 0.5 µm) sodium superoxide (NaO2) crystals with only 9% of side products were found in the hybrid electrolyte containing the pyrrolidinium IL with a linear alkyl chain. The long-term cyclability of Na-O2 batteries with high Coulombic efficiency (>90%) was obtained for this electrolyte with fewer side products (20 cycles at 0.5 mA h cm-2). In contrast, rapid failure was observed with the use of the phosphonium-based electrolyte, which strongly stabilizes the superoxide anion. A high discharge capacity (4.46 mA h cm-2) was obtained for the hybrid electrolyte containing the pyrrolidinium-based IL bearing a linear alkyl chain with a slightly lower value (3.11 mA h cm-2) being obtained when the hybrid electrolyte contained similar pyrrolidinium-based IL bearing an alkoxy chain.

Trustworthy AI: Closing the gap between development and integration of AI systems in ophthalmic practice.

An increasing number of artificial intelligence (AI) systems are being proposed in ophthalmology, motivated by the variety and amount of clinical and imaging data, as well as their potential benefits at the different stages of patient care. Despite achieving close or even superior performance to that of experts, there is a critical gap between development and integration of AI systems in ophthalmic practice. This work focuses on the importance of trustworthy AI to close that gap. We identify the main aspects or challenges that need to be considered along the AI design pipeline so as to generate systems that meet the requirements to be deemed trustworthy, including those concerning accuracy, resiliency, reliability, safety, and accountability. We elaborate on mechanisms and considerations to address those aspects or challenges, and define the roles and responsibilities of the different stakeholders involved in AI for ophthalmic care, i.e., AI developers, reading centers, healthcare providers, healthcare institutions, ophthalmological societies and working groups or committees, patients, regulatory bodies, and payers. Generating trustworthy AI is not a responsibility of a sole stakeholder. There is an impending necessity for a collaborative approach where the different stakeholders are represented along the AI design pipeline, from the definition of the intended use to post-market surveillance after regulatory approval. This work contributes to establish such multi-stakeholder interaction and the main action points to be taken so that the potential benefits of AI reach real-world ophthalmic settings.

Adversarial attack vulnerability of medical image analysis systems: Unexplored factors.

Adversarial attacks are considered a potentially serious security threat for machine learning systems. Medical image analysis (MedIA) systems have recently been argued to be vulnerable to adversarial attacks due to strong financial incentives and the associated technological infrastructure. In this paper, we study previously unexplored factors affecting adversarial attack vulnerability of deep learning MedIA systems in three medical domains: ophthalmology, radiology, and pathology. We focus on adversarial black-box settings, in which the attacker does not have full access to the target model and usually uses another model, commonly referred to as surrogate model, to craft adversarial examples that are then transferred to the target model. We consider this to be the most realistic scenario for MedIA systems. Firstly, we study the effect of weight initialization (pre-training on ImageNet or random initialization) on the transferability of adversarial attacks from the surrogate model to the target model, i.e., how effective attacks crafted using the surrogate model are on the target model. Secondly, we study the influence of differences in development (training and validation) data between target and surrogate models. We further study the interaction of weight initialization and data differences with differences in model architecture. All experiments were done with a perturbation degree tuned to ensure maximal transferability at minimal visual perceptibility of the attacks. Our experiments show that pre-training may dramatically increase the transferability of adversarial examples, even when the target and surrogate's architectures are different: the larger the performance gain using pre-training, the larger the transferability. Differences in the development data between target and surrogate models considerably decrease the performance of the attack; this decrease is further amplified by difference in the model architecture. We believe these factors should be considered when developing security-critical MedIA systems planned to be deployed in clinical practice. We recommend avoiding using only standard components, such as pre-trained architectures and publicly available datasets, as well as disclosure of design specifications, in addition to using adversarial defense methods. When evaluating the vulnerability of MedIA systems to adversarial attacks, various attack scenarios and target-surrogate differences should be simulated to achieve realistic robustness estimates. The code and all trained models used in our experiments are publicly available.3.

Quantification of Key Retinal Features in Early and Late Age-Related Macular Degeneration Using Deep Learning.

PURPOSE: We sought to develop and validate a deep learning model for segmentation of 13 features associated with neovascular and atrophic age-related macular degeneration (AMD). DESIGN: Development and validation of a deep-learning model for feature segmentation. METHODS: Data for model development were obtained from 307 optical coherence tomography volumes. Eight experienced graders manually delineated all abnormalities in 2712 B-scans. A deep neural network was trained with these data to perform voxel-level segmentation of the 13 most common abnormalities (features). For evaluation, 112 B-scans from 112 patients with a diagnosis of neovascular AMD were annotated by 4 independent observers. The main outcome measures were Dice score, intraclass correlation coefficient, and free-response receiver operating characteristic curve. RESULTS: On 11 of 13 features, the model obtained a mean Dice score of 0.63 ± 0.15, compared with 0.61 ± 0.17 for the observers. The mean intraclass correlation coefficient for the model was 0.66 ± 0.22, compared with 0.62 ± 0.21 for the observers. Two features were not evaluated quantitatively because of a lack of data. Free-response receiver operating characteristic analysis demonstrated that the model scored similar or higher sensitivity per false positives compared with the observers. CONCLUSIONS: The quality of the automatic segmentation matches that of experienced graders for most features, exceeding human performance for some features. The quantified parameters provided by the model can be used in the current clinical routine and open possibilities for further research into treatment response outside clinical trials.

High Current Cycling in a Superconcentrated Ionic Liquid Electrolyte to Promote Uniform Li Morphology and a Uniform LiF-Rich Solid Electrolyte Interphase.

High-energy-density systems with fast charging rates and suppressed dendrite growth are critical for the implementation of efficient and safe next-generation advanced battery technologies such as those based on Li metal. However, there are few studies that investigate reliable cycling of Li metal electrodes under high-rate conditions. Here, by employing a superconcentrated ionic liquid (IL) electrolyte, we highlight the effect of Li salt concentration and applied current density on the resulting Li deposit morphology and solid electrolyte interphase (SEI) characteristics, demonstrating exceptional deposition/dissolution rates and efficiency in these systems. Operation at higher current densities enhanced the cycling efficiency, e.g., from 64 ± 3% at 1 mA cm-2 up to 96 ± 1% at 20 mA cm-2 (overpotential <±0.2 V), while resulting in lower electrode resistance and dendrite-free Li morphology. A maximum current density of 50 mA cm-2 resulted in 88 ± 3% cycling efficiency, displaying tolerance for high overpotentials at the Ni working electrode (0.5 V). X-ray photoelectron microscopy (XPS), time-of-flight secondary-ion mass spectroscopy (ToF-SIMS), and scanning electron microscopy (SEM) surface measurements revealed that the formation of a stable SEI, rich in LiF and deficient in organic carbon species, coupled with nondendritic and compact Li morphologies enabled enhanced cycling efficiency at higher currents. Reduced dendrite formation at high current is further highlighted by the use of a highly porous separator in coin cell cycling (1 mAh cm-2 at 50 °C), sustaining 500 cycles at 10 mA cm-2.

Iterative Augmentation of Visual Evidence for Weakly-Supervised Lesion Localization in Deep Interpretability Frameworks: Application to Color Fundus Images.

Interpretability of deep learning (DL) systems is gaining attention in medical imaging to increase experts' trust in the obtained predictions and facilitate their integration in clinical settings. We propose a deep visualization method to generate interpretability of DL classification tasks in medical imaging by means of visual evidence augmentation. The proposed method iteratively unveils abnormalities based on the prediction of a classifier trained only with image-level labels. For each image, initial visual evidence of the prediction is extracted with a given visual attribution technique. This provides localization of abnormalities that are then removed through selective inpainting. We iteratively apply this procedure until the system considers the image as normal. This yields augmented visual evidence, including less discriminative lesions which were not detected at first but should be considered for final diagnosis. We apply the method to grading of two retinal diseases in color fundus images: diabetic retinopathy (DR) and age-related macular degeneration (AMD). We evaluate the generated visual evidence and the performance of weakly-supervised localization of different types of DR and AMD abnormalities, both qualitatively and quantitatively. We show that the augmented visual evidence of the predictions highlights the biomarkers considered by experts for diagnosis and improves the final localization performance. It results in a relative increase of 11.2± 2.0% per image regarding sensitivity averaged at 10 false positives/image on average, when applied to different classification tasks, visual attribution techniques and network architectures. This makes the proposed method a useful tool for exhaustive visual support of DL classifiers in medical imaging.

An investigation of commercial carbon air cathode structure in ionic liquid based sodium oxygen batteries.

In order to bridge the gap between theoretical and practical energy density in sodium oxygen batteries challenges need to be overcome. In this work, four commercial air cathodes were selected, and the impacts of their morphologies, structure and chemistry on their performance with a pyrrolidinium-based ionic liquid electrolyte are evaluated. The highest discharge capacity was found for a cathode with a pore size ca. 6 nm; this was over 100 times greater than that delivered by a cathode with a pore size less than 2 nm. The air cathode with the highest specific surface area and the presence of a microporous layer (BC39) exhibited the highest specific capacity (0.53 mAh cm-2).

A Deep Learning Model for Segmentation of Geographic Atrophy to Study Its Long-Term Natural History.

PURPOSE: To develop and validate a deep learning model for the automatic segmentation of geographic atrophy (GA) using color fundus images (CFIs) and its application to study the growth rate of GA. DESIGN: Prospective, multicenter, natural history study with up to 15 years of follow-up. PARTICIPANTS: Four hundred nine CFIs of 238 eyes with GA from the Rotterdam Study (RS) and Blue Mountain Eye Study (BMES) for model development, and 3589 CFIs of 376 eyes from the Age-Related Eye Disease Study (AREDS) for analysis of GA growth rate. METHODS: A deep learning model based on an ensemble of encoder-decoder architectures was implemented and optimized for the segmentation of GA in CFIs. Four experienced graders delineated, in consensus, GA in CFIs from the RS and BMES. These manual delineations were used to evaluate the segmentation model using 5-fold cross-validation. The model was applied further to CFIs from the AREDS to study the growth rate of GA. Linear regression analysis was used to study associations between structural biomarkers at baseline and the GA growth rate. A general estimate of the progression of GA area over time was made by combining growth rates of all eyes with GA from the AREDS set. MAIN OUTCOME MEASURES: Automatically segmented GA and GA growth rate. RESULTS: The model obtained an average Dice coefficient of 0.72±0.26 on the BMES and RS set while comparing the automatically segmented GA area with the graders' manual delineations. An intraclass correlation coefficient of 0.83 was reached between the automatically estimated GA area and the graders' consensus measures. Nine automatically calculated structural biomarkers (area, filled area, convex area, convex solidity, eccentricity, roundness, foveal involvement, perimeter, and circularity) were significantly associated with growth rate. Combining all growth rates indicated that GA area grows quadratically up to an area of approximately 12 mm2, after which growth rate stabilizes or decreases. CONCLUSIONS: The deep learning model allowed for fully automatic and robust segmentation of GA on CFIs. These segmentations can be used to extract structural characteristics of GA that predict its growth rate.

Evaluation of a deep learning system for the joint automated detection of diabetic retinopathy and age-related macular degeneration.

PURPOSE: To validate the performance of a commercially available, CE-certified deep learning (DL) system, RetCAD v.1.3.0 (Thirona, Nijmegen, The Netherlands), for the joint automatic detection of diabetic retinopathy (DR) and age-related macular degeneration (AMD) in colour fundus (CF) images on a dataset with mixed presence of eye diseases. METHODS: Evaluation of joint detection of referable DR and AMD was performed on a DR-AMD dataset with 600 images acquired during routine clinical practice, containing referable and non-referable cases of both diseases. Each image was graded for DR and AMD by an experienced ophthalmologist to establish the reference standard (RS), and by four independent observers for comparison with human performance. Validation was furtherly assessed on Messidor (1200 images) for individual identification of referable DR, and the Age-Related Eye Disease Study (AREDS) dataset (133 821 images) for referable AMD, against the corresponding RS. RESULTS: Regarding joint validation on the DR-AMD dataset, the system achieved an area under the ROC curve (AUC) of 95.1% for detection of referable DR (SE = 90.1%, SP = 90.6%). For referable AMD, the AUC was 94.9% (SE = 91.8%, SP = 87.5%). Average human performance for DR was SE = 61.5% and SP = 97.8%; for AMD, SE = 76.5% and SP = 96.1%. Regarding detection of referable DR in Messidor, AUC was 97.5% (SE = 92.0%, SP = 92.1%); for referable AMD in AREDS, AUC was 92.7% (SE = 85.8%, SP = 86.0%). CONCLUSION: The validated system performs comparably to human experts at simultaneous detection of DR and AMD. This shows that DL systems can facilitate access to joint screening of eye diseases and become a quick and reliable support for ophthalmological experts.