Search

The properties of transition metal borides and silicides are intimately linked to the covalent character of the chemical bonds within their crystal structures. Bringing boron and silicon together within metal borosilicides can then engender different competing covalent networks and complex charge distributions. This situation results in unique structures and atomic environments, which can impact charge transport and catalytic properties. Metal borosilicides, however, hold the status of unusual exotic species, difficult to synthesize and with poor knowledge of their properties. Our strategy consists of developing a redox pathway to synthesize transition metal borosilicides in inorganic molten salts as high-temperature solvents. By studying the formation of Ni6Si2B, Co4.75Si2B, Fe5SiB2, and Mn5SiB2with in situ X-ray diffraction, we highlight how new reaction routes, maintaining covalent structural building blocks, draw a general scheme of their formation. This pathway is driven by the covalence of the chemical bonds within the boron coordination framework. Next, we demonstrate high efficiency for water oxidation electrocatalysis, especially for Ni6Si2B. We ascribe the strongly increased resistance to corrosion, high stability, and electrocatalytic activity of the Ni6Si2B-derived material to three factors: (1) the two entangled boron and silicon covalent networks; (2) the ability to codope with boron and silicon an in situ generated catalytic layer; and (3) a rare electron enrichment of the transition metal by back-donation from boron atoms, previously unknown within this compound family. With this work, we then unveil a new chemical dimension for Earth-abundant water oxidation electrocatalysts by bringing to light a new family of materials.

Obtaining valid estimates of nutrient intake in infants is currently limited by the difficulties of accurately measuring human milk intake. Current methods are either unsuitable for large-scale studies (i.e., the gold standard dose-to-mother stable isotope technique) or use set amounts, regardless of known variability in individual intake.

•Candidacan colonize and invade human surfaces, causing severe infections.•Incidence of candidemia in our study was higher than Latin American.•Non-albicans Candidaspecies were the most frequent, including C. parapsilosis.•Septic shock was the only one with a risk factor for mortality.•Implementing infection control strategies to prevent candidemia and adverse outcomes.

ABSTRACTIntroductionAsthma encompasses a spectrum of phenotypes often categorized into two groups- type 2 high (T2 high) and type 2 low (T2 low). T2 high includes atopic and eosinophilic presentations whereas T2 low is non-atopic, non-eosinophilic, and oft associated with neutrophilic inflammation. Eosinophilic asthma is often driven by IgE, IL-4, IL-5, and IL-13 and TSLP. This can lead to eosinophilic inflammatory response in the airways which in turn can be used as target for treatment.Areas coveredThe article will focus on biologic therapy that is currently being used in eosinophilic asthma management in mainly the adult population including clinical trials and co-morbidities that can be treated using the same biologics. A review on asthma biologics for pediatric population has been reviewed elsewhere.Expert opinionBiological therapy for asthma targeting the IgE, IL-4, IL-5, IL-13, and TSLP pathways are shown to have benefit for the treatment of eosinophilic asthma, as exemplified in real-world studies. When choosing the right biological agent factors such as phenotype, comorbidities, and cost-effectiveness of the biologic agent must be taken into consideration.

To analyze data available in the literature regarding a possible prognostic value of the platelet-to-lymphocyte ratio (PLR) and neutrophil-to-lymphocyte ratio (NLR) in prostate cancer (PCa) patients stratified in non-metastatic and metastatic diseases.

IMPORTANCE: Management of sporadic vestibular schwannoma with radiosurgery is becoming increasingly common globally; however, limited data currently characterize patient outcomes in the setting of microsurgical salvage for radiosurgical failure. OBJECTIVE: To describe the clinical outcomes of salvage microsurgery following failed primary stereotactic radiosurgery (SRS) or fractionated stereotactic radiotherapy (FSRT) among patients with sporadic vestibular schwannoma. DESIGN, SETTING, AND PARTICIPANTS: This was a cohort study of adults (≥18 years old) with sporadic vestibular schwannoma who underwent salvage microsurgery following failed primary SRS/FSRT in 7 vestibular schwannoma treatment centers across the US and Norway. Data collection was performed between July 2022 and January 2023, with data analysis performed between January and July 2023. EXPOSURE: Salvage microsurgical tumor resection. MAIN OUTCOMES AND MEASURES: Composite outcome of undergoing less than gross total resection (GTR) or experiencing long-term facial paresis. RESULTS: Among 126 patients, the median (IQR) age at time of salvage microsurgery was 62 (53-70) years, 69 (55%) were female, and 113 of 117 (97%) had tumors that extended into the cerebellopontine angle at time of salvage. Of 125 patients, 96 (76%) underwent primary gamma knife SRS, while 24 (19%) underwent linear accelerator-based SRS; the remaining patients underwent FSRT using other modalities. Postoperative cerebrospinal fluid leak was seen in 15 of 126 patients (12%), hydrocephalus in 8 (6%), symptomatic stroke in 7 (6%), and meningitis in 2 (2%). Each 1-mm increase in cerebellopontine angle tumor size was associated with a 13% increased likelihood of foregoing GTR (64 of 102 patients [63%]) or long-term postoperative House-Brackmann grade higher than I (48 of 102 patients [47%]) (odds ratio, 1.13; 95% CI, 1.04-1.23). Following salvage microsurgery, tumor growth–free survival rates at 1, 3, and 5 years were 97% (95% CI, 94%-100%), 93% (95% CI, 87%-99%), and 91% (95% CI, 84%-98%), respectively. CONCLUSIONS: In this cohort study, more than half of patients who received salvage microsurgery following primary SRS/FSRT underwent less than GTR or experienced some degree of facial paresis long term. These data suggest that the cumulative risk of developing facial paresis following primary SRS/FSRT by the end of the patient’s journey with treatment approximates 2.5% to 7.5% when using published primary SRS/FSRT long-term tumor control rates.

Early exit neural networks (EENs) reduce the processing times of deep convolutional neural networks by means of internal classifiers (ICs) that allow jobs, being the input of the EEN, to exit early from the processing pipeline. However, the current designs used in pervasive systems ignore variability in data arrival rates, exposing EEN-based services to potential loss of the incoming jobs, due to finite input buffer capacity. Motivated by this issue, we introduce and study the early exit scheduling problem, which aims at dynamically configuring IC thresholds at runtime to achieve effective trade-offs between job classification accuracy, processing time, and job loss ratio. We argue that deciding the EEN exit layer for a job at the start of its processing makes the problem mathematically tractable, allowing us to develop policies to control buffer backlog, classification accuracy, and processing time across the EEN layers. The main contribution of the paper is the introduction of single-exit IC threshold configurations as a mechanism to allow the scheduling policy to reliably predict the best EEN exit layer of each input job. Three scheduling policies that leverage this idea are proposed to dynamically schedule job arrivals to an EEN-based service. The proposed solution, here tailored to EENs based on convolutional neural networks (CNNs), is fairly general and can be applied to different use cases. The two application scenarios considered in this paper focus on image classification and intrusion detection. Experiments on some popular CNNs for the two aforementioned application scenarios indicate that the proposed policies can achieve significant savings in processing times and improve job loss ratio compared to both ordinary EENs and CNNs while still providing high mean classification accuracy.

Emerging serverless computing technologies, such as function as a service (FaaS), enable developers to virtualize the internal logic of an application, simplifying the management of cloud-native services and allowing cost savings through billing and scaling at the level of individual functions. Serverless computing is therefore rapidly shifting the attention of software vendors to the challenge of developing cloud applications deployable on FaaS platforms. In this vision paper, we present the research agenda of the RADON project (http://radon-h2020.eu), which aims to develop a model-driven DevOps framework for creating and managing applications based on serverless computing. RADON applications will consist of fine-grained and independent microservices that can efficiently and optimally exploit FaaS and container technologies. Our methodology strives to tackle complexity in designing such applications, including the solution of optimal decomposition, the reuse of serverless functions as well as the abstraction and actuation of event processing chains, while avoiding cloud vendor lock-in through models.

BackgroundSocial cognition (SC) deficits are included in the amyotrophic lateral sclerosis-frontotemporal spectrum disorder (ALS-FTDS) revised diagnostic criteria. However, the impact of SC assessment on cognitive classification and the cognitive–behavioural correlates of SC remain unclear. This cross-sectional study aimed to assess the impact of SC assessment on ALS-FTDS categorisation and explore the relationship of SC with executive functions (EF) and behaviour changes in a cohort of ALS patients.Methods121 patients and 56 healthy controls from the Turin ALS Centre underwent cognitive/behavioural testing, including the SC subdomains of facial emotion recognition, and cognitive and affective theory of mind (ToM).ResultsPatients performed significantly worse than controls in all SC explored domains, and 45% of patients exhibited a deficit in at least one SC test, dissociated from the presence of EF deficits. In 13% of cases, the SC deficit was isolated and subclinical. SC assessment contributed to the attribution of cognitive impairment in 10% of patients. Through a statistical clustering approach, we found that ToM only partially overlaps with EF while behaviour changes are associated with emotional disorders (anxiety and depression).ConclusionsSC is overall independent of EF in ALS, with ToM only partially associated with specific EF measures, and behaviour changes associated with emotional disorders. The influence of SC on cognitive categorisation and the frequent identification of a subclinical SC impairment have implications in a clinical setting, considering the substantial impact of cognitive impairment on disease burden and therapeutic choices.

ObjectiveGlycaemia in newborns changes significantly after birth; however, little is known about these changes. The objective was to describe continuous interstitial glucose values in term newborns who were exclusively breast fed on the first day of life.DesignWe studied 159 newborns with appropriate weights for gestational age, who were exclusively breast fed on the first day of life, using a continuous glucose monitoring device that calculates interstitial glucose every 5 min. The device was removed after 24 hours, and the results were analysed using the R program, which provides the minimum, maximum, median and a standard curve with centiles.ResultsAt the second hour of life, the moment in which the sensor started to identify the newborn’s glycaemia, interstitial glucose levels were 2.59–4.43 mmol/L (46.7–79.9 mg/dL). The median interstitial glucose level of the newborns during the first day of life was 3.33±0.48 mmol/L (60±8.6 mg/dL). Interstitial glucose levels dropped until the sixth hour of life, reaching 2.19–3.95 mmol/L (39.5–71.1 mg/dL), and then increased again. The maximum values were found at the 20th and 21st hours of life, which were 2.81–4.64 mmol/L (50.6–83.6 mg/dL).ConclusionThe interstitial glucose during the first 24 hours of life declined until the sixth hour of life, then increased around the 20th hour and remained stable until the end of the first day of life.

Typical mobile edge computing infrastructures have to contend with unreliable computing devices at their end-points. The limited resource capacities of mobile edge devices gives rise to frequent contentions, node overloads or failures. This is exacerbated by the strict deadlines of modern applications. To avoid failures, fault-tolerant approaches utilize preemptive migration to transfer active tasks across nodes and prevent nodes running at capacity. However, prior work struggles to dynamically adapt in settings with highly volatile workloads or even accurately detect and diagnose anomalies for optimal remediation. To meet the strict service level objectives of contemporary workloads, there is a need for dynamic fault-tolerant methods that can quickly adapt to changes in edge environments while having parsimonious remediation in the form of preemptive migration to avoid stressing the system network. This work proposes PreGAN, featuring a Generative Adversarial Network (GAN) based approach to predict contentions, pinpoint specific resource types with high chance of overload, and generate migration decisions to proactively avoid system downtime. PreGAN leverages coupled-simulations to train the GAN model at run-time and a few-shot fault classifier to update decisions of an underpinning scheduler. We also extend it to PreGAN+ that also periodically tunes the decision model using semi-supervised training and a Transformer based neural network for low tuning time, albeit with higher memory overheads. Experiments on a Raspberry-Pi based edge environment demonstrate that both models outperform state-of-the-art baselines in fault detection and diagnosis scores by up to 12.5% and 31.2% respectively. This also translates in improvements in Quality of Service against baseline approaches.

Layered queueing networks (LQNs) are a class of performance models for software systems in which multiple distributed resources may be possessed simultaneously by a job. Estimating response times in a layered system is an essential but challenging analysis dimension in Quality of Service (QoS) assessment. Current analytic methods are capable of providing accurate estimates of mean response times. However, accurately approximating response time distributions used in service-level objective analysis is a demanding task. This paper proposes a novel hybrid framework that leverages phase-type (PH) distributions and neural networks to provide accurate density estimates of response times in layered queueing networks. The core step of this framework is to recursively obtain response time distributions in the submodels that are used to analyze the network by means of decomposition. We describe these response time distributions as a mixture of density functions for which we learn the parameters through a Mixture Density Network (MDN). The approach recursively propagates MDN predictions across software layers using PH distributions and performs repeated moment-matching based refitting to efficiently estimate end-to-end response time densities. Extensive numerical experiment results show that our scheme significantly improves density estimations compared to the state-of-the-art.

Background Physiotherapy in Palliative Care (PC) is effective in the management of a series of respiratory, circulatory and motor symptoms, and often has a positive impact on the patient's mood. The Edmonton Functional Assessment Tool (EFAT) is the only existing validated tool specifically designed for functional assessment in PC, and its use has been recommended in clinical practice. To date, no Italian version of the tool has been validated. The aim of this study was to translate, cross-culturally adapt, and evaluate the psychometric properties of the Italian version of the EFAT2.Method After receiving formal permission from the author, Beaton guidelines for cross-cultural adaptation were followed, namely: (1) forward translation; (2) a multidisciplinary focus group (including 4 physiotherapists, 1 physician, 3 nurses, 1 occupational therapist, 1 psychologist) to assess semantic, idiomatic, experiential, and conceptual equivalence; (3) backward translation. The Content Validity Index (CVI) was used to assess content validity of the tool. Construct and concurrent validity were also evaluated. To evaluate the reliability of the EFAT2-I, reliability was measured using Cronbach alfa, item-total correlation, and Cohen's Kappa.Results 119 patients admitted to a Palliative Care Unit (Italy) agreed to participate in the study. The EFAT2-I mean score was 11.3, ranging from 0 to 30. Very good CVI scores were achieved, both in terms of single item validity (I-CVI) and of whole scale validity (S-CVI). Positive results were obtained from construct, concurrent validity assessment and measures of reliability.Discussion The EFAT2-I showed good psychometric properties and can be used as a rehabilitation assessment tool in palliative care settings. The validation of the Italian version will allow comparison of different centres and palliative care facilities on national and international levels.

TRANSCRIBEDDevoAs heard on FREEDOM OF CHOICEPHOTO (COLOR)PHOTO (COLOR)PHOTO (COLOR)PHOTO (COLOR)PHOTO (COLOR)PHOTO (COLOR)PHOTO (COLOR)PHOTO (COLOR)PHOTO (COLOR)PHOTO (COLOR) [Extracted from the article]

Purpose: This research study provides insight into students’ perceptions of teaching through virtual and face-to-face clinicals in an introductory education course in a pre-education program at a minority-serving institution. Design/methodology/approach: This study took place at an urban–suburban-centered community college in the Midwestern United States and was reviewed by the higher education institutional review board (IRB). Data were collected from pre-education majors enrolled in a four-hour Introduction to Education with field experiences. Findings: The findings indicated that both virtual and face-to-face clinicals were beneficial to the development of pre-service teachers, particularly in an early introduction to education course. Research limitations/implications: The finding that virtual clinicals are significant to teacher growth is significant to teacher recruitment and preparation. Practical implications: The flexibility of a virtual clinical provides greater opportunities for low-income and marginalized populations with limited means and access. Social implications: This finding can lead to strategies to diversify teacher candidates. Originality/value: This study sought to answer the following question: how do pre-education students reflect to understand the roles and responsibilities of teaching through virtual options vs face-to-face clinicals? The interest of this research is to expand pathways into the teaching profession to nontraditional, ethnically and culturally marginalized groups and historically underrepresented groups.

The rise of distributed cloud computing technologies has been pivotal for the large-scale adoption of Artificial Intelligence (AI) based applications for high fidelity and scalable service delivery. Systematic resource management is central in maintaining optimal Quality of Service (QoS) in cloud platforms and is divided into three fundamental types: resource provisioning, AI model deployment and workload placement. To exploit the synergy among these decision types, it becomes imperative to concurrently design (co-design) the provisioning, deployment and placement decisions for optimal QoS. As users and cloud service providers shift to non-stationary AI-based workloads, frequent decision making imposes severe time constraints on the resource management models. Existing AI-based solutions often optimize decision types independently and tend to ignore the dependencies across various system performance aspects such as energy consumption and CPU utilization, making them perform poorly in large-scale cloud systems. To address this, we propose a novel method, called SciNet, that leverages a co-simulated digital-twin of the infrastructure to capture inter-metric dependencies and accurately estimate QoS scores. To avoid expensive simulation overheads at test time, SciNet trains a neural network based imitation learner that aims to mimic an oracle, which takes optimal decisions based on co-simulated QoS estimates. Offline model training and online decision making based on the imitation learner, enables SciNet to take optimal decisions while being time-efficient. Experiments with real-life AI-based benchmark applications on a public cloud testbed show that SciNet gives up to 48% lower execution cost, 79% higher inference accuracy, 71% lower energy consumption and 56% lower response times compared to the current state-of-the-art methods.

To construct a symptoms-based prediction tool to assess the likelihood of superior canal dehiscence (SSCD) on high-resolution CT.

Rasmussen Encephalitis (RE) is a rare inflammatory neurodegenerative disease associated with refractory seizures, hemiparesis, and cognitive deterioration, due to lateralized cortical atrophy. Hemispheric surgery (hemispherotomy) is the mainstay of treatment, but its unavoidable motor deficits and lack of long‐term data regarding seizure outcomes can make patients and families apprehensive to undergo this procedure. The present study aimed at analyzing the results of surgical treatment for RE from a motor and epilepsy standpoint, and mitigate such concerns. Clinical and operative data were retrospectively collected from medical records of pharmacoresistant patients treated with functional hemispherectomy at a tertiary reference center for epilepsy surgery, during a 24‐year period (1996–2020). Variables such as age of epilepsy onset, seizure semiology, seizure frequency, immunomodulatory therapy, age at surgery, duration of epilepsy, surgical procedures and complications, number of medications used preoperatively and postoperatively were described and statistically analyzed. Forty‐three (43) patients were included in this study. Mean age of epilepsy onset was 6.14 years, the average interval between epilepsy onset and hemispherotomy was 2.21 years. and the mean age at surgery was 8.28 years. Thirty patients (69.7%) were Engel I at their last follow‐up, of whom 23 (56.4%) were Engel Ia, within a mean follow‐up of 11.3 years. Duration of epilepsy, seizure frequency, and age at surgery, among others, did not correlate with seizure outcome, except the use of immunotherapy which led to worse outcomes (p< .05). Also, after surgery, motor functionality was significantly recovered (i.e., most patients returned to their previous status) with time. This study tackled some issues regarding the surgical treatment of this disease, particularly showing that hemispherotomy is safe and leads to potentially recoverable disability of motor functions while providing high rates of effective and long‐lasting seizure control; therefore, early surgical indication should be warranted once medical refractoriness has been established.

Nanoscale heterostructures of covalent intermetallics should give birth to a wide range of interface-driven physical and chemical properties. Such a level of design however remains unattainable for most of these compounds, due to the difficulty to reach a crystalline order of covalent bonds at the moderate temperatures required for colloidal chemistry. Herein, we design heterostructured cobalt silicide nanoparticles to trigger magnetic and catalytic properties in silicon-based materials. Our strategy consists in controlling the diffusion of cobalt atoms into silicon nanoparticles, by reacting these particles in molten salts. By adjusting the temperature, we tune the conversion of the initial silicon particles toward homogeneous CoSi nanoparticles and core–shell nanoparticles made of a CoSi shell and a silicon-rich core. The increased interface-to-volume ratio of the CoSi component in the core–shell particles yields distinct properties compared to the bulk and homogeneous nanoparticles. First, the core–shell particles exhibit increased ferromagnetism, despite the bulk diamagnetic properties of cobalt monosilicide. Second, the core–shell nanoparticles act as efficient precatalysts for alkaline water oxidation, where the nanostructure is converted in situ into a layered cobalt silicon oxide/(oxy)hydroxide with high and stable oxygen evolution reaction (OER) electrocatalytic activity. This work demonstrates a route to design heterostructured nanocrystals of covalent intermetallic compounds and shows that these new structures exhibit very rich, yet poorly explored, interface-based physical properties and reactivity.

In recent years, deep learning models have become ubiquitous in industry and academia alike. Deep neural networks can solve some of the most complex pattern-recognition problems today, but come with the price of massive compute and memory requirements. This makes the problem of deploying such large-scale neural networks challenging in resource-constrained mobile edge computing platforms, specifically in mission-critical domains like surveillance and healthcare. To solve this, a promising solution is to split resource-hungry neural networks into lightweight disjoint smaller components for pipelined distributed processing. At present, there are two main approaches to do this: semantic and layer-wise splitting. The former partitions a neural network into parallel disjoint models that produce a part of the result, whereas the latter partitions into sequential models that produce intermediate results. However, there is no intelligent algorithm that decides which splitting strategy to use and places such modular splits to edge nodes for optimal performance. To combat this, this work proposes a novel AI-driven online policy, SplitPlace, that uses Multi-Armed-Bandits to intelligently decide between layer and semantic splitting strategies based on the input task's service deadline demands. SplitPlace places such neural network split fragments on mobile edge devices using decision-aware reinforcement learning for efficient and scalable computing. Moreover, SplitPlace fine-tunes its placement engine to adapt to volatile environments. Our experiments on physical mobile-edge environments with real-world workloads show that SplitPlace can significantly improve the state-of-the-art in terms of average response time, deadline violation rate, inference accuracy, and total reward by up to 46, 69, 3 and 12 percent respectively.

Background: T1 high-grade (HG) non-muscle invasive bladder cancer (NMIBC) has a significant risk of recurrence and progression, and the European Association of Urology recommends a second transurethral resection of the bladder (ReTUR). Stage at ReTUR has been shown to be a reliable predictor of survival, therefore, we sought to assess clinical and pathological predictors associated with the persistence of T1 at ReTUR in our retrospective multicentric cohort.Methods: This is a retrospective multicentric study of T1 HG patients at transurethral resection of the bladder (TURB) who underwent subsequent ReTUR. All histological samples were sub-classified according to Rete Oncologica Lombarda (ROL) T1 sub-staging system.Results: One hundred and sixty-six patients were enrolled. Forty-four (26.5%) had T1 HG tumor at ReTUR while 93 (56%) had residual tumor of any stage. Lesion size was significantly greater in T1 HG patients at ReTUR, as well as the prevalence of multifocality. The multivariable logistic regression model showed lesion dimension and multifocality as predictors of T1 HG at ReTUR, after adjusting for significant covariables (CIS and detrusor muscle presence). ROL sub-staging system was not a significant predictor, but ROL2 prevalence was higher in the T1 HG at ReTUR group.Conclusions: Lesion size and multifocality were independent predictors of T1 HG persistence at ReTUR, and patients at risk should be promptly identified and treated accordingly. Our results could help physicians make patient-tailored decisions by identifying those most likely to benefit from a second resection.