Search

BACKGROUND: Penetrating trauma to the head and neck (HN) has increased during the past decade in Sweden. This aim of this study was to characterize these injuries and to evaluate the outcome in patients treated at a tertiary trauma centre. METHODS: Swedish trauma registry (SweTrau) data was extracted on patients with HN injuries admitted to Karolinska University Hospital, Stockholm, between 2011 and 2019. Outcome information was extracted from hospital records, with primary endpoints focusing of physiological outcome measures and secondary endpoints on surgical and radiological outcomes. RESULTS: Of 1436 penetrating trauma patients, 329 patients with penetrating HN injuries were identified. 20% (n=66) suffered a gunshot wound (GSW), 73% (n=240) a stab wound (SW), and 7% (n=23) other trauma mechanisms (OTM). The median age for GSW, SW and OTM were 25, 33, and 21, respectively. Assault was primary intent, GSW (81.8%, n=54) and SW (65.8%, n=158). Patients suffering GSWs had severer injuries, worse admission GCS-M and higher intubation rate at the injury site. Most GSW patients underwent major surgery (59.1%) as an initial procedure and more likely had intracranial haemorrhage (ICH) (21.2%). The 30-day mortality: 45.5% (n=30) for GSWs, 5.4% (n=13) for SWs and 0% (n=0) for OTMs. There was a yearly increase in incidence and mortality for GSW and SW. CONCLUSION: Between 2011 and 2019, there was an increasing yearly trend of incidence and mortality from penetrating HN trauma in Stockholm, Sweden. GSW patients suffered more severe injuries, ICHs, and underwent more surgical interventions compared to SW and OTM.

Personalized Food Recommendation Systems (Food-RecSys) are recognized for their potential to enhance dietary practices and address pressing health issues. Despite their promise, current food recommendation methodologies, predominantly rule-based and classification-driven, encounter substantial barriers in achieving practical applicability. This challenge is largely attributed to two main factors: the poor understanding of food recommendation specific components as a whole and inadequate research on a holistic integration between all the key components which enables the final methodology to achieve effective results, and also, the limitations of conventional machine learning models, which falter in the face of a virtually limitless array of food categories and the inherent imbalance within datasets. In this context, the advent of Large Language Models (LLMs) presents an intriguing alternative, suggesting a path forward for recommendation systems. Nevertheless, existing approaches that employ LLMs generally adopt a generic Recommendation as Language Processing (RLP) strategy, which falls short of incorporating the nuanced components essential for successful Food-RecSys.To bridge this research gap, this thesis provides a comprehensive overview on an intensely complex big picture, identifies the key components and provides innovative exploratory models for each the multiplex key components required for personalized contextual food recommendation, additionally, articulates a novel feasibility study of a comprehensive integrated personal contextual Food-RecSys framework, underpinned by several pivotal innovations aimed at fortifying the effectiveness of personalized food recommendations. Key among these is the development of a multimedia food logging platform and the World Food Atlas, the latter of which facilitates geolocation-based food queries—a capability notably absent in current offerings. Moreover, this work pioneers the Food Recommendation as Language Processing (

A common type of cancer treatment is chemotherapy, which involves using anticancer chemicals to kill cancerous cells. Despite all advances in chemotherapy, this method still has undesired side effects on healthy cells and suffers from nonspecific targeting. Dendrimers are nanoparticles capable of delivering drugs, DNA, and RNA to tumor cells without harming normal cells, so they have been utilized extensively in targeted cancer treatment. A dendrimer is a highly branched polymeric macromolecule with excellent physicochemical properties, intramolecular cavities, and easy surface modification. These structures can be applied as anticancer drug carriers through the entrapment of drugs in their internal voids or binding to the functional groups on their surface. In addition, further surface modifications, such as targeting ligands, can increase selectivity, delivery, and biocompatibility. This review will briefly summarize dendrimer’s history and focus on dendrimer-based nanotherapeutics for cancer therapy. In a nutshell, functionalized dendrimers are presented here as potential drug and gene carriers for cancer treatment.

Background: The early management of polytrauma patients with traumatic spinal cord injury (tSCI) is a major challenge. Sparse data is available to provide optimal care in this scenario and worldwide variability in clinical practice has been documented in recent studies. Methods: A multidisciplinary consensus panel of physicians selected for their established clinical and scientific expertise in the acute management of tSCI polytrauma patients with different specializations was established. The World Society of Emergency Surgery (WSES) and the European Association of Neurosurgical Societies (EANS) endorsed the consensus, and a modified Delphi approach was adopted. Results: A total of 17 statements were proposed and discussed. A consensus was reached generating 17 recommendations (16 strong and 1 weak). Conclusions: This consensus provides practical recommendations to support a clinician’s decision making in the management of tSCI polytrauma patients.

Sexual and gender-based harassment is a work environment concern, especially in hierarchical and male-dominant organisations such as the police. This scoping review aims to examine the existing literature on such harassment against police employees in Europe to identify the available knowledge and highlight the gaps for future studies. Following the Arksey and O’Malley framework, 16 relevant articles were analysed. Most of the studies were focused on sexual harassment, and gender-based harassment was mostly absent or mixed with other discriminations. The analysis resulted in six main themes: existence of sexual and gender-based harassment; perpetrators; associated factors; consequences; individual response; and impact of organisational policies. The findings demonstrate that such harassment remains prevalent among police employees, emphasising the need for continuous and rigorous supervision, evaluation, and comprehensive revision of policies by police authorities. This study recommends further research to address the identified gaps in this field., Originally included in thesis in manuscript form.

In this theoretical investigation, we analyze light-induced nonlinear spin Hall currents in a gated single-layer 1T ′ -WTe2, flowing transversely to the incident laser polarization direction. Our study encompasses the exploration of the second and third-order rectified spin Hall currents using an effective low-energy Hamiltonian and employing the Kubo’s formalism. We extend our analysis to a wide frequency range spanning both transparent and absorbing regimes, investigating the influence of light frequency below and above the optical band gap. Additionally, we investigate the influence of an out-of-plane gate potential on the system, disrupting inversion symmetry and effectively manipulating both the strength and sign of nonlinear spin Hall responses. We predict a pronounced third-order spin Hall current relative to its second-order counterpart. The predicted nonlinear spin currents show strong anisotropic dependence on the laser polarization angle. The outcomes of our study contribute to a generalized framework for nonlinear response theory within the spin channel will impact the development of emerging field of opto-spintronic., QC 20240308

This study investigates 30 lone actors in Sweden with a register-based design using a group of male lone actors and two reference groups: same-sex siblings and other male violent extremists. We compare lone actors to the reference groups along social background, criminal background, and co-offending relations (1995 –2016), and mental health (1980–2016). Our results show that lone actors are primarily born in Sweden to two Swedish-born parents. They have a high degree of criminality and co-offending, indicating that they are not completely loners in their criminal behaviour. They have higher enrolment in secondary education than the reference groups, but lower enrolment in higher education than other male violent extremists. Additionally, they suffer considerably more from mental disorders compared to the reference groups. An analysis of criminality and in- and outpatient hospitalisation over the life course indicates that lone actors may have had problems in their transition into middle age.

We investigate the wetting properties of PDMS (Polydimethylsiloxane) pseudo-brush anchored on glass substrates. These PDMS pseudobrushes exhibit a significantly lower contact angle hysteresis compared to hydrophobic silanized substrates. The effect of different molar masses of the used PDMS on the wetting properties seems negligible. The surface roughness and thickness of the PDMS pseudo-brush are measured by atomic force microscopy and x-ray reflectivity. The outcome shows that these surfaces are extremely smooth (topologically and chemically), which explains the reduction in contact angle hysteresis. These special features make this kind of surfaces very useful for wetting experiments. Here, the dynamics of the four-phase contact point are studied on these surfaces. The four-phase contact point dynamics on PDMS pseudo-brushes deviate substantially from its dynamics on other substrates. These changes depend only a little on the molar mass of the used PDMS. In general, PDMS pseudo-brushes increase the traveling speed of four-phase contact point on the surface and change the associated power law of position vs time.

With the evolving role of Model Integrated Evidence (MIE) in generic drug development and regulatory applications, the need for improving Model Sharing, Acceptance, and Communication with the FDA is warranted. Model Master File (MMF) refers to a quantitative model or a modeling platform that has undergone sufficient model Verification & Validation to be recognized as sharable intellectual property that is acceptable for regulatory purposes. MMF provides a framework for regulatorily acceptable modeling practice, which can be used with confidence to support MIE by both the industry and the U.S. Food and Drug Administration (FDA). In 2022, the FDA and the Center for Research on Complex Generics (CRCG) hosted a virtual public workshop to discuss the best practices for utilizing modeling approaches to support generic product development. This report summarizes the presentations and panel discussions of the workshop symposium entitled "Model Sharing, Acceptance, and Communication with the FDA". The symposium and this report serve as a kick-off discussion for further utilities of MMF and best practices of utilizing MMF in drug development and regulatory submissions. The potential advantages of MMFs have garnered acknowledgment from model developers, industries, and the FDA throughout the workshop. To foster a unified comprehension of MMFs and establish best practices for their application, further dialogue and cooperation among stakeholders are imperative. To this end, a subsequent workshop is scheduled for May 2-3, 2024, in Rockville, Maryland, aiming to delve into the practical facets and best practices of MMFs pertinent to regulatory submissions involving modeling and simulation methodologies.

This letter studies the performance of reconfigurable intelligent surface (RIS)-aided communications for a fluid antenna system (FAS) enabled receiver. Specifically, a fixed single-antenna base station (BS) transmits information through a RIS to a mobile user which is equipped with a planar fluid antenna in the absence of a direct link. We first analyze the spatial correlation structures among the positions (or ports) in the planar FAS, and then derive the joint distribution of the equivalent channel gain at the user by exploiting the central limit theorem. Furthermore, we obtain compact analytical expressions for the outage probability (OP) and delay outage rate (DOR). Numerical results illustrate that using FAS with only one activated port into the RIS-aided communication network can greatly enhance the performance, when compared to traditional antenna systems (TAS). IEEE

Nanocellulose has emerged as a widely utilized building block in nanostructured materials due to its availability, sustainability, large surface area, and high stiffness and aspect ratio. The wet or dry elastoplastic properties of these materials are determined by the fibrils' stiffness, chemical properties, hemicellulose content, and the number of fibril contacts. However, the specific contributions and relative importance of each factor remain unclear. Therefore, this work was devoted to systematically comparing the material properties of gels, aerogels, and wet and dry sheets prepared from CNFs with different aspect ratios, chemical functionality, and hemicellulose content. The fibrils were prepared by chemical and mechanical processing of different pulps. By preserving the native structure as much as possible, higher aspect ratio fibrils can be obtained, which allows for the development of more mechanically robust materials. The results demonstrate that higher aspect ratios lead to more interconnected networks at a lower solids concentration, resulting in a more evenly distributed stress and longer-range stress transfer, yielding stiffer and more ductile materials. The most important finding was that the aspect ratio influences the network formation, resulting in different network topologies. The results were also compared to earlier published data and integrated into a theoretical beam-bending model for a complete elastoplastic description of the network properties, including the influence of fibril aspect ratio and chemical functionality. This information improves our understanding and description of nanofibril networks for which general models have been missing. It can be used to optimize nanofibril preparation and, hence, the resulting eco-friendly materials.

Research in the field of traumatic brain injury has until now heavily relied on the use of animal models to identify potential therapeutic approaches. However, a long series of failed clinical trials has brought many scientists to question the translational reliability of pre-clinical results obtained in animals. The search for an alternative to conventional models that better replicate human pathology in traumatic brain injury is thus of the utmost importance for the field. Recently, orthotopic xenotransplantation of human brain organoids into living animal models has been achieved. This review summarizes the existing literature on this new method, focusing on its potential applications in preclinical research, both in the context of cell replacement therapy and disease modelling. Given the obvious advantages of this approach to study human pathologies in an in vivo context, we here critically review its current limitations while considering its possible applications in traumatic brain injury research.

This study studies the challenges of effective communication and collaboration in remote design review meetings (DRMs) and explores the potential of Extended Reality (XR) technologies to address these challenges. The research focuses on identifying recurring communication issues and the preferences of companies within the context of remote DRMs. The study involves qualitative content analysis and industry workshops to uncover the current problems with conventional approaches and the aspirations of companies regarding improved collaboration in the DRM process. Drawing upon the insights gathered from both the workshop and design review observations, this paper highlights the features that are critical for collaborative software to handle online design reviews. XR technologies offer immersive and interactive experiences that can transform communication and collaboration in the context of DRMs. By identifying the specific challenges faced in remote DRMs and understanding the desires of companies, this study sets the stage for a more efficient and effective collaborative process. It emphasizes the adaptability of XR technologies to meet industry needs and integrate seamlessly into existing workflows. The study concludes by highlighting the potential for XR technologies to enhance collaboration in DRMs, making them a valuable tool for various industries., The authors would like to thank the Swedish innovation agency Vinnova for their funding of the PLENUM project, grant number: 2022-01704 as well as the partners in the projects that made this work possible.

Highly anisotropic cellulose nanofibrils can solidify liquid water, creating self-supporting structures by incorporating a tiny number of fibrils. These fibrillar hydrogels can contain as much as 99.99 wt% water. The structure and mechanical properties of fibrillar networks have so far not been completely understood, nor how they solidify the bulk water at such low particle concentrations. In this work, the mechanical properties of cellulose fibrillar hydrogels in the dilute regime from a wt% perspective have been studied, and an elastoplastic model describing the network structure and its mechanics is presented. A significant insight from this work is that the ability of the fibrils to solidify water is very dependent on particle stiffness and the number of contact points it can form in the network structure. The comparison between the experimental results and the theoretical model shows that the fibrillar networks in the dilute regime form via a non-stochastic process since the fibrils have the time and freedom to find contact points during network formation by translational and rotational diffusion. The formed, dilute fibrillar network deforms by sliding fibril contacts upon straining the network beyond its elastic limit. Our results also show that before macroscopic failure, the fibril contacts are restored once the load is released. The exceptional properties of this solidified water are exploited to host fluidic channels, allowing directed fluid transportation in water. Finally, the microfluidic channels formed in the hydrogels are tailored by the layer-by-layer technique to be interactive against external stimuli, a characteristic envisioned to be useful in biomedical applications.

Introduction: Increased levels of extracellular vesicles (EVs) are associated with haemostatic disturbances in various clinical settings. However, their role in COVID-19 patients is still not fully clear. In the present study we investigated EVs in plasma from patients with COVID-19 and neurological symptoms in relation to the activation of coagulation. Methods: Nineteen COVID-19 patients with neurological symptoms and twenty-three aged-matched healthy individuals were included. Global coagulation assays were performed and levels of EVs were determined by flow-cytometry in plasma and cerebrospinal fluid (CSF). Results: A procoagulant state characterized by significantly increased overall coagulation- (OCP) and overall haemostatic potential (OHP), diminished overall fibrinolytic potential (OFP) together with a denser fibrin structure was found in patients with COVID-19. Flow cytometry revealed elevated levels of plasma circulating EVs derived from neutrophils (MPO+) and platelets (CD61+), as well as EVs expressing phosphatidylserine (PS+) and complement component C5b-9 (TCC+) in patients with COVID-19 compared with controls. The concentrations of PS+, CD61+ and TCC+ EVs were positively correlated with OCP and OHP in COVID-19 patients. Moreover, we identified CD61+, MPO+ and endothelial cell-derived EVs, as well as EVs exposing PS and TCC in the CSF of patients suffering from neurological symptoms during COVID-19. Conclusion: The unique finding in this study was the presence of EVs in the CSF of COVID-19 patients with neurologic manifestations as well as higher expression of complement protein on circulating plasma EVs. EVs may indicate blood-brain barrier damage during SARS-COV-2 infection.

Background: The early management of polytrauma patients with traumatic spinal cord injury (tSCI) is a major challenge. Sparse data is available to provide optimal care in this scenario and worldwide variability in clinical practice has been documented in recent studies. Methods: A multidisciplinary consensus panel of physicians selected for their established clinical and scientific expertise in the acute management of tSCI polytrauma patients with different specializations was established. The World Society of Emergency Surgery (WSES) and the European Association of Neurosurgical Societies (EANS) endorsed the consensus, and a modified Delphi approach was adopted. Results: A total of 17 statements were proposed and discussed. A consensus was reached generating 17 recommendations (16 strong and 1 weak). Conclusions: This consensus provides practical recommendations to support a clinician's decision making in the management of tSCI polytrauma patients.

Cellulose-based aerogels are emerging bio-based materials for a range of applications in the quest toward a circular and carbon-neutral society. Owing to their lightweight nature, high porosity, high specific surface area, biocompatibility, and biodegradability, cellulose aerogels are suitable for packaging, insulation, wound care products, hygiene products, and water purification. However, their commercial use is hampered by complicated time- and energy-consuming fabrication processes. Hence, industrially relevant processes with upscaling opportunities need to be developed for cellulose-based aerogels to reach their full potential. This thesis explores different scalable and simple methods for preparing and designing highly porous aerogels with high wet integrity using cellulose-rich fibers and cellulose nanofibrils (CNFs). As wet integrity is crucial for specific applications and enables further functionalization of the aerogels using water-based chemistry, different methods were developed to achieve wet integrity without complicated crosslinking procedures. The effects of the raw materials and processing methods on the final material properties were also carefully studied to optimize the performance for the targeted applications. Moreover, the role of the network-forming ability of CNFs in the development of functional materials with structural integrity was explored by incorporating small amounts of CNFs in aerogel systems based on macroscopic cellulose-rich fibers and nanosized metal-organic frameworks. Finally, the potential of the different developed cellulose-based or cellulose-reinforced aerogels with high wet integrity was demonstrated in applications for which the aerogels’ structural integrity and physical and mechanical properties are highly advantageous, such as biomedical applications, gas storage and separation, flame retardancy, and hygiene products. As demonstrated in this thesis, these functional aerogel materials could be a bio-based alternativ, Cellulosabaserade aerogeler har på senare tid visat sig vara användbara biobaserade material för olika tillämpningar i strävan mot ett cirkulärt och kolneutralt samhälle. Materialens mycket låga densitet, höga porositet, höga specifika yta, biokompatibilitet och biologiska nedbrytbarhet innebär att cellulosaaerogeler är lämpliga för förpackningar, isolering, sårvårdsprodukter, hygienprodukter och vattenrening. Den kommersiella användningen har dock bromsats av komplicerade, tid- och energikrävande tillverkningsmetoder. Därmed måste industriellt relevanta processer med uppskalningsmöjligheter utvecklas för att cellulosabaserade aerogeler ska nå sin fulla potential. Denna avhandling utforskar olika skalbara och enkla metoder för att bereda och skräddarsy högporösa och våtstabila aerogeler från cellulosafibrer och cellulosananofibriller (CNFer). Eftersom våtstabilitet är avgörande för vissa tillämpningar och möjliggör ytterligare funktionalisering med vattenbaserad kemi, har ett omfattande arbete genomförts för att identifiera nya metoder för att skapa en god våtstabilitet utan att använda komplicerade tvärbindningsprocedurer. Ett stort fokus har även lagts på att klarlägga råvarans och bearbetningsmetodernas inverkan på de slutliga materialegenskaperna för att optimera prestandan för riktade tillämpningar. Dessutom har CNFers unika nätverksbildande egenskaper också utforskats i att skapa funktionella material med strukturell integritet från mycket små mängder CNFer i aerogelsystem baserade på makroskopiska cellulosarika fibrer och nanopartiklar av metallorganiska nätverk. Slutligen demonstrerades potentialen av de framställda cellulosabaserade och cellulosaförstärkta aerogelerna med utmärkt våtstyrka i tillämpningar där deras strukturella integritet, fysikaliska och mekaniska egenskaper kan användas på ett mycket fördelaktigt sätt, till exempel biomedicinska applikationer, gaslagring och -separering, flamskydd, och hygienprodukter. Mot bakgrund av dessa resultat är d, QC 20240522Embargo godkänt av skolchef Amelie Eriksson Karlström via e-post 2024-05-14

Generating a pure spin current using electrons, which have degrees of freedom beyond spin, such as electric charge and valley index, presents challenges. In response, we propose a mechanism based on intervalley exciton dynamics in arc-shaped strained transition metal dichalcogenides (TMDs) to achieve the exciton spin Hall effect in an electrically insulating regime, without the need for an external electric field. The interplay between strain gradients and strain-induced pseudomagnetic fields results in a net Lorentz force on long-lived intervalley excitons in WSe2, carrying nonzero spin angular momentum. This process generates an exciton-mediated pure spin Hall current, resulting in opposite-sign spin accumulations and local magnetization on the two sides of the single-layer arc-shaped TMD. We demonstrate that the magnetic field induced by spin accumulation, at approximately ∼mT, can be detected using techniques such as superconducting quantum interference magnetometry or spatially resolved magneto-optical Faraday and Kerr rotations., QC 20240618

While there is a substantial research literature on Western ‘foreign fighters’ – those young men and women from Europe, North America, Australia and elsewhere who traveled to Syria and Iraq, from around 2011–2017, to join jihadist groups engaged in combat – there is a dearth of comparative studies examining the backgrounds of these fighters. National variations in the levels of recruitment have been measured and samples examined to determine the demographics of these fighters, indicating some national variations in who went, how, and maybe why. More fulsome comparative data is needed, however, to detect and measure such differences to gain insight into the factors conditioning the radicalization of these foreign fighters. Calling on original and unique datasets, this study presents the results of a comparative analysis of Canadian and Swedish foreign fighters. In each case the findings are compared with other domestic jihadists as well to delineate if those drawn to fight in Syria and Iraq differ. Clear differences emerge in the basic demographics of these national samples, highlighting some empirical and interpretive issues in need of further analysis.

Objective:The majority of traumatic brain injuries (TBIs) are classified as mild and occur in young individuals. The course of recovery varies but can result in chronic or troubling outcomes. The impact of age on TBI outcomes in young adults before complete brain maturation is not well studied.Methods:In this study, we compared the effects of mild TBI on cognitive performance and self-reported TBI symptoms and posttraumatic stress disorder (PTSD) in 903 soldiers in 3 different age groups: 24 years or younger, 25 to 27 years, and 28 to 40 years. The soldiers had returned from war zones in Iraq and were screened for TBI within a few days of return. Cognitive performance was measured with the Automated Neuropsychological Assessment Metrics of Military TBI Version 4 (ANAM4). Symptoms associated with mild TBI were self-reported on the Neurobehavioral Symptom Inventory, and the PTSD Checklist-Civilian Version (PCL-C).Results:Soldiers with TBI in every age group had significantly higher prevalence of most symptoms than those with no TBI. Soldiers with TBI also reported more chronic pain sites, regardless of age. Soldiers aged 28 to 40 years with TBI had the lowest cognitive performance scores (ANAM) across several subtests, both unadjusted and adjusted. The Global Deficit Score was significantly higher for soldiers aged 28 to 40 years and 25 to 27 years with TBI than for soldiers younger than 24 years with no TBI. After adjusting for PTSD symptoms, education, and number of lifetime TBIs, the overall test battery mean for soldiers aged 28 to 40 years with TBI was significantly lower than for soldiers younger than 24 years with no TBI.Conclusion:Soldiers with mild TBI in the younger age group show more symptoms associated to frontal lobe function while soldiers in the older group suffer more cognitive impairment. This may warrant further study as it may indicate a propensity to later cognitive decline among soldiers who were older at the time of injury.

Police officers encounter challenging and stressful situations at work, which negatively impact their health, job performance, and aspects of their lives. Regarding the gendered structure and competitive work environment of police organizations, women are under extra pressure from the pervasive gendered norms within the police organization. The aim of this dissertation is to study sexual and gender-based harassment and work-related stress as important work environment elements from a gender perspective, and to explore if and how these elements can affect police officers’ job satisfaction in the context of Swedish police work. This thesis is based on three studies: a scoping review, a quantitative study, and a qualitative study, resulting in four articles. In the scoping review study on sexual and gender-based harassment in police work in the European context 16 articles on sexual and gender-based harassment against police employees were studied. Thematic analysis was applied to obtain the main patterns across and within the included studies. In the qualitative study, data collection included one group interview and 12 individual interviews with male and female police officers. Thematic and content analysis were conducted to analyze the interviews. During the quantitative study, data were collected from 152 male and female police officers applying a set of questionnaires including sociodemographic questions, the Police Stress Identification Questionnaire (PSIQ), Sexual and gender-based harassment questions, and Job Descriptive Index (JDI). The scoping review showed that most of the European studies were focused on sexual harassment, while gender-based harassment was often overlooked or mixed with other types of harassment and discrimination. Six main themes were found in the studies; the existence of sexual and gender-based harassment, perpetrators, associated factors, consequences, individual response, and impact of organizational policies. In the mixed methods a

Purpose: This paper aims to investigate how congruent keywords are used in information security policies (ISPs) to pinpoint and guide clear actionable advice and suggest a metric for measuring the quality of keyword use in ISPs. Design/methodology/approach: A qualitative content analysis of 15 ISPs from public agencies in Sweden was conducted with the aid of Orange Data Mining Software. The authors extracted 890 sentences from these ISPs that included one or more of the analyzed keywords. These sentences were analyzed using the new metric - keyword loss of specificity - to assess to what extent the selected keywords were used for pinpointing and guiding actionable advice. Thus, the authors classified the extracted sentences as either actionable advice or other information, depending on the type of information conveyed. Findings: The results show a significant keyword loss of specificity in relation to pieces of actionable advice in ISPs provided by Swedish public agencies. About two-thirds of the sentences in which the analyzed keywords were used focused on information other than actionable advice. Such dual use of keywords reduces the possibility of pinpointing and communicating clear, actionable advice. Research limitations/implications: The suggested metric provides a means to assess the quality of how keywords are used in ISPs for different purposes. The results show that more research is needed on how keywords are used in ISPs. Practical implications: The authors recommended that ISP designers exercise caution when using keywords in ISPs and maintain coherency in their use of keywords. ISP designers can use the suggested metrics to assess the quality of actionable advice in their ISPs. Originality/value: The keyword loss of specificity metric adds to the few quantitative metrics available to assess ISP quality. To the best of the authors' knowledge, applying this metric is a first attempt to measure the quality of actionable advice in ISPs.

In recent years, the study of terrorism and violent extremism has increasingly benefited from the use of register data such as population data and judicial data. The use of this type of data brings with it many advantages, such as the ability to study complete populations, to combine different sources of data, and, often, it enables researchers to take on a longitudinal and comparative approach. However, the use of register data also brings with it some significant challenges, in terms of acquiring the data (practical challenges) and in terms of biases in the data (content/ethical challenges). In this paper we discuss the usage of register data in terrorism research, with a focus on the Dutch and Swedish context. Whereas both countries are to some extent similar in that they both have a rich data-infrastructure, differences also exist regarding the type of data, agencies responsible for collecting and distributing the data, and ethical approaches. In the chapter, we will go into the strengths and weaknesses of using register data in research on terrorism and violent extremism. Moreover, we will elaborate on how the use of register data can be improved, what type of research questions can and cannot be answered based on register data, and on potential opportunities of integrating register data into mixed-methods designs.

It is fascinating to predict the mass and width of the ordinary and exotic mesons solely based on their quark content and quantum numbers. Such prediction goes beyond conventional methodologies traditionally employed in hadron physics for calculating or estimating these quantities. The relation between the quantum numbers and the properties of the mesons, such as the mass and width, is complicated in the world of particle physics. However, the deep neural network (DNN) as a subfield of machine learning techniques provides a solution to this problem. By analyzing large datasets, deep learning algorithms can automatically identify complex patterns among the particles' quantum numbers, and their mass and width, that would otherwise require complex calculations. In this study, we present two approaches using the DNNs to estimate the mass of some ordinary and exotic mesons. Also for the first time, the DNNs are trained to predict the width of ordinary and exotic mesons, whose widths have not been experimentally known. Our predictions obtained through the DNNs, will be useful for future experimental searches., Comment: 27 Pages, 2 Figures and 16 Tables

Diffusion models have been widely used for conditional data cross-modal generation tasks such as text-to-image and text-to-video. However, state-of-the-art models still fail to align the generated visual concepts with high-level semantics in a language such as object count, spatial relationship, etc. We approach this problem from a multimodal data fusion perspective and investigate how different fusion strategies can affect vision-language alignment. We discover that compared to the widely used early fusion of conditioning text in a pretrained image feature space, a specially designed intermediate fusion can: (i) boost text-to-image alignment with improved generation quality and (ii) improve training and inference efficiency by reducing low-rank text-to-image attention calculations. We perform experiments using a text-to-image generation task on the MS-COCO dataset. We compare our intermediate fusion mechanism with the classic early fusion mechanism on two common conditioning methods on a U-shaped ViT backbone. Our intermediate fusion model achieves a higher CLIP Score and lower FID, with 20% reduced FLOPs, and 50% increased training speed compared to a strong U-ViT baseline with an early fusion.

This letter investigates the performance of content caching in a heterogeneous cellular network (HetNet) consisting of fluid antenna system (FAS)-equipped mobile users (MUs) and millimeter-wave (mm-wave) single-antenna small base stations (SBSs), distributed according to the independent homogeneous Poisson point processes (HPPP). In particular, it is assumed that the most popular contents are cached in the SBSs to serve the FAS-equipped MUs requests. To assess the system performance, we derive compact expressions for the successful content delivery probability (SCDP) and the content delivery delay (CDD) using the Gauss-Laguerre quadrature technique. Our numerical results show that the performance of cache-enabled mm-wave HetNets can be greatly improved, when the FAS is utilized at the MUs instead of traditional fixed-antenna system deployment.

Cross-modal feature extraction and integration have led to steady performance improvements in few-shot learning tasks due to generating richer features. However, existing multi-modal object detection (MM-OD) methods degrade when facing significant domain-shift and are sample insufficient. We hypothesize that rich text information could more effectively help the model to build a knowledge relationship between the vision instance and its language description and can help mitigate domain shift. Specifically, we study the Cross-Domain few-shot generalization of MM-OD (CDMM-FSOD) and propose a meta-learning based multi-modal few-shot object detection method that utilizes rich text semantic information as an auxiliary modality to achieve domain adaptation in the context of FSOD. Our proposed network contains (i) a multi-modal feature aggregation module that aligns the vision and language support feature embeddings and (ii) a rich text semantic rectify module that utilizes bidirectional text feature generation to reinforce multi-modal feature alignment and thus to enhance the model's language understanding capability. We evaluate our model on common standard cross-domain object detection datasets and demonstrate that our approach considerably outperforms existing FSOD methods.

Training a deep neural network using gradient-based methods necessitates the calculation of gradients at each level. However, using backpropagation or reverse mode differentiation, to calculate the gradients necessities significant memory consumption, rendering backpropagation an inefficient method for computing gradients. This paper focuses on analyzing the performance of the well-known Frank-Wolfe algorithm, a.k.a. conditional gradient algorithm by having access to the forward mode of automatic differentiation to compute gradients. We provide in-depth technical details that show the proposed Algorithm does converge to the optimal solution with a sub-linear rate of convergence by having access to the noisy estimate of the true gradient obtained in the forward mode of automated differentiation, referred to as the Projected Forward Gradient. In contrast, the standard Frank-Wolfe algorithm, when provided with access to the Projected Forward Gradient, fails to converge to the optimal solution. We demonstrate the convergence attributes of our proposed algorithms using a numerical example.

Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) is a gene editing technology that has revolutionized the fields of biology and medicine. However, one of the challenges of using CRISPR is predicting the on-target efficacy and off-target sensitivity of single-guide RNAs (sgRNAs). This is because most existing methods are trained on separate datasets with different genes and cells, which limits their generalizability. In this paper, we propose a novel ensemble learning method for sgRNA design that is accurate and generalizable. Our method combines the predictions of multiple machine learning models to produce a single, more robust prediction. This approach allows us to learn from a wider range of data, which improves the generalizability of our model. We evaluated our method on a benchmark dataset of sgRNA designs and found that it outperformed existing methods in terms of both accuracy and generalizability. Our results suggest that our method can be used to design sgRNAs with high sensitivity and specificity, even for new genes or cells. This could have important implications for the clinical use of CRISPR, as it would allow researchers to design more effective and safer treatments for a variety of diseases.

Few-shot learning (FSL) is a challenging machine learning problem due to a scarcity of labeled data. The ability to generalize effectively on both novel and training tasks is a significant barrier to FSL. This paper proposes a novel solution that can generalize to both training and novel tasks while also utilizing unlabeled samples. The method refines the embedding model before updating the outer loop using unsupervised techniques as ``meta-tasks''. The experimental results show that our proposed method performs well on novel and training tasks, with faster and better convergence, lower generalization, and standard deviation error, indicating its potential for practical applications in FSL. The experimental results show that the proposed method outperforms prototypical networks by 3.9%.

Small-diameter TBMs are increasingly deployed for utility tunnelling projects, but they face diverse geological and technical challenges. With the aid of a recently compiled database, new approaches for performance analysis and prediction for small-diameter TBMs are presented. Furthermore, a comprehensive geotechnical testing on microwave pre-conditioned rock samples, resulting in strength reduction ratios of over 50%, was performed, which could significantly boost the advance rates and profitability of future mechanized hard rock tunnel boring and mining equipment., TBMs mit kleinen Durchmessern werden zunehmend für Projekte mit kleinen Durchmessern eingesetzt, stehen aber vor diversen geologischen und technischen Herausforderungen. Mit Hilfe einer neu zusammengestellten Datenbank werden neue Ansätze für die Leistungsanalyse und -vorhersage vorgestellt. Weiterhin wurden umfangreiche geotechnische Tests an mikrowellenbestrahlten Gesteinen durchgeführt. Die erzielte Festigkeitsreduzierung von über 50 % könnte die Vortriebsraten und die Rentabilität künftiger Tunnelbohr- und Bergbaumaschinen erheblich steigern.

In this article, we explore the communication performance of vehicle-to-vehicle (V2V) networks assisted by a reconfigurable intelligent surface (RIS) and a simultaneous transmitting and reflecting intelligent omni-surface (STAR-IOS) under non-orthogonal multiple access (NOMA) and orthogonal multiple access (OMA) schemes. In particular, we consider that the RIS is close to the transmitter vehicle while the STAR-IOS is near the receiver vehicles. In addition, it is assumed that the STAR-IOS exploits the energy-splitting (ES) protocol for communication and the fading channels between the RIS and STAR-IOS follow composite Fisher-Snedecor F distribution. Under such assumptions, we first use the central limit theorem (CLT) to derive the PDF and the CDF of equivalent channels at receiver vehicles, and then, we derive the closed-form expression of outage probability (OP) under NOMA/OMA scenarios. Additionally, by exploiting Jensen's inequality, we propose an upper bound of the ergodic capacity (EC) and subsequently develop an analytical expression of the energy efficiency (EE) for both NOMA and OMA cases. Further, our analytical results, which are double-checked with the Monte-Carlo simulation, reveal that applying RIS/STAR-RIS in V2V communications significantly enhances the communication performance of intelligent transportation systems (ITS). Besides, the results indicate that considering the NOMA scheme provides better performance in terms of the OP, EC, and EE as compared with the OMA case for the considered V2V communication system.

State-of-the-art rule-based and classification-based food recommendation systems face significant challenges in becoming practical and useful. This difficulty arises primarily because most machine learning models struggle with problems characterized by an almost infinite number of classes and a limited number of samples within an unbalanced dataset. Conversely, the emergence of Large Language Models (LLMs) as recommendation engines offers a promising avenue. However, a general-purpose Recommendation as Language Processing (RLP) approach lacks the critical components necessary for effective food recommendations. To address this gap, we introduce Food Recommendation as Language Processing (F-RLP), a novel framework that offers a food-specific, tailored infrastructure. F-RLP leverages the capabilities of LLMs to maximize their potential, thereby paving the way for more accurate, personalized food recommendations.

Computing systems have been evolving to be more pervasive, heterogeneous, and dynamic. An increasing number of emerging domains now rely on diverse edge to cloud continuum where the execution of applications often spans various tiers of systems with significantly heterogeneous computational capabilities. Resources in each tier are often handled in isolation due to scalability and privacy concerns. However, better overall resource utilization could be achieved if different tiers of systems had the means to communicate their computational capabilities. In this paper, we propose H-EYE, a universal approach to holistically capture diverse computational characteristics of edge-cloud systems with arbitrary topologies and to manage the assignment of tasks to the computational resources with the whole continuum in the scope. Our proposed work introduces two significant innovations: (1) We present a multi-layer, graph-based hardware (HW) representation and a modular performance modeling interface that could capture interactions and inference between different computing and communication resources in the system at desired level of detail. (2) We introduce a novel orchestrator mechanism that leverages the graph-based HW representation to hierarchically locate target devices that a given set of tasks could be mapped to. Orchestrator provides isolation for various device groups and allows hierarchical abstraction to scalably find mappings that satisfy system deadlines. The orchestrator internally relies on a novel traverser that takes shared resource slowdown into account. We demonstrate the utility and flexibility of H-EYE on edge-server systems that are deployed on the field in two different disciplines, improving up to 47% latency over baselines with less than 2% scheduling overhead

This paper investigates the performance of physical layer security (PLS) in fluid antenna-aided communication systems under arbitrary correlated fading channels. In particular, it is considered that a single fixed-antenna transmitter aims to send confidential information to a legitimate receiver equipped with a planar fluid antenna system (FAS), while an eavesdropper, also taking advantage of a planar FAS, attempts to decode the desired message. For this scenario, we first present analytical expressions of the equivalent channel distributions at the legitimate user and eavesdropper by using copula, so that the obtained analytical results are valid for any arbitrarily correlated fading distributions. Then, with the help of Gauss-Laguerre quadrature, we derive compact analytical expressions for the average secrecy capacity (ASC), the secrecy outage probability (SOP), and the secrecy energy efficiency (SEE) for the FAS wiretap channel. Moreover, for exemplary purposes, we also obtain the compact expression of ASC, SOP, and SEE by utilizing the Gaussian copula under correlated Rayleigh fading channels as a special case. Eventually, numerical results indicate that applying the fluid antenna with only one active port to PLS can guarantee more secure and reliable transmission, when compared to traditional antenna systems (TAS) exploiting maximal ratio combining (MRC).

Timing vulnerabilities in processors have emerged as a potent threat. As processors are the foundation of any computing system, identifying these flaws is imperative. Recently fuzzing techniques, traditionally used for detecting software vulnerabilities, have shown promising results for uncovering vulnerabilities in large-scale hardware designs, such as processors. Researchers have adapted black-box or grey-box fuzzing to detect timing vulnerabilities in processors. However, they cannot identify the locations or root causes of these timing vulnerabilities, nor do they provide coverage feedback to enable the designer's confidence in the processor's security. To address the deficiencies of the existing fuzzers, we present WhisperFuzz--the first white-box fuzzer with static analysis--aiming to detect and locate timing vulnerabilities in processors and evaluate the coverage of microarchitectural timing behaviors. WhisperFuzz uses the fundamental nature of processors' timing behaviors, microarchitectural state transitions, to localize timing vulnerabilities. WhisperFuzz automatically extracts microarchitectural state transitions from a processor design at the register-transfer level (RTL) and instruments the design to monitor the state transitions as coverage. Moreover, WhisperFuzz measures the time a design-under-test (DUT) takes to process tests, identifying any minor, abnormal variations that may hint at a timing vulnerability. WhisperFuzz detects 12 new timing vulnerabilities across advanced open-sourced RISC-V processors: BOOM, Rocket Core, and CVA6. Eight of these violate the zero latency requirements of the Zkt extension and are considered serious security vulnerabilities. Moreover, WhisperFuzz also pinpoints the locations of the new and the existing vulnerabilities., Comment: Accepted to USENIX Sec'24

As humans advance toward a higher level of artificial intelligence, it is always at the cost of escalating computational resource consumption, which requires developing novel solutions to meet the exponential growth of AI computing demand. Neuromorphic hardware takes inspiration from how the brain processes information and promises energy-efficient computing of AI workloads. Despite its potential, neuromorphic hardware has not found its way into commercial AI data centers. In this article, we try to analyze the underlying reasons for this and derive requirements and guidelines to promote neuromorphic systems for efficient and sustainable cloud computing: We first review currently available neuromorphic hardware systems and collect examples where neuromorphic solutions excel conventional AI processing on CPUs and GPUs. Next, we identify applications, models and algorithms which are commonly deployed in AI data centers as further directions for neuromorphic algorithms research. Last, we derive requirements and best practices for the hardware and software integration of neuromorphic systems into data centers. With this article, we hope to increase awareness of the challenges of integrating neuromorphic hardware into data centers and to guide the community to enable sustainable and energy-efficient AI at scale., Comment: 11 pages, 2 figures, submitted to NICE 2024

We introduce an algorithm for tackling the problem of unsupervised domain adaptation (UDA) in continual learning (CL) scenarios. The primary objective is to maintain model generalization under domain shift when new domains arrive continually through updating a base model when only unlabeled data is accessible in subsequent tasks. While there are many existing UDA algorithms, they typically require access to both the source and target domain datasets simultaneously. Conversely, existing CL approaches can handle tasks that all have labeled data. Our solution is based on stabilizing the learned internal distribution to enhances the model generalization on new domains. The internal distribution is modeled by network responses in hidden layer. We model this internal distribution using a Gaussian mixture model (GMM ) and update the model by matching the internally learned distribution of new domains to the estimated GMM. Additionally, we leverage experience replay to overcome the problem of catastrophic forgetting, where the model loses previously acquired knowledge when learning new tasks. We offer theoretical analysis to explain why our algorithm would work. We also offer extensive comparative and analytic experiments to demonstrate that our method is effective. We perform experiments on four benchmark datasets to demonstrate that our approach is effective.

Transformer neural networks are increasingly replacing prior architectures in a wide range of applications in different data modalities. The increasing size and computational demands of fine-tuning large pre-trained transformer neural networks pose significant challenges for the widespread adoption of these models for applications that demand on-edge computing. To tackle this challenge, continual learning (CL) emerges as a solution by facilitating the transfer of knowledge across tasks that arrive sequentially for an autonomously learning agent. However, current CL methods mainly focus on learning tasks that are exclusively vision-based or language-based. We propose a transformer-based CL framework focusing on learning tasks that involve both vision and language, known as Vision-and-Language (VaL) tasks. Due to the success of transformers in other modalities, our architecture has the potential to be used in multimodal learning settings. In our framework, we benefit from introducing extra parameters to a base transformer to specialize the network for each task. As a result, we enable dynamic model expansion to learn several tasks in a sequence. We also use knowledge distillation to benefit from relevant past experiences to learn the current task more efficiently. Our proposed method, Task Attentive Multimodal Continual Learning (TAM-CL), allows for the exchange of information between tasks while mitigating the problem of catastrophic forgetting. Notably, our approach is scalable, incurring minimal memory and time overhead. TAM-CL achieves state-of-the-art (SOTA) performance on challenging multimodal tasks

This paper studies the performance of backscatter communications (BC) over emerging fluid antenna (FA) technology. In particular, a single-antenna source sends information to a FA reader through the wireless forward (i.e., source-to-tag) and backscatter (tag-to-reader) channels. For the considered BC, we first derive the cumulative distribution function (CDF) of the equivalent channel at the FA receiver, and then we obtain closed-form expressions of the outage probability (OP) and delay outage rate (DOR) under a correlated Rayleigh distribution. Moreover, in order to gain more insights into the system performance, we present analytical expressions of the OP and DOR at the high SNR regime. Numerical results indicate that considering the FA at the reader can significantly improve the performance of BC in terms of the OP and DOR compared with a single-antenna reader.

A major technique for tackling unsupervised domain adaptation involves mapping data points from both the source and target domains into a shared embedding space. The mapping encoder to the embedding space is trained such that the embedding space becomes domain agnostic, allowing a classifier trained on the source domain to generalize well on the target domain. To further enhance the performance of unsupervised domain adaptation (UDA), we develop an additional technique which makes the internal distribution of the source domain more compact, thereby improving the model's ability to generalize in the target domain.We demonstrate that by increasing the margins between data representations for different classes in the embedding space, we can improve the model performance for UDA. To make the internal representation more compact, we estimate the internally learned multi-modal distribution of the source domain as Gaussian mixture model (GMM). Utilizing the estimated GMM, we enhance the separation between different classes in the source domain, thereby mitigating the effects of domain shift. We offer theoretical analysis to support outperofrmance of our method. To evaluate the effectiveness of our approach, we conduct experiments on widely used UDA benchmark UDA datasets. The results indicate that our method enhances model generalizability and outperforms existing techniques.

This paper investigates the performance of physical layer security (PLS) in a vehicle-to-vehicle (V2V) communication system, where a transmitter vehicle exploits a dual reconfigurable intelligent surface (RIS) to send confidential information to legitimate receiver vehicles under the non-orthogonal multiple access (NOMA) scheme in the presence of an eavesdropper vehicle. In particular, it is assumed that an RIS is near the transmitter vehicle and another RIS is close to the receiver vehicles to provide a wider smart radio environment. Besides, we suppose that the channels between two RISs suffer from the Fisher-Snedecor F fading model. Under this scenario, we first provide the marginal distributions of equivalent channels at the legitimate receiver vehicles by exploiting the central limit theorem (CLT). Then, in order to evaluate the PLS performance of the considered secure communication system, we derive analytical expressions of the average secrecy capacity (ASC), secrecy outage probability (SOP), and secrecy energy efficiency (SEE) by using the Gauss-Laguerre quadrature and the Gaussian quadrature techniques. Moreover, to gain more insights into the secrecy performance, the asymptotic expression of the ASC is obtained. The numerical results indicate that incorporating the dual RIS in the secure V2V communication under the NOMA scheme can significantly provide ultra-reliable transmission and guarantee more secure communication for intelligent transportation systems (ITS).

Large language models demonstrate remarkable proficiency in various linguistic tasks and have extensive knowledge across various domains. Although they perform best in English, their ability in other languages is notable too. In contrast, open-source models, such as LLaMa, are primarily trained on English datasets, resulting in poor performance in non-English languages. In this paper, we introduce PersianMind, an open-source bilingual large language model which demonstrates comparable performance to closed-source GPT-3.5-turbo in the Persian language. By expanding LLaMa2's vocabulary with 10,000 Persian tokens and training it on a dataset comprising nearly 2 billion Persian tokens, we show that our approach preserves the model's English knowledge and employs transfer learning to excel at transferring task knowledge from one language to another.