Mid Sweden University
Not a member yet
15329 research outputs found
Sort by
Laboratory and synchrotron validation of µ-XRF for sulfur mapping in CTMP paper samples
No full textThe transition toward renewable, fiber-based packaging requires an improved understandingof chemical modifications in high-yield pulps such as chemithermomechanical pulp (CTMP). Sulfonationuniformity is essential for the energy-efficient production of high-strength CTMP pulp. However,laboratory methods only measure total sulfur and cannot illustrate its distribution at the fiber level,which can be visualized using μ-XRF. In this work, we present a laboratory μ-XRF system developedat Mid Sweden University and assess its capability to detect light elements in CTMP paper handsheets.A 32 × 32 point grid scan (1.6 × 1.6mm2 field of view, 50 μm step, 300 s/point) successfully resolvedsulfur Kα (2.31 keV) and calcium Kα (3.69 keV) fluorescence without helium flushing. Comparativemeasurements at the Elettra synchrotron confirmed consistency of sulfur peak position and spatialdistribution, with higher spectral resolution and signal-to-noise ratio. Histogram analysis usingWassersteindistance metrics demonstrated close agreement between datasets despite differing acquisitionconditions. These results demonstrate that laboratory XRF can reproducibly detect and map sulfur inCTMP fibers under ambient conditions, providing a practical tool to complement synchrotron studiesand supporting the development of energy-efficient, fiber-based packaging materials.Open access: https://iopscience.iop.org/article/10.1088/1748-0221/20/11/C11002</p
Interpretable Intrusion Detection for Automotive Ethernet Networks : A Logical Neural Network Approach for Symbolic Reasoning
No full textThe rapid integration of Automotive Ethernet as the backbone for modern vehicle communication systems has introduced unprecedented advantages in speed and scalability for Advanced Driver Assistance Systems (ADAS) and autonomous driving technologies in in-vehicle Networks (IVNs). However, this technological advancement has also increased the cybersecurity risks, as the Automotive Ethernet’s uniquearchitecture demands specialized security solutions. Intrusion Detection Systems (IDS) are critical to mitigating these threats, yet the demands of the automotive domain including real-time constraints, reliability, and adherence to safety standards set them apart from IDS inother domains. Traditional IDS approaches often rely on deep learningmodels that, while accurate, lack interpretability, posing challengesfor safety-critical automotive applications where explainability is essential. However, eXplainable artificial intelligence (XAI) provides a pathway to achieve transparency, enabling stakeholders to understand, validate, and trust system outputs. But this thesis studies a novel Neuro-Symbolic approach to intrusion detection by integrating Logical Neural Networks (LNN) ’a form of explainable AI developed by IBM’ into the IDS framework for Automotive Ethernet, by combining the pattern recognition strengths of neural networks with the logical reasoning capabilities of symbolic AI. The proposed framework aims to address the dual challenges of detection performance and explainability. The results demonstrate that LNN achieve competitive detection accuracy while significantly enhancing interpretability, offering a promising pathway toward trustworthy and transparent IDS in modern vehicular networks
How am I Supposed to Read this Whale? : Allegorical and Counter-Allegorical Representational Strategies in Cetopoetic Narratives
No full textAnimal stories are ubiquitous in oral and literary traditions the world over, but they have only recently become an object of sustained scholarly inquiry. Cultural and literary animal studies (CLAS), a vibrant subfield of the multi- and interdisciplinary research field of animal studies challenges many of the unspoken assumptions about nonhuman and human animals which also underlie humanism as well as the self-understanding of disciplines like literary studies. As CLAS reconceptualizes nonhuman animals as a constitutive part of human cultures, this paradigm shift prompts a re-thinking of how we conceptualize animals in texts and their relationship to living animals as well. Anchored in this research area, the compilation thesis "How Am I Supposed to Read This Whale? Allegorical and Counter-Allegorical Representational Strategies in Cetopoetic Texts" comprises four published papers on selected anglophone narratives that prominently feature a specific group of animals, namely cetaceans (whales, dolphins, and porpoises): The science fiction film Star Trek IV: The Voyage Home, the activist documentaries The Cove and Blackfish, Herman Melville's novel Moby-Dick, and Witi Ihimaera's novel The Whale Rider. In each paper, I examine how the representations of cetaceans in the text the facilitate specific cultural work. Thinking with and about whales offers ways to process anxieties about environmental risks on a planetary scale; consider the possibility of nonhuman personhood and its ethical implications; reflect on questions of moral and legal standing; and to be immersed in Maori cosmology as it reaffirms itself in the face of an uncertain future. Collectively, the papers suggest that expanding the notion of animal representation from a rather straightforward translation gesture to a more comprehensive zoopoetics-and in the selected whale stories, specifically a cetopoetics-illuminates the specific ultural work facilitated by these texts in greater nuance.Djurberättelser finns i muntliga och litterära traditioner över hela världen, men det är först på senare tid som de har blivit föremål för kontinuerligt vetenskaplig forskning. Cultural and Literary Animal Studies (CLAS), ett växande delområde inom det mång- och tvärvetenskapliga forskningsfältet Animal Studies (djurstudier), ifrågasätter många av de outtalade antaganden om icke-mänskliga och mänskliga djur som ligger till grund för såväl humanismen såsom litteraturvetenskap. Eftersom CLAS utgår ifrån tanken att icke-mänskliga djur är en konstituerande del av mänskliga kulturer, leder detta paradigmskifte till ett omprövande av hur vi förstår djur i texter och deras relation till levande djur. Avhandlingen “How am I Supposed to Read this Whale? Allegorical and Counter-Allegorical Representational Strategies in Cetopoetic Texts” är förankrad i detta forskningsområde och består av fyra publicerade artiklar om utvalda engelskspråkiga berättelser som behandlar cetaceans (valar, delfiner och tumlare): science fiction-filmen Star Trek IV: The Voyage Home, aktivistdokumentärerna The Cove och Blackfish, Herman Melvilles roman Moby-Dick och Witi Ihimaeras roman The Whale Rider. Varje artikel undersöker hur representationer av valar i texten möjliggör kulturellt arbete. Att tänka med och om valar erbjuder sätt att bearbeta oron över världsomfattande miljörisker; att reflektera över möjligheten att identifiera icke-mänskliga djur som personer och de etiska konsekvenser som detta betraktelsesätt kan få för djurs moraliska och juridiska ställning; samt att sätta sig in i māorisk kosmologi som är hotad men bekräftas av berättelser. Sammantaget visar artiklarna att förståelsen för begreppet ‘djurrepresentation’ behöver utvidgas från att beskrivna ett tolkningsförfarande som närmar sig djur i text främst, eller enbart, som metafor för mänskliga problem, till en mer omfattande zoopoetik (eller snarare en cetopoetik sett i relation till avhandlingsmaterialet) som på ett nyanserat sätt belyser det specifika kulturella arbete som zoopoetiska eller cetopoetiska element möjliggör
MultiEpilepsyNet : An EEG and MRI data based multimodal seizure detection model using hybrid deep learning model
No full textEpilepsy is a critical neurological disorder that requires accurate and privacy-preserving diagnostic solutions to enable early detection and effective management. However, existing approaches face key challenges, including reliance on centralized data, poor generalizability across modalities, suboptimal feature extraction, and vulnerability to noise. To address these issues, we propose MultiEpilepsyNet, a novel multimodal seizure detection framework that integrates federated learning with hybrid deep learning models. At its core, the system introduces SeizureFed-Net, a federated architecture that enables collaborative learning from EEG and MRI data while safeguarding patient privacy. For detection, we design SeizureShieldNet, a hybrid model that fuses the temporal learning capabilities of BBIDNet (Boosted BiLSTM Intrusion Detector Network) with the adaptive decision-making of FD-TMS (Fuzzy-DQN Threat Mitigation System) under uncertainty. To further enhance model efficiency, a Jackal-Wolf Hybrid Optimizer (JWHO)—a novel combination of Golden Jackal Optimization and Grey Wolf Optimizer—is employed for optimal feature subset selection. On the imaging side, MRI preprocessing is improved through EpiSkullNet+ +, a modified 3D UNet+ + architecture tailored for precise brain segmentation. Extensive experiments on two benchmark datasets demonstrate the superiority of our approach, achieving 99.36 % accuracy on the CHB-MIT EEG dataset and 99.38 % accuracy on an epilepsy MRI dataset. Beyond accuracy, MultiEpilepsyNet demonstrates improved robustness to missing modalities, reduced training overhead via federated aggregation, and enhanced privacy preservation compared to centralized deep learning models, thereby addressing critical barriers in practical clinical deployment. These outcomes highlight the effectiveness, scalability, and real-world clinical potential of the proposed framework for epilepsy diagnosis and management.
Nonlinear Evolution Equations of the Soliton Type : Old and New Results
No full textAn overview on the study of nonlinear evolution equations of soliton type is provided. In addition, 5th-order nonlinear evolution equations are shown to be connected to the Caudrey–Dodd–Gibbon–Sawada–Kotera (CDGSK) equation via Bäcklund transformations. The links are depicted in a wide net of links which we term a Bäcklund Chart. The links obtained previously by Rogers and Carillo and by Carillo and Fuchssteiner are revisited, and new results are obtained. A 5th-order nonlinear evolution equation, which does not seem to appear in any list of integrable equations, is provided. All the connected equations exhibit a very interesting symmetry structure enjoyed by the corresponding full hierarchies. Indeed, they all admit a hereditary recursion operator. Hence, each one of the mentioned equations represents the base member of a corresponding hierarchy of equations. These hierarchies are constructed via the recursive application of the respective recursion operators. The symmetry properties of such equations are recalled. Finally, we compare the net of links, derived via Bäcklund transformations, in the case of the fifth-order nonlinear evolution equations with an analog net of links connecting third-order Korteweg-de Vries (KdV) and modified Korteweg-de Vries (mKdV) equations. Analogies and discrepancies between the connections established in the case of fifth-order equations with respect to those established in the case of third-order equations are analyzed. This study aims to open the way for the construction of corresponding non-Abelian equations of the fifth order
Reimagining Professional Development in the Age of Artificial Intelligence
No full textAs Artificial Intelligence (AI) reshapes education, professional development (PD) must go beyond tool training to foster critical, meaningful integration. Initial PD should introduce AI’s uses and challenges, but also address the impact on teaching and learning. This paper explores and reflects upon Phase II of the FAITH project, a transatlantic design-based initiative developing an AI and Education (AI&ED) model for higher education. Effective AI pedagogy is grounded in socially constructed, hands-on experiences where educators design lessons, generate content, and critically assess AI outputs. Such approaches build confidence, competence, and prevent mechanical adoption. Leadership and policy must further support a dual PD strategy: immediate classroom applications alongside preparation for broader societal shifts. Early FAITH findings show introductory courses spark essential dialogue, but PD must remain dynamic, ethical, and intentional. Phase II combines theoretical exploration (e.g., sustainability, ethics) with context-relevant practice. Ultimately, AI&ED should be understood as a lifelong professional learning journey.FAIT
Resource Optimization in Multi-Hop IAB Networks : Balancing Data Freshness and Spectral Efficiency
No full textThis work proposes a multi-objective resource optimization framework for integrated access and backhaul (IAB) networks, tackling the dual challenges of timely data updates and spectral efficiency under dynamic wireless conditions. Conventional single-objective optimization is often impractical for IAB networks, where objective preferences are unknown or difficult to predefine. Therefore, we formulate a multi-objective problem that minimizes the age of information (AoI) and maximizes spectral efficiency, subject to a risk-aware AoI constraint, access-backhaul throughput fairness, and other contextual requirements. A lightweight proportional fair (PF) scheduling algorithm first handles user association and access resource allocation. Subsequently, a Pareto Q-learning-based reinforcement learning (RL) scheme allocates backhaul resources, with the PF scheduler’s outcomes integrated into the state and constrained action spaces of a Markov decision process (MDP). The reward function balances AoI and spectral efficiency objectives while explicitly capturing fairness, thereby resulting in robust long-term performance without imposing fixed weights. Furthermore, an adaptive value-difference-based exploration technique adjusts exploration rates based on Q-value estimate variances, promoting strategic exploration for optimal trade-offs. Simulations show that the proposed method outperforms baselines, reducing the convexity gap between approximated and optimal Pareto fronts by 68.6% and improving fairness by 16.9%
Mandatory sustainability reporting and the disclosure-performance gap : Insights from the EU Directive
No full textPurpose – This study aims to examine the influence of mandatory sustainability reporting on corporate transparency by determining whether such soft regulations drive a change toward substantive or symbolic transparency. It also investigates the role of country and industry in moderating a company’s compliance with sustainability-related regulations. Design/methodology/approach – The study uses difference-in-differences analysis to examine the influence of EU Directive implementation on corporate transparency, measured as the disclosure-performance gap. The treatment and control groups comprise companies from 17 European Union (EU) and 11 non-European Union Organization for Economic Cooperation and Development (non-EU OECD) countries, respectively. The timeperiod spans from 2013 to 2022, with 2017 being the year of the treatment, i.e. the EU Directive implementation.The study uses various approaches to address endogeneity issues, including the multiple specifications method, parallel trend analysis, propensity score matching and control variables. Findings – In the post implementation period of the EU Directive, EU companies exhibited a shift toward a wider disclosure-performance gap, i.e. symbolic transparency. Additionally, civil-law (common-law) countries tend toward substantive (symbolic) transparency, and industries under higher stakeholder scrutiny are engaged in symbolic transparency. Research implications – This study provides empirical evidence that mandatory sustainability reporting leads toward symbolic transparency, reinforcing legitimacy theory. It examines transparency by using the novel approach of assessing the gap between sustainability disclosures and actual performance. The ten-year period adds to the significance of the results. Practical implications – It provides insights for policymakers for designing future regulatory frameworks. Originality/value – First, it offers a novel approach to analyzing the effectiveness of mandatory sustainability reporting by emphasizing on distinguishing between substantive and symbolic transparency by evaluating the disclosure-performance gap. Second, while it provides a comprehensive analysis of all EU companies’ transparency, it also uncovers the country-level and industry-level heterogeneity arising from variations in institutional environmental and stakeholder pressures. Third, it fills the frequently highlighted industry-, regional and temporal gaps in the literature of mandatory sustainability reporting in general and the EU Directive inspecific.
A Gaussian fitting-based analysis method for multiple image radiography with integrated angular calibration, MIR2
No full textMultiple image radiography (MIR) is an X-ray phase-contrast technique that enhances soft-tissuevisibility by rejecting Compton scatter and capturing absorption, refraction, and ultra-smallangle x-ray scattering (USAXS) signals. Conventional MIR workflows depend on normalizationbetween object and reference datasets and precise angular alignment, making them sensitive todrift and prone to artifacts such as banding. We present an improved analysis framework, MIR2,which eliminates normalization and alignment by independently analyzing object and referencedata and applying angular calibration based on the dynamical theory of diffraction. Like MIR, itemploys pixel-wise Gaussian fitting of angular intensity profiles, but the MIR2 pipeline is simpler,less error-prone, and more robust against alignment-related artifacts. Importantly, artifact suppression is achieved intrinsically, without relying on additional correction algorithms. MIR2 wasimplemented in Python and validated at the BMIT beamline (Canadian Light Source, 33.3 keV,Si(220) double-crystal monochromator) using both test objects (PMMA step wedge, layeredpaper) and in vivo imaging of a live anesthetized mouse lung. Across both studies, MIR2 producedmore stable and artifact-reduced images than MIR. The method simplifies analysis workflows andsupports streamlined application of MIR in biomedical and material imaging under dose-limitedconditions.
The Data Lake as an Archive: Assessing the Suitability of Data Lakes for Long-Term Preservation of Big Data from thePerspective of Archival Theory
No full textThis study aimed to investigate, from an archival and information science perspective, whether data lakes can be considered a suitable option for longterm preservation of big data created in the public sector. As such, the study is an attempt to map and exemplify how archival and information science theory can be used to manage big data, a phenomenon that in itself illustrates the transformation of archives due to digitalization. The original purpose of the study was to conduct a case study of a practical implementation of a data lake in the Swedish public sector, but its direction changed due to unforeseen difficulties during data collection. The case study therefore took on a more theoretical outlook on the concept of the data lake, resulting in a literature review through which the central aspects, challenges, and opportunities of data lakes were identified. Challenges were deemed the most important points of analysis, and key challenges identified were the lack of context and structure in data lakes, their real-time collection of unstructured raw data, and their use of the schema-on-read principle. Utilizing a thematic analysis, these challenges were then analyzed using three archival and information science frameworks: the Records Continuum Model (RCM), the principle of provenance, and recordkeeping informatics. The analysis showed that the data lake concept, as described in the literature review, is incompatible with these frameworks given three main themes. Schema-on-read counteracts the proactive approach to metadata of the RCM and thereby long-term preservation. The lack of structure and context makes it difficult to apply the principle of provenance and thus to make collected data accessible, a process through which transparency and accountability can be ensured according to recordkeeping informatics. Finally, data lakes pose a risk from an information security and confidentiality perspective, as their structure and the way they complicate archival management processes such as appraisal and classification give rise to several risks in relation to GDPR and other personal data regulations governing the public sector. The result of the analysis led to the conclusion that the data lake concept in its current state is unsuitable for the long-term preservation of big data, but an excellent tool for short-term analysis and value creation. However, the conclusion leaves room for technological development which, with the help of archival and information science principles, could result in the implementation of data lakes that are compatible with such principles.Denna studie syftade till att utifrån ett arkiv- och informationsvetenskapligt perspektiv undersöka huruvida datasjöar kan anses utgöra ett lämpligt alternativ för långsiktigt bevarande av stordata som skapas i offentliga verksamheter. Därigenom utgör studien ett försök att kartlägga och exemplifiera hur arkiv- och informationsvetenskaplig teori kan användas för att hantera stordata, ett fenomen som i sig illustrerar arkivens förvandling i takt med digitaliseringen. Studiens syfte var ursprungligen att genomföra en fallstudie av en praktisk implementering av en datasjö i svensk offentlig verksamhet, men ändrade riktning på grund av problem i datainsamlingen. Fallstudien tog därmed en mer teoretisk inriktning på konceptet datasjö och mynnade ut i en litteraturstudie genom vilken datasjöns centrala aspekter, utmaningar och möjligheter identifierades. Utmaningar bedömdes utgöra de viktigaste analyspunkterna och centrala sådana som identifierades var datasjöars brist på kontext och struktur, deras realtidsinsamling av ostrukturerade rådata samt deras användning av schema-on-read-principen. I en tematisk analys har dessa utmaningar sedan analyserats med hjälp av tre arkiv- och informationsvetenskapliga ramverk, nämligen records continuum model (RCM), proveniensprincipen och recordkeeping informatics. Analysen visade att datasjökonceptet, så som litteraturstudien beskriver det, är oförenligt med dessa ramverk sett till tre huvudsakliga teman. Schema-on-read motverkar RCM:s proaktiva utgångspunkt för metadata och därigenom långsiktighet. Bristen på struktur och kontext innebär svårigheter att applicera proveniensprincipen och således tillgängliggöra insamlade data, en process genom vilken transparens och ansvarsutkrävande kan säkras enligt recordkeeping informatics. Slutligen innebär datasjöar en risk sett ur ett informationssäkerhets- och sekretessperspektiv, detta då dess uppbyggnad och försvårande av arkivhanteringsprocesser så som värdering och klassificering ger upphov till flertalet risker i förhållande till GDPR och andra personuppgiftsbestämmelser som styr offentlig verksamhet. Resultatet av analysen medförde att slutsatsen blev en bedömning av datasjökonceptet i dess nuvarande läge som olämpligt för långsiktigt bevarande av stordata, men som ett utmärkt verktyg för kortsiktig analys och värdeskapande. Slutsatsen lämnar dock utrymme för teknologisk utveckling som med hjälp av arkiv- och informationsvetenskapliga principer kan mynna ut i implementering av datasjöar som är förenliga med sådana.