Network Of Experts Research Articles

Haemovigilance resources available for everyone: A report from the International Haemovigilance Tools Collaborative Project.

Haemovigilance systems are intended to collect and analyse data, and report findings relating to transfusion complications, such as blood product safety, procedural incidents, and adverse reactions in donors and patients. A common problem among developing haemovigilance programs is the lack of resources and tools available to countries striving to establish or enhance their haemovigilance system. World Health Organization, in collaboration with International Society for Blood Transfusion (ISBT), International Haemovigilance Network and other haemovigilance experts embarked on a Haemovigilance Tools Project to collect and provide materials and resources to assist with the stepwise implementation of haemovigilance. Resources are housed as a virtual compendium on the ISBT website under the Haemovigilance Working Party. These are managed by a subcommittee of the Working Party and are freely available and downloadable to all without requiring ISBT membership.

Dynamic polarization fusion network (DPFN) for imaging in different scattering systems.

Deep learning has broad applications in imaging through scattering media. Polarization, as a distinctive characteristic of light, exhibits superior stability compared to light intensity within scattering media. Consequently, the de-scattering network trained using polarization is expected to achieve enhanced performance and generalization. For getting optimal outcomes in diverse scattering conditions, it makes sense to train expert networks tailored for each corresponding condition. Nonetheless, it is often unfeasible to acquire the corresponding data for every possible condition. And, due to the uniqueness of polarization, different polarization information representation methods have different sensitivity to different environments. As another of the most direct approaches, a generalist network can be trained with a range of polarization data from various scattering situations, however, it requires a larger network to capture the diversity of the data and a larger training set to prevent overfitting. Here, in order to achieve flexible adaptation to diverse environmental conditions and facilitate the selection of optimal polarization characteristics, we introduce a dynamic learning framework. This framework dynamically adjusts the weights assigned to different polarization components, thus effectively accommodating a wide range of scattering conditions. The proposed architecture incorporates a Gating Network (GTN) that efficiently integrates multiple polarization features and dynamically determines the suitable polarization information for various scenarios. Experimental result demonstrates that the network exhibits robust generalization capabilities across continuous scattering conditions.

Tag-based self-learning task recommendation for mobile crowdsensing via collaborative multi-expert system

With the powerful sensing capabilities of mobile smart devices, sensing users can complete mobile crowd sensing tasks through smart devices in the Internet of Things. When a user completes a sensing task, the sensing user may want to complete more related items from the previous task as an extended task. However, it is difficult for traditional recommender systems to satisfy sensing user needs for tasks because the recommendation results should be diverse. In this paper, we propose a multi-expert tag recommendation framework (MTRS) with the help of tags for sensing tasks. The main purpose is to predict whether the sensing user wants to extend the task, and provide appropriate related tasks as suggestions. The framework mainly consists of two parts: recommendation migration and recommendation purification. We extract features from multiple aspects. In the information gain expert network, information gain is obtained through the relationship between tags. In the feature interaction expert network, multi-head self-attention is used to model the feature interaction between different feature fields. In similar, the relevance between tags is explicitly emphasized in the degree expert network. Feature information is obtained through different expert networks, and a multi-critic strategy is proposed to judge the importance of each expert network. At the same time, different optimization goals are set for recommendation migration and recommendation purification. In recommendation purification, the user's satisfaction with task tags and the delay cost of task recommendation are specially considered. Finally, experiments demonstrate the correctness, effectiveness and robustness of the MTRS approach.

Stakeholder perspectives from 15 countries in Africa on barriers in snakebite envenoming research and the potential role of research hubs.

Snakebite envenoming is a debilitating neglected tropical disease disproportionately affecting the rural poor in low and middle-income countries in the tropics and sub-tropics. Critical questions and gaps in public health and policy need to be addressed if major progress is to be made towards reducing the negative impact of snakebite, particularly in the World Health Organisation (WHO) Africa region. We engaged key stakeholders to identify barriers to evidence-based snakebite decision making and to explore how development of research and policy hubs could help to overcome these barriers. We conducted an electronic survey among 73 stakeholders from ministries of health, health facilities, academia and non-governmental organizations from 15 countries in the WHO Africa region. The primary barriers to snakebite research and subsequent policy translation were limited funds, lack of relevant data, and lack of interest from policy makers. Adequate funding commitment, strong political will, building expert networks and a demand for scientific evidence were all considered potential factors that could facilitate snakebite research. Participants rated availability of antivenoms, research skills training and disease surveillance as key research priorities. All participants indicated interest in the development of research and policy hubs and 78% indicated their organization would be willing to actively participate. In conclusion, our survey affirms that relevant stakeholders in the field of snakebite perceive research and policy hubs as a promising development, which could help overcome the barriers to pursuing the WHO goals and targets for reducing the burden of snakebite.

Quick spreading of the exotic amphipod Laticorophium baconi (Shoemaker, 1934): another small stowaway overlooked?

Studies of non-indigenous species (NIS) often tend to focus on medium and large-sized taxa with potential for remarkable ecological and/or economic impact, whereas the early detection of small invertebrates is often delayed due to taxonomic challenge, lack of consistent, standardised monitoring efforts and limited funding. This study represents the first records of the marine amphipod Laticorophium baconi (Shoemaker, 1934) in Morocco, Tunisia, Corsica (France), Italy, Greece, Egypt, Saudi Arabia and New Caledonia (France). Furthermore, it reports an expansion of its known distribution in Portugal (earliest and northernmost record for the country and first record for Macaronesia), Spain (earliest record for Atlantic and Mediterranean European waters), and Australia (first record for Indian Ocean). Recreational boating and commercial shipping, mainly through hull fouling and secondarily ballast waters, are proposed as vectors for introduction and secondary spread of L. baconi. The following traits, analysed during the present study, could contribute to its invasive potential: (i) quick and extensive spread of the species worldwide, (ii) high densities in marinas, harbours, hull fouling and other artificial habitats, including aquaculture facilities and floating debris, (iii) high ability for short-term colonisation of empty artificial niches, (iv) diet based on detritus suggesting an opportunistic feeding behaviour, and (v) population survival during seasonal fluctuations in different regions. Taxonomic expertise and scientific collaboration, based on multidisciplinary networks of experts, are crucial for the early detection, distribution updates, and risk assessment of small and overlooked stowaways in marine environments.

Making sense of the COVID-19 crisis: information-seeking practices and attitudes towards information providers among Baltic audiences

ABSTRACT Drawing on a comparative cross-country survey and qualitative interview data, this article explores audiences’ information-seeking practices and sentiments related to news media and other information providers in Estonia and Latvia during the COVID-19 crisis. It concludes that members of both the ethno-linguistic majority and the Russian-speaking minority relied primarily on their social network and local experts to make sense of the pandemic, with media playing a rather moderate role. Russian speakers reoriented their news media repertoires away from Russian television toward local news providers. Despite popular sentiments of media skepticism in both countries, audiences in Estonia express more confidence in local news media. Alongside ethno-linguistic divides, the study illuminates socio-economic rifts in media-society relations.

Interagency deployment of a shared low‐cost flood monitoring system to improve flood resilience across Southeast Texas: A case study

AbstractIn 2021, eight counties in Southeast Texas formed a non‐tax authority flood control partnership to provide a forum for communication, create a regional database and improve flood modeling, early warning, and flood mitigation decisions. This partnership is able to successfully deploy and manage an early warning flood system network at relatively low‐cost by: (i) assigning the university as technical expert and general network administrator so that no additional agency workforce is required to operate and maintain the system, (ii) the flood monitoring system is a simple system and (iii) the data was able to be displayed in real‐time using an existing web‐based dashboard owned and operated by another agency within the partnership. The low‐cost sensors are lightweight and easily mountable which allowed them to be rapidly deployed within a short period of time (73 sensor sites installed within 7‐month window) over a 6000 square mile area. The data is shared with the public, and local, state, and federal agencies to monitor drainage infrastructure, and provide the community with flash flood and roadway closure alerts.

Predicting the next pandemic: VACCELERATE ranking of the WorldHealth Organization's Blueprint forAction toPreventEpidemics

IntroductionThe World Health Organization (WHO)'s Research and Development (R&D) Blueprint for Action to Prevent Epidemics, a plan of action, highlighted several infectious diseases as crucial targets for prevention. These infections were selected based on a thorough assessment of factors such as transmissibility, infectivity, severity, and evolutionary potential. In line with this blueprint, the VACCELERATE Site Network approached infectious disease experts to rank the diseases listed in the WHO R&D Blueprint according to their perceived risk of triggering a pandemic. VACCELERATE is an EU-funded collaborative European network of clinical trial sites, established to respond to emerging pandemics and enhance vaccine development capabilities. MethodsBetween February and June 2023, a survey was conducted using an online form to collect data from members of the VACCELERATE Site Network and infectious disease experts worldwide. Participants were asked to rank various pathogens based on their perceived risk of causing a pandemic, including those listed in the WHO R&D Blueprint and additional pathogens. ResultsA total of 187 responses were obtained from infectious disease experts representing 57 countries, with Germany, Spain, and Italy providing the highest number of replies. Influenza viruses received the highest rankings among the pathogens, with 79 % of participants including them in their top rankings. Disease X, SARS-CoV-2, SARS-CoV, and Ebola virus were also ranked highly. Hantavirus, Lassa virus, Nipah virus, and henipavirus were among the bottom-ranked pathogens in terms of pandemic potential. ConclusionInfluenza, SARS-CoV, SARS-CoV-2, and Ebola virus were found to be the most concerning pathogens with pandemic potential, characterised by transmissibility through respiratory droplets and a reported history of epidemic or pandemic outbreaks.

CoSoGMIR: A Social Graph Contagion Diffusion Framework using the Movement-Interaction-Return Technique

Besides the inherent benefits of exchanging information and interactions between nodes on a social graph, they can also become a means for the propagation of knowledge. Social graphs have also become a veritable structure for the spread of disease outbreaks. These and its set of protocols are deployed as measures to curb its widespread effects as it has also left network experts puzzled. The recent lessons from the COVID-19 pandemic continue to reiterate that diseases will always be around. Nodal exposure, adoption/diffusion of disease(s) among interacting nodes vis-a-vis migration of nodes that cause further spread of contagion (concerning COVID-19 and other epidemics) has continued to leave experts bewildered towards rejigging set protocols. We model COVID-19 as a Markovian process with node targeting, propagation and recovery using migration-interaction as a threshold feat on a social graph. The migration-interaction design seeks to provision the graph with minimization and block of targeted diffusion of the contagion using seedset(s) nodes with a susceptible-infect policy. The study results showed that migration and interaction of nodes via the mobility approach have become an imperative factor that must be added when modeling the propagation of contagion or epidemics.

Expert networks, alternative data and managing risks of material non-public information

Expert networks and alternative data providers can be sources of invaluable information with which to make investment decisions, and they can play an important role in maximising returns from financial markets, but with the use of the sophisticated information they provide comes certain risks of regulatory scrutiny and even potential investigation for insider trading. However, a well-designed compliance programme can be tailored to monitor for, prevent, and detect red flags to avoid investigations and charges by financial enforcement authorities. Due diligence, the establishment of proper policies and procedures, and adherence to best practices will make the difference, not only in avoiding prohibited trading but in ameliorating the potential consequences if allegations of illegal behaviour arise. Financial advisers looking for an edge are well advised, themselves, to understand the risks and prepare against them.

The Protein Gap: The Rise and Fall of a Charismatic Nutrient in International Public Health.

From the early 1950s to the early 1970s, international nutritionists considered childhood protein malnutrition the world's most serious public health threat. By 1974, many believed that this "protein gap" had been exaggerated. Two questions remain: why protein, and why this period? Four converging developments created a network that maintained protein's "charisma": new food technology, a growing international health infrastructure, the nominal demise of eugenics, and new geopolitical priorities in a world shaped by both the Cold War and decolonization struggles. A transnational network of nutrition experts argued that protein deficiencies could explain bodily and population differences that would have, in an earlier era, been attributed to race or inheritance. Protein malnutrition could help explain "backward" economies and cultures, they claimed, and protein supplementation would help spur development. The protein gap theory thus framed difference in the language of modernization theory, but left intact older hierarchies of bodies, nations, and races.

Container port truck dispatching optimization using Real2Sim based deep reinforcement learning

In marine container terminals, truck dispatching optimization is often considered as the primary focus as it provides crucial synergy between the sea-side operations and yard-side activities and hence can greatly affect the terminal throughput and quay crane utilization. However, many existing studies rely on strong assumptions that often overlook the uncertainties and dynamics innate to real-life applications. In this work, we propose a dynamic truck dispatching system for container ports equipped with the latest IoT technologies. The system is comprised of Real2Sim simulation and a truck dispatch agent, trained through a spatial-attention based deep reinforcement learning module, supported by an expert network. The proposed Real2Sim framework has the ability to model the non-linear complexities and non-deterministic events while our attention-aware deep reinforcement learning module is capable of making full use of both historical and real-time port data to learn a high-quality truck dispatching policy under uncertainties. Extensive experiments show our proposed method has good generalization and achieves the state-of-the-art results on the problems derived from real-life data of a large international port.

DMMP: A distillation-based multi-task multi-tower learning model for personalized recommendation

Recommendation systems involve the matching and ranking stages. The click-through rate (CTR) and the conversion rate (CVR) predictions are two fundamental modules in recommendation systems. Most candidate generation matching models leverage a two-tower architecture to model the CTR prediction task. However, items with low-quality but attractive titles, i.e., click baits, may be recommended to the user, which worsens the user’s experience. Therefore, both click and conversion tasks should be modeled during the matching stage to improve user engagement. An intuitive way is to model these two tasks with a three-tower matching model. However, its efficiency and effectiveness are limited. By inheriting the merits of knowledge distillation and multi-task learning, we propose a distillation-based multi-task multi-tower model (DMMP) for personalized recommendation. Specifically, the MTL-based teacher network builds the task-shared and task-specific expert networks, and employs a customized multi-gate control network to merge these expert networks adaptively within the task embedding layer. Two auxiliary pCTR and pCTCVR tasks are incorporated to model CVR directly across the whole space. By sharing the feature representation parameters with the CTR modeling, the CVR modeling can be trained with richer samples. Three-tower-based candidate generation student network incorporates a user preference gating network to learn the task scores so that personalized candidates can be produced. Furthermore, we design an adaptive weighting strategy for total loss to eventually adjust the importance and relevance of different networks. Extensive experiments on public and industrial datasets validate the effectiveness of DMMP.

KGDiff: towards explainable target-aware molecule generation with knowledge guidance.

Designing 3D molecules with high binding affinity for specific protein targets is crucial in drug design. One challenge is that the atomic interaction between molecules and proteins in 3D space has to be taken into account. However, the existing target-aware methods solely model the joint distribution between the molecules and proteins, disregarding the binding affinities between them, which leads to limited performance. In this paper, we propose an explainable diffusion model to generate molecules that can be bound to a given protein target with high affinity. Our method explicitly incorporates the chemical knowledge of protein-ligand binding affinity into the diffusion model, and uses the knowledge to guide the denoising process towards the direction of high binding affinity. Specifically, an SE(3)-invariant expert network is developed to fit the Vina scoring functions and jointly trained with the denoising network, while the domain knowledge is distilled and conveyed from Vina functions to the expert network. An effective guidance is proposed on both continuous atom coordinates and discrete atom types by taking advantages of the gradient of the expert network. Experiments on the benchmark CrossDocked2020 demonstrate the superiority of our method. Additionally, an atom-level explanation of the generated molecules is provided, and the connections with the domain knowledge are established.

National clinical Genetic Networks - GENets - Establishment of expert collaborations in Denmark

Genetic conditions are often familial, but not all relatives receive counseling from the same institution. It is therefore necessary to ensure consistency in variant interpretation, counseling practices, and clinical follow up across health care providers. Furthermore, as new possibilities for gene-specific treatments emerge and whole genome sequencing becomes more widely available, efficient data handling and knowledge sharing between clinical laboratory geneticists and medical specialists in clinical genetics are increasingly important.In Denmark, these needs have been addressed through the establishment of collaborative national networks called Genetic Expert Networks or "GENets". These networks have enhanced patient and family care significantly by bringing together groups of experts in national collaborations. This promotes coordinated clinical care, the dissemination of best clinical practices, and facilitates the exchange of new knowledge.

Expert-guided Knowledge Distillation for Semi-supervised Vessel Segmentation.

In medical image analysis, blood vessel segmentation is of considerable clinical value for diagnosis and surgery. The predicaments of complex vascular structures obstruct the development of the field. Despite many algorithms have emerged to get off the tight corners, they rely excessively on careful annotations for tubular vessel extraction. A practical solution is to excavate the feature information distribution from unlabeled data. This work proposes a novel semi-supervised vessel segmentation framework, named EXP-Net, to navigate through finite annotations. Based on the training mechanism of the Mean Teacher model, we innovatively engage an expert network in EXP-Net to enhance knowledge distillation. The expert network comprises knowledge and connectivity enhancement modules, which are respectively in charge of modeling feature relationships from global and detailed perspectives. In particular, the knowledge enhancement module leverages the vision transformer to highlight the long-range dependencies among multi-level token components; the connectivity enhancement module maximizes the properties of topology and geometry by skeletonizing the vessel in a non-parametric manner. The key components are dedicated to the conditions of weak vessel connectivity and poor pixel contrast. Extensive evaluations show that our EXP-Net achieves state-of-the-art performance on subcutaneous vessel, retinal vessel, and coronary artery segmentations. Code is available at https://github.com/shennbit/EXP-Net.

Quantitative tools in microbial and chemical risk assessment.

The popularity of biological origin food protection substances is driven by demands from consumers for natural and clean label product, increasing various food-related safety and health concerns and sustainability issues. Lactic acid bacteria (LAB) are most promising because they are a large group of beneficial microorganisms commonly used in food protection due to their ability to inhibit the growth of pathogenic bacteria and enhance food safety. Extensive scientific research has been conducted to understand the mechanisms by which LAB exert their protective effects in various food systems. Even though LAB activity against various food pathogens and spoilers is distinguished, use of cell-free supernatant (CFS) is still under investigation. This report is dedicated to present how qualitative measures can elaborate in new bacteria-origin food additive investigation. As part of the EU-FORA programme, the fellow was involved in the risk assessment tasks and projects which include gaining basic knowledge in predicative microbiology fundamentals, including different types of modelling strategies; delivering essential understanding about experimental design, knowledge in three specific software tools (MATLAB, GInaFiT and DMFit) and gained overall understanding what are the main differences while modelling growth or inactivation models. Secondary activities were included as a way to expand competences beyond qualitative measures to overall all activities done regarding risk assessment and build a strong network of food safety experts and professionals to continue engaging in risk assessment beyond fellowship programme.

Preface

IC3DDose 2022 - 12th International Conference on 3D and Advanced Dosimetry The 12th IC3DDose conference on advanced radiation dosimetry was held in person in Quebec City, Quebec, Canada from June 20 to 23, 2022. The meeting was scheduled back-to-back with the Canadian Organization of Medical Physicists’ (COMP) annual scientific meeting, offering unique opportunities for scientific knowledge sharing and extensive networking. It was a refreshing restart for the IC3DDose community after the cancelled meeting in 2020 and a reduced scope virtual meeting in 2021.A total of 71 participants from 11 countries met for 3.5 days in a mix of refresher/review sessions and proffered presentations covering state-of-the-art dosimetry encompassing all major dosimetry techniques used for standard external beam and brachytherapy. Presenters also covered advanced applications addressing the challenges of radiation measurement in small field, MR-Linac, microbeam and FLASH dosimetry. A total of 14 invited presentations and 25 contributed talks were given the at meeting including papers in a Rising Star Competition and in a single hybrid session with virtual participants from Australia, China and Japan. The associated papers presented in these proceedings add to an already extensive collection of 118 review and 574 proffered papers in 9 previous volumes of the Journal of Physics Conference Series from the past conferences.A feature for this year’s meeting was that all contributed speakers on a given day were asked to mount their presentation slides on poster boards for a dedicated closing discussion session entitled “Meet the speakers of the day”. This allowed attendees to engaged in deeper discussions, beyond the question periods at the end of their talks. The daily sessions reinforced the IC3DDose philosophy to foster strong mingling of all participants, from trainees to more experienced attendees. A keynote address and the final morning of sessions were open to COMP registrants also. This made the topics of the IC3DDose meeting available to a much wider audience.The meeting was a great success and the attendees developed new expertise and new networks through the many interactions over the meeting’s four days. Colleagues from Aarhus Denmark announced at the end of the meeting that they will now undertake to organize a 13th meeting in 2024.The local organizing committee sincerely thanks the sponsors for our meeting: Elekta and Varian as bronze sponsor and MedScint in the start-up category. We further acknowledge support from COMP, l’ Association Québécoise des Physicien(ne)s Médicaux Cliniques, l’ Université Laval and CRCHU de Quebec-Université Laval.Louis Archambault, Luc Beaulieu and L. John Schreiner(for the organizing and scientific committee)List of International Scientific Committee, Local Organizing Committee, Detailed Program of the 12th International Conference on 3D and Advanced Dosimetry (IC3Ddose) June 2022, Château Frontenac, Quebec City, Contents, About, A Warm Thank You to our Sponsors, at a Glance, Day 1 – June, Day 2 – Day 3 – June 22nd, Day 4 –, a Warm Thank You to our Sponsors are available in this pdf.

Investigating the barriers and pathways to implementing satellite data into air quality monitoring, regulation and policy design in the United Kingdom

Satellite measurements relevant to air quality could help to inform policy and assist the monitoring and management of key air quality issues, both in the United Kingdom and internationally. This study investigates the barriers facing experts, across both scientific research and air quality management, in translating and incorporating satellite data into end user activities, and identifies pathways for the use of current and future satellite data by potential end users. Through an iterative Delphi-inspired two-round study, two groups of experts were issued parallel, near-identical surveys. These groups were Group A: current users of satellite data such as satellite product developers, researchers and air quality scientists, and Group B: potential end users of satellite data such as regulatory bodies, local authorities, government departments and independent air quality consultancies. This study confirmed that satellite data has had minimal penetration into Group B activities, and that a number of barriers currently prevent its adoption into regulation, namely: personnel-related issues such as lack of resources, technical skills and training, and technical limitations such as spatio-temporal resolution, uncertainties and data storage concerns. Despite this, there was a strong desire across both groups for collaboration and joint data exploration, but current progress has been minimal. Moving forward, we propose the creation of a network of experts, facilitated by a dedicated boundary organisation, designed to promote collaboration and knowledge exchange between these two groups, with the objective of transferring satellite data sets into end user work flows and to prepare end users to utilise future data sets.

Career connections: transnational expert networks and multilateral cybercrime negotiations

ABSTRACT The Council of Europe’s Budapest Convention on cybercrime (2001) remains a key landmark in multilateral governance on cybersecurity. Negotiated over two decades ago, this regional agreement gains renewed significance today through its enduring impact, its rarity as a binding multilateral agreement on cyber-issues, and its role in shaping the debate on free speech and the criminalization of online content. This article asks how the almost-total exclusion of content crimes from the Convention can be explained. Drawing on new documents published for the twentieth anniversary of the Convention, including detailed testimonials from its chair and principal drafters, the article highlights the importance of long-term career connections among technically-savvy and diplomatically experienced negotiators, creating a transnational policy network that overcame differences and power imbalances. Understanding how the Budapest Convention resolved disagreements helps contextualize current crises in multilateral cybersecurity and cybercrime governance and provides a path away from growing digital authoritarianism.