I denna stuide undersöks hur lärare planerar, organiserar och utvärderar sin läsundervisning för flerspråkiga elever. Studien har genomförts med kvalitativa semi-strukturerade intervjuer med verksamma undervisande lärare i årskurs F-3. Syftet med arbetet är att bidra med kunskap om utifrån frågeställningarna hur lärare planerar, organiserar, genomför, anpassar och utvärderar sin läsundervisning för flerspråkiga elever. Urvalet hos lärarna har då varit att de är verksamma i ett flerspråkigt klassrum. Empirin analyserades sedan utifrån frågeställningarna och identifierades och presenterades i olika teman. 

I resultatens övergirpande fyra delar visade det sig att organisering och framför allt genomförandet av läsundervisningen beskrivs mer av lärarna än själva planeringen av undervisningen. Beskrivingen av lärarnas planering var nästintill obefintligt men däremot beskrev lärarna tydligt vilka arbetssätt och vilken arbetsgång deras läsundervisning har. Resultaten visar även att bedömningen av flerspråkiga elever upplevs som svår och osäker hos lärarna. Vidare visar resultaten att lärarna är överens om att ett bra sätt att utvärdera sin undervisning är genom olika analyser. De menar att genom analyserna kan de se framgångar och utvecklingsområden av sin undervisning och vilka anpassningar som bör sättas in. 

En av slutsatserna är att lärarnas planering verkar nästintill så självklar att den inte behöver någon större beskrivning. En annan slutsats är att det inte är någon skillnad på läsinlärningen oavsett vilket modersmål eleverna har, de använder samma metod till alla elever. En tredje slutsats är att lärarna upplever bedömningen av flerspråkiga elever är svårare då de upplever att det finns fler förutsättningar att ta hänsyn till, vilket kan göra att de ibland gör för snälla bedömningar. 

Denna studies syfte var att undersöka grundskollärares syn på bedömning av skriftliga uppgifter i en tid då generativ-AI finns tillgängligt för eleverna och hur man hanterar det. Syftet uppfylldes genom att göra en enkätstudie med lärare i grundskolan 6-9 och utifrån den göra en tematisk analys. Resultatet visar att generativ-AI ger en stor påverkan i bedömningssituationer för lärarna och att de står inför en stor förändring i både lärar- och bedömningspraktiken. Resultaten visar också att det saknas och behövs riktlinjer och kompetensutveckling för att lärarna ska kunna hantera den nya situationen. 

Syftet med denna studie är att undersöka lärares egna uppfattningar om elevers matematiska kunskapsnivå och hur relevanta baskunskaperna är i den frågan. Motivationen att göra denna studie är dels de senaste låga resultaten i ämnet matematik som elever i årskurs 8 och årskurs 9 i Sverige har uppvisat enligt PISA-undersökningens rapport (Programme for International Student Assessment), dels de åtgärder som lärarna vidtar i sin dagliga undervisning när de upptäcker bristfälliga kunskaper hos sina elever. Syftet med undersökningen är att få en uppfattning om vad lärarna själva anser om elevernas kunskapsnivå i de klasser de ansvarar för, samt vilka svårigheter lärarna upplever vara förekommande hos eleverna och hur de gör för att undervisningen ska uppfylla de syften som läroplanen beskriver. Enligt Skolverket (2022) är matematikundervisningens mål i Lgr22 att bidra till att eleverna ska utveckla förmågan att använda och beskriva matematiska begrepp och sambanden mellan dem, utveckla förmågan att använda matematiska metoder för att lösa problem och kunna föra och följa matematiska resonemang, och så vidare. 

I denna studie har en kvalitativ metod använts, där totalt åtta matematik- och klasslärare intervjuas med fokus på grundläggande färdigheter i matematik. De berörda lärarna är verksamma vid olika skolor för årskurs 4 till 9 i Linköpings kommun. Intervjuerna är semistrukturerade, vilket innebär att det finns en lista med tydliga frågor till samtliga deltagare i studien, men frågorna behöver inte ställas i en bestämd ordning. För analysen av insamlade data används tematisk analysmetod med induktiv ansats. Resultaten visar att de intervjuade lärarna har samma uppfattning om att det finns elever vars presterandenivå är låg. Andelen elever med lägre kunskapsnivå än förväntat varierar mellan en fjärdedel och en femtedel per klass. Lärarna anser också att elevernas kunskapsutveckling påverkas negativt av bristande grundläggande matematikkunskaper. Därför behöver de lägga extra tid och resurser på att dela klassen i mindre grupper för att kunna ge extra stöd åt elever som behöver det.

Gene regulation is a fundamental process in development and disease. Transcription factors (TFs) play a pivotal role by binding specific genomic regions to regulate target genes. In this thesis, we explored the landscape of TF binding through CUT&RUN, with a particular focus on Wnt signaling and its key mediator, β-catenin.

Paper I introduced CUT&RUN Low Volume Urea (LoV-U), optimized for co-factors like β-catenin. This method allowed for high-quality profiling of diverse targets in both cell lines and mouse tissues. Paper II identified common artifacts in CUT&RUN data, establishing a list of "Suspect" regions for data filtration. In Paper III, we used these tools to examine time-resolved β-catenin binding in two cell types and discovered that binding is dynamic over time and cell-type specific. Paper IV addressed one of the central challenges in TF and chromatin research — signal reproducibility. We developed ICEBERG (Increased Capture of Enrichment by Exhaustive Replicate aGgregation), a pipeline to improve the detection of TF binding across the genome. ICEBERG classified binding sites based on detection probability and uncovered previously missed, rare regulatory associations. In Paper V, we shifted focus to a broader landscape of mouse development. We created a CUT&RUN resource dataset of twelve targets in four embryonic tissues and identified "popular regions" bound by multiple TFs, enriched for essential developmental genes. Finally, in Paper VI, we identified a set of genomic regions where CTCF binding changes in response to Wnt activation. These regions overlap with β-catenin and are associated with changes in 3D genome architecture. By disrupting CTCF binding, we demonstrated that CTCF contributes to the regulation of key Wnt target genes.

Together, these studies represent a methodological and conceptual advance in the study of gene regulation, shedding new light on the nuclear mechanisms of Wnt/β-catenin signaling, and providing tools and methods for future research.

A 5 W flyback converter with multiple adjustable voltage outputs in the 5–12 V range has been constructed. The converter has a switching frequency of 200 kHz and operates in discontinuous-conduction mode (DCM). A current mode PWM controller has been used, with feedback delivered using a voltage reference (TL431) and an optocoupler. Simulations have been carried out to validate the design choices and further development of a prototype. The prototype has been measured and an improved final product has been developed. The resulting efficiency was measured to approximately 87 %, which is considered an acceptable result for such a low power converter. Some tests have deliberately been left out. To ensure that the final product will continue to deliver no matter condition, some additional testing might be required.

Med utgångspunkt i Henri Lefebvres och Michel Foucaults respektive rumsteori undersöker denna uppsats relationen mellan makt, människa och rum i Ursula K. Le Guins science fiction-roman The Left Hand of Darkness. Genom att tillämpa Lefebvres begrepp rumslig triad och social produktion i kombination med Foucaults idéer kring hur makt uttrycks och fungerar i rumsliga sammanhang, granskar analysen dels hur maktstrukturer påverkar och tar sig form i romanens rum och platser, och dels hur romanens huvudkaraktärer påverkas av och förhåller sig till dessa platser. Därtill jämförs dessa rum och platser för att se hur de liknar varandra och skiljer sig åt beroende på deras respektive maktordning, samhällsstruktur och vilken karaktärs perspektiv som används för att skildra platserna i fråga. Uppsatsen visar att de makt- och samhällsstrukturer som skildras i romanen har en märkbar inverkan på hur berättelsens olika rum och platser utformas, upplevs och navigeras av karaktärerna i berättelsen. Dessutom påverkas karaktärerna av de rum och platser de vistas i, vilket också syns i hur de beskriver och förhåller sig till platserna i fråga. Vidare påverkas huvudkaraktärernas förhållanden till olika platser dels av deras upplevelser av andra platser, dels av deras relationer till varandra. uppsatsen uppmuntrar till fortsatta rumsteoretiska analyser av romanen, i synnerhet utifrån den rumsliga triaden för att granska symboliska, kroppsliga och fysiologiska upplevelser och skildringar av rum.

The Toffoli gate is an important universal quantum gate, and will alongside the Clifford gates be available in future fault-tolerant quantum computing hardware. Many quantum algorithms rely on performing arbitrarily small single-qubit rotations for their function, and these rotations may also be used to construct any unitary from a limited (but universal) gate set. How to carry out such rotations is then of significant interest. In this work, we evaluate the performance of a recently proposed single-qubit rotation algorithm using the Clifford plus Toffoli gate set by implementation of a one-shot version on both a real and a simulated quantum computer. We test the algorithm under various simulated noise levels using a per-qubit depolarizing error noise model and examine how the probabilities and process fidelities are affected. We then conduct live runs and find that the results reasonably match the simulated results. We also attempt to model the hardware noise by combining a number of noise models, matching the results to results of the live runs to approximate the hardware noise. Our results suggest that the algorithm will perform well for up to 1%noise under the noise models we chose. We further posit the use of our algorithm as a benchmark for quantum processing units, given that it has a low complexity that is easy to fine-tune in small steps. We provide details for how to do this.

The aviation industry faces the urgent challenge of reconciling projected traffic growth with stringent climate targets, including the European Union’s goal of climate neutrality by 2050. While techno-logical innovations such as sustainable aviation fuels and hydrogen propulsion are critical for long-term decarbonization, their near-term impact remains limited. In this thesis, we address the immediate opportunity of improving the environmental efficiency of aircraft arrival operations in Terminal Manoeuvring Areas (TMAs), where congestion and complex sequencing often lead to excess fuel burn, emissions and noise.

Existing research has demonstrated the benefits of Continuous Descent Operations (CDOs) and structured arrival procedures such as Point Merge (PM). However, for TMA performance evaluations, cur-rent practices are limited in real-world assessments of environmental efficiency beyond Carbon Dioxide (CO2). In terms of optimization of the arrival operations, the exploration of dynamic PM usage and early speed adjustments as a sequencing tool, integrated within the optimization framework, remains relatively unexplored. Furthermore, most optimization frameworks do not fully integrate arrival and departure scheduling in mixed-mode runway environments.

The thesis addresses four key research questions: (1) What are the environmental benefits of using fuel-efficient CDOs within TMA and how to quantify them? (2) How to evaluate the arrival aircraft performance within TMAs implementing PM procedures? (3) Can the performance of current operations in a TMA where PM is implemented be improved by using optimization? (4) How can speed adjustment during cruise and descent be used, with or without combining it with PM procedures, to safely separate and sequence arriving flights? The overarching aim is to develop methodologies for assessing and optimizing arrival operations to reduce fuel consumption, emissions and noise while maintaining safety and capacity.

The research combines data-driven performance evaluation with mathematical optimization. We use real-world Automatic Dependent Surveillance Broadcast (ADS-B) trajectory data from the OpenSky Network, together with meteorological data from ECMWF ERA5 and the performance modeling capabilities of EUROCONTROL BADA, to quantify inefficiencies in current operations. For optimization, we develop a Mixed-Integer Programming (MIP)-based framework to assign conflict-free, fuel-efficient arrival routes and descent profiles, incorporating wake turbulence separation and runway scheduling constraints. We perform case studies at Dublin and Oslo-Gardermoen airports, both operating with PM procedures.

Empirical analysis reveals substantial inefficiencies in current TMA operations, with level-flight segments contributing significantly to excess fuel burn. Our studies show that CDOs reduce fuel consumption and noise exposure, although Carbon Monoxide (CO) and Hydrocarbon (HC) emissions may increase under idle-thrust conditions. Optimization results demonstrate that dynamic PM usage and early speed adjustments improve horizontal and vertical efficiency, reduce time in TMA and decrease fuel burn compared to actual operations.

Operational improvements in arrival management can deliver immediate environmental benefits, complementing long-term technological solutions. The proposed frameworks support SESAR and NextGen objectives for greener TMAs and are adaptable to future contexts, including Urban Air Mobility (UAM). By integrating rigorous performance evaluation with advanced optimization, this thesis provides actionable insights for reducing aviation’s environmental footprint while enhancing predictability.

Platforms mediating care services are increasingly reshaping the geographies of social reproduction, offering care fixes to some, while exacerbating the crisis of care for others. In this paper, we draw on research on healthcare, deliveries and cleaning platforms in Sweden to argue that platforms reinforce and redistribute flows of care privilege and care poverty between the Global North and the Global South, between urban and rural locations, and within cities, thus deepening the uneven geographies of care. These uneven geographies apply to those performing social reproductive work as well. While the working conditions of migrant platform cleaners have clear repercussions for their own social reproduction, other segments, such as healthcare professionals, may experience platform work as a fix to both working conditions and work–life balance. Ultimately, we propose that the political economy of social reproduction unfolds as a spatially uneven process, making life easier for some and harder for others.

In this research, we explore how the physical setting within refugee camps contributes to shaping social interactions of displaced children aged 6-12. By examining the spatial and social dimensions of refugee camps / accommodations, we aim to investigate how such challenging environments shape children's social connections and interactions with their communities and surroundings and what adaptive strategies they follow to get by. 

We chose to conduct the study across two different settings, Sweden and Syria, since these countries represent two completely different stages of refugees’ route to one of their final destinations, Northern Europe. Syria is one of the very first countries refugee people go through, whereas Sweden is one of their final ones, or even the final in many cases. This fact also allowed for both diversity in capturing displaced children’s experiences, and a deeper understanding of how different physical environments (camps or accommodations) shape their social interactions. In order to achieve this, we gathered data from those who work daily and closely with children in camps. We conducted semi-structured interviews with camp workers, allowing them to share detailed observations on how camps as physical environments affect children’s social interactions. Later on, thematic analysis was applied to the data collected, focusing on the spatial and social dimensions of children's experiences as observed by workers. 

We believe this research contributes to child studies by providing insight into the complex experiences of displaced children in refugee camps, through focusing on daily social activities of children who face extreme conditions and challenges that affect all aspects of their lives and exploring their ways of adapting with these settings and practicing agency. We expect our findings to highlight how children adapt within the camp community and, hopefully, provide concrete insights for NGOs, educators, and policymakers working in refugee camps, thus contributing to the creation of supportive environments, responsive to the unique needs of displaced children for well-rounded social interactions.

This thesis studies the representation theory of finite groups, with an emphasis on tensor product decompositions. We begin by introducing the basic concepts of group representations, modules, homomorphisms and equivalence of representations. We then develop the theory of characters, inner products of characters, and the decomposition of the group algebra. The symmetric group S₃ is used throughout as a running example to illustrate the theory. We then investigate the irreducible representations of S₄. We first compute the character table of S₄ followed by finding its Clebsch-Gordan table. By these calculations, we derive a general formula for the number of simple components contained in a k-tensor product of simple modules of S₄. These results illustrate key ideas in representation theory and provide a concrete introduction to more general methods.

Minns du vad du gjorde den 21 oktober 2015? För många av oss är det ett datum med särskild betydelse, vi som delar den generationserfarenhet som brukar gå under rubriken ”Generation X” och som är födda någonstans mellan 1965 och 1985. Själv högtidlighöll jag detta datum, som ganska många andra, genom att titta på filmen Tillbaka till framtiden del II. Det är inte en film som har åldrats med värdighet. Och om vi ska vara riktigt ärliga var den inte särskilt bra ens då det begav sig...

Debugging is essential for maintaining reliable software systems. However, it is often a time‑consuming process constrained by the manual effort required to both reproduce reported bugs and reduce the action sequences that trigger them. This thesis asks how effectively large language model (LLM)–based agents can autonomously reproduce industrial control software bugs and minimise their Steps to Reproduce (S2R), and how agent architecture and tool access shape this capability. Three agent architectures are evaluated: one Single‑Agent mode and two multi‑agent modes (Producer–Reviewer and Orchestrator–Workers). These are tested under controlled conditions that vary tool access (a restricted shell‑primitive set versus an unrestricted command‑line interface) on a real‑world dataset of industrial control software bug reports whose S2Rs are augmented with injected redundant steps. Experimental results indicate that the Single Agent achieves the highest reproduction success, stronger S2R minimisation, and lower runtime than the multi‑agent variants. Specifically, both the Single Agent and Producer–Reviewer reproduce 55.56% of bugs, while Orchestrator–Workers reproduces 44.44%. When reproduction succeeds, the best‑performing agent removes an average of 90% of injected redundant steps, producing substantially shorter and clearer reproduction sequences. Across all architectures, unrestricted command execution outperforms curated shell‑primitive tools in both reproduction success and minimisation quality. Overall, the findings demonstrate that LLM‑driven agents can autonomously reproduce real industrial bugs and simplify the action sequences required to trigger them, and that architectural simplicity outperforms more complex multi‑agent coordination for this task. This unified treatment of reproduction and reduction informs practical design choices for autonomous debugging in industrial test environments.

This thesis aims to bridge classical geometric control theory with the theory of structured systems within a network-based framework. The proposed methodology integrates structural and geometric control techniques and leverages the key observation that the subspaces of interest contain or are contained in a subspace generated by canonical basis vectors. Since these canonical vectors correspond to individual network nodes, the algebraic analysis of such subspaces can be directly translated into an analysis over node sets.

The first paper included in this thesis investigates the controllability of temporal networks, modeled as linear, piecewise-constant, time-varying dynamical systems. In particular, an upper bound on the number of snapshots, i.e., time intervals during which the system matrices remain constant, necessary to achieve controllability, is derived. The proof exploits structural properties by decomposing the controllability subspace associated with each snapshot into a component which varies depending on the specific entries of the system matrix, and a fixed component, independent of those entries.

The second paper introduces the minimal node cardinality disturbance decoupling problem, studied under three feedback paradigms: state feedback, output feedback, and dynamic feedback. The formulation and analysis of this problem rely on classical concepts from geometric control theory, such as controlled and conditioned invariant subspaces. By exploiting the fixed structure of these subspaces, they can be reformulated as sets of nodes, providing a network-based interpretation that enhances both intuition and visual interpretability.

The third paper applies the developed theoretical framework to the minimal input cardinality disturbance decoupling problem via output feedback in a system of coupled oscillators, linearized around a stable phase-locked synchronized state, and applies it to the disturbance decoupling of power grids. This application illustrates the practical relevance and versatility of the proposed approach.

Overall, the thesis shows that, under mild assumptions, complex geometric control structures can be represented in an intuitive and visually meaningful manner when analyzed through the lens of networks. This perspective not only facilitates the study of large-scale systems, but also opens new research directions at the intersection of geometric control and network theory.

Bakgrund

Trots att skadeförebyggande arbete är avgörande för en hållbar idrottskarriär inom barn- och ungdomsidrott i badminton, saknas det specifik forskning som belyser tränarnas roll och erfarenheter av skadeförebyggande strategier för barn och unga. 

Syfte

Syftet med denna studie var att undersöka tränarnas tankar och erfarenheter av skadeförebyggande arbete inom barn- och ungdomsbadminton med elitsatsning.

Metod

Studiens design är en kvalitativ intervjustudie med induktiv ansats där intervjuerna analyserades med hjälp av innehållsanalys.

Resultat

Analysen av intervjuerna landade i tre kategorier med tillhörande subkategorier; Skadeförebyggande insatser värderas högt men genomförs varierat, Ledarskap, kunskap och vägledning som grund för skadeförebyggande insatser, Utmaningar och utvecklingsmöjligheter i skadeförebyggande arbete. 

Konklusion

Studien visar att badmintontränare inom barn- och ungdomselit upplevde skadeförebyggande arbete som viktigt för spelarnas utveckling, men samarbeten med fysioterapeuter sker oftast först vid skada. Trygga relationer och god kommunikation mellan tränare och spelare är centrala, medan begränsade resurser, bristande utbildning och svårigheter att sprida riktlinjer utgör utmaningar. Förebyggande arbete sker redan men bygger mest på individuella lösningar och erfarenhet, vilket tyder på ett behov av mer strukturerade riktlinjer och gemensam kunskapsbas för tränare.

Bakgrund: Vid träning frisläpps kreatinkinas från musklerna vilket indikerar på muskelskada. Kallvattenimmersion är en återhämtningsmetod som har visats påskynda muskelåterhämtningen efter träning. 

Syfte: Syftet var att systematisk kritiskt granska litteraturen för hur muskelåterhämtning avseende kreatinkinas påverkas efter kallvattenimmersion ≤10 minuter efter träning.

Metod: Databaserna PubMed, Scopus och CINAHL användes för att hitta stöd i befintlig forskning inom området. För att värdera artiklarnas metodologiska kvalité samt bedömning av evidens användes Fowkes & Fultons granskningsmall respektive GRADE. 

Resultat: Tio studier inkluderades där sju var randomiserade kontrollerade studier och tre var crossover studier. Samtliga studier hade en interventionsgrupp som genomförde kallvattenimmersion i ≤10 minuter och en kontrollgrupp som genomförde passiv vila. Vattentemperaturen varierade mellan 6–15°C. Två studier visade stor effektstorlek för att kallvattenimmersion hade en påskyndande effekt för muskelåterhämtning. Åtta studier visade ingen påskyndande effekt för muskelåterhämtningen. Det fanns både större och mindre brister i studiernas metodologiska kvalitet och den sammanlagda evidensen bedömdes ha mycket låg tillförlitlighet. 

Konklusion: Evidens med mycket låg tillförlitlighet för att kallvattenimmersion i ≤10 minuter efter träning inte påskyndar muskelåterhämtningen jämfört med en passiv återhämtning hos personer som har tränat.

In Sweden, one in four students in an average school class is estimated to be at risk of developing reading and writing difficulties. The purpose of this study is to investigate the experiences of primary school teachers, special education teachers, and special educational needs coordinators in creating inclusive education for students with reading and writing difficulties, as well as the concrete adaptations used to foster inclusion in the classroom. Furthermore, the study examines how inclusion work affects students' self-esteem. The study is based on a qualitative research design, where eight semi-structured interviews were conducted with active teachers, special education teachers, and special educational needs coordinators in grades 1–3. The collected empirical data was analyzed using thematic analysis, which formed three main themes: Individualized inclusion work, When ideals meet reality, and Students' self-esteem is positively affected by leaving the classroom. The results show that teachers operate from an inclusive pedagogical perspective and use individualized strategies to meet the needs of all students. Key success factors for this work include clear structure, visual aids, varied working methods, relationships, and opportunities for students to demonstrate their knowledge in different ways. At the same time, it emerges that many educators experience obstacles in the form of insufficient time, resources, and a gap between theory and practice to address the needs of students with reading and writing difficulties. The results also indicate that the self-esteem of students with reading and writing difficulties is crucial to whether they feel included or not. The conclusion is that contributing to classroom inclusion requires systematic support from school leadership. Both learning and self-esteem are strengthened when students feel secure and their needs are met.

This article bridges optimization and control and presents a novel closed-loop control framework based on natural gradient descent, offering a trajectory-oriented alternative to traditional cost-function tuning. We leverage the Fisher information matrix to formulate a preconditioned gradient descent update that offers flexibility in shaping closed-loop behavior. To this end, the proposed method parameterizes closed-loop dynamics in terms of stationary covariance and an unknown cost function, providing a geometric interpretation of control adjustments. We establish theoretical stability conditions. The simulation results on a rotary inverted pendulum benchmark highlight the advantages of natural gradient descent in trajectory shaping.

The current third party Ada syntax highlighter in Emacs called Ada-mode is prone to cras-hing. This bachelor’s thesis investigated the possibility of using the Storm server to providesyntax-highlighting instead. The aim was to investigate whether Storm could offer fastersyntax highlighting. This was achieved by transcribing the Ada grammar into Storm’s TSLformat and conducting performance tests, primarily focused on editing times in Ada fi-les from student projects. Even though the results from the testing show that Ada-modeoutperformed Storm-mode, Storm-mode held its own and could in the future be a viablealternative to Ada-mode.

Background

Surgery may be the only chance for curative treatment of liver tumors and ensuring long-term survival. The extent of major liver resection is limited by the volume of liver parenchyma remaining after surgery. When the remaining liver volume is too small, the patient is at risk of posthepatectomy liver failure, which is a highly feared complication after extensive hepatic resection. In attempts to avoid posthepatectomy liver failure, hypertrophy of the future liver remnant can be induced before surgery. The aim of this thesis is to provide new and/or greater insights, along with possible benefits, problems and limitations, of the materials, methods, and measurements used for portal vein embolization and associated techniques for inducing liver hypertrophy.

Methods/materials

The studies reported in Papers I, III and IV were conducted as retrospective, multicenter cohort studies, and the outcomes after portal vein embolization were analyzed. Paper II presents a long-term follow-up survival analysis from a randomized controlled trial.

Results

Paper I: Compared with microparticles, the use of N-butyl-cyanoacrylate as an embolizing material results in a greater degree of hypertrophy and a higher resection rate, as well as lower radiation exposure, a shorter fluoroscopy time, and less contrast used, within the context of portal vein embolization.

Paper II: With a median follow-up time of almost eight years, no significant differences in long-term survival were detected between patients who underwent two-stage hepatectomy and patients who underwent associating liver partition and portal vein ligation for staged hepatectomy. In a multivariable analysis, the factors influencing long-term survival were extrahepatic disease, an increasing size of the largest liver metastasis, liver tumor-free status at the first follow-up, and a primary rectal tumor.

Paper III: Compared with portal vein embolization characterized by normal P-bilirubin levels, hyperbilirubinemia at the time of portal vein embolization did not result in decreased liver hypertrophy, as measured by absolute and relative growth, degree of hypertrophy, or kinetic growth rate. N-butyl-cyanoacrylate and elevated P-bilirubin levels were associated with increased liver hypertrophy.

Paper IV: Hypertrophy of the future liver remnant, as measured by absolute and relative growth, degree of hypertrophy, or kinetic growth rate, did not significantly differ between groups categorized by the timing of tumor clearance from the future liver remnant in relation to portal vein embolization.

Conclusions

The use of N-butyl-cyanoacrylate as an embolizing material is superior to the use of microparticles within the context of portal vein embolization.

Portal vein embolization is safe and exhibits a high success rate.

Two-stage hepatectomy and associating liver partition and portal vein ligation for staged hepatectomy yield similar long-term oncological outcomes in patients with advanced colorectal liver metastasis.

Hyperbilirubinemia at the time of portal vein embolization does not impair induced liver hypertrophy.

Similar results of liver hypertrophy are observed after portal vein embolization, regardless of whether the future liver remnant is cleared before the portal vein embolization procedure.

Achieving both high hardness and toughness in protective coatings is a formidable challenge. Here, we harness an original superlattice architecture that synergistically combines Koehler hardening with coherent interfaces to reduce the crack driving forces and enhance toughness, enabling coatings with improved damage tolerance. We engineer epitaxial HfN1.33/Hf0.76Al0.24N1.15 superlattices on MgO(0 0 1) substrates using low-energy, high-flux ion-assisted reactive magnetron sputtering. These superlattices, with bilayer periods ranging from 6 to 20 nm exhibit a unique three-fold superstructure, where each layer forms a three-dimensional checkerboard pattern. HfN1.33 forms a checkerboard period of 7.5 Å from self-organization of Hf-vacancies and N-interstitials, whereas Hf0.76Al0.24N1.15 forms a period of 12.5 Å due to spinodal decomposition. Mechanical testing reveals high hardness (∼36 GPa), matching that of Hf0.76Al0.24N1.15 and exceeding the softer HfN1.33 (∼27 GPa), reflecting interface-driven Koehler strengthening. Micropillar compression tests show distributed cracking along {1 1 0}〈1 1 0〉 slip systems and faster mechanical recovery than Hf0.76Al0.24N1.15 for improved toughness, corroborated by cube-corner indentation fracture analysis, while not reaching the superb toughness and plasticity of HfN1.33. These results demonstrate that epitaxial HfN1.33/Hf0.76Al0.24N1.15 superlattices combine high hardness and strength with good toughness for an improved film damage tolerance.

The aim of this thesis is to develop an electroactive in vitro model of the bone marrow(BM), termed "bone-on-a-chip," designed to replicate the BM microenvironment.Polypyrrole (PPy) doped with succinic acid (SUCA) was electropolymerized ontoa gold surface using a three-electrode setup. Chondrocyte-derived plasma membranenanofragments (PMNFs) of KUSA cells origin were covalently attached to the PPy(SUCA)surface via EDC/NHS coupling. The redox states of PPy were modulated from asfabricated (AF) to oxidized (OX) and reduced (RE) to mimic variations in bone density.The PPy surfaces were mineralized through the growth of minerals during a three-dayincubation in cell culture media. Characterization techniques assessed surface propertiesat each fabrication step. Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) measured electrochemical performance, while water contact angle (WCA)gauged hydrophobicity. SEM and energy-dispersive X-ray spectroscopy (EDX) evaluatedsurface morphology and mineral content post-mineralization. Further cell studies wereconducted on the mineralized PPy surfaces to assess their ability to support stem cell viability and remain in their undifferentiated state as stem cells without further differentiationinto other cell types. Mouse hematopoietic stem cells (HSCs) were subsequently isolated,sorted, and cultured on the surfaces.The RE state demonstrated the highest WCA and mineral content, followed by OXand AF. EDX confirmed significant differences in calcium (Ca) and phosphorus (P) levelsbetween surfaces with switched redox states (OX and RE) compared to AF and the negativecontrol. SEM images revealed clear differences seen in morphology between the redoxstates, although these differences appeared challenging to control. Preliminary cell studiessuggested that redox-modified surfaces might support undifferentiated HSCs better thanthe AF state and controls; however, reproducibility remains a challenge.This study highlights the potential of electroactive surfaces in creating a bone-on-achip model to better understand bone marrow dynamics and stem cell behavior, pavingthe way for applications in stem cell research and therapeutic advancements. 

This master thesis was carried out at SSAB Oxelösund and aims to improve production efficiency in Quenching Line 4 by developing an optimization model for the plate flow. The background is SSAB’s transition toward fossil free steel production, where a new electric furnace replaces coal based blast furnaces. This change increases the production load on Quenching Line 4, creating higher demands on planning and resource utilization. An integer programming model was developed in Pyomo to optimize the sequence of plates through the furnace, considering capacity constraints, cooling requirements, setup times, and LIFO rules. The objective function was formulated to balance the total throughput time and furnace utilization. The results show that the developed model improves both throughput time and furnace utilization compared with the current operator strategy. With an effective balance setting, the makespan was reduced by 24 percent, while the average utilization increased from 36.5 to 51.9 percent. The study demonstrates that the model can serve as a foundation for a future decision support tool for crane operators. With further development, including temperature depen dent heating times, improved data integration, and shorter computation time, it could be used in real time for production planning and decision making. The work thus contributes to a more efficient and sustainable production process in the transition toward fossil free steelmaking.

MoonBoard is a standardized climbing wall with an accompanying mobile app used by climbers worldwide to create, share and find climbing problems. The system has over 100,000 users and more than 200,000 unique problems. Despite this large dataset, the current app’s search and filtering options mainly rely on grades, repeats and other basic parameters. These simpler filters cannot fully capture important aspects such as climbing styles or hold types, which are crucial for climbers aiming to train specific techniques or reach personal goals. This thesis addresses this limitation by applying machine learning and visualization techniques to better characterize and recommend climbing problems on the MoonBoard. This thesis presents a system that applies machine learning and interactive visualization to enhance problem selection on the MoonBoard. Climbing problems were classified based on hold types (crimp, pinch, jug, pocket, and wood) and movement characteristics (dynamic, reachy, and cross-over). Data from the MoonBoard was preprocessed, and climbing sequences were computed using BetaMove. These sequences were analyzed with various machine learning models to identify relevant problem attributes. Dimensionality reduction and clustering were then applied to visualize similarities between problems, which were integrated into a web-based interface allowing users to explore, filter, and receive recommendations for both similar and diverse problems. A user evaluation showed that the system met its main objectives: climbers were able to find suitable problems more efficiently and appreciated the added filters and similaritybased ecommendations. The tool achieved an average SUS score of 88.75 indicating excellent usability, although limitations were noted, particularly in the accuracy of movement classification and problem similarity calculations. These findings demonstrate that combining machine learning with interactive visualization can enhance problem selection on the MoonBoard. While the current system only covers a subset of climbing styles, it provides a foundation for future extensions incorporating more advanced models, additional movement types, and broader applications within indoor and outdoor climbing contexts.

This thesis investigates how low-power edge AI can be used for privacy-conscious crowd analytics by deploying an end-to-end system on a Raspberry Pi 5 equipped with a Hailo-8 accelerator. The work targets doorway-level crowd monitoring in indoor environments, where traditional cloud-based surveillance raises concerns about latency, bandwidth, cost, and data protection. Two main questions are addressed: how effectively a YOLO-based detection and lightweight person re-identification (Re-ID) pipeline can be deployed on this hardware for real-time crowd counting, and how an interactive dashboard can turn the resulting data into actionable insights for decision-makers. The implemented system consists of three subsystems: an AI pipeline, a privacy-preserving event database, and a web-based visualization layer. The pipeline uses a YOLOv8s detector and a multi-object tracker that combines centroid distance, Intersection over Union, and motion prediction to maintain stable identities. A person Re-ID module links entry and exit events at the doorway. Instead of storing video, the system writes anonymized visit records (entry time, exit time, dwell time) to a local SQLite database. A Dash-based dashboard running on the Pi presents these events through coordinated time-based views and summary indicators. Evaluation on a realistic video sequence shows that the pipeline sustains approximately 25–30 FPS, with detection latency of 0.03–0.04 s per frame and stable CPU and memory usage, confirming real-time feasibility on-device. The baseline full-precision Re-ID model achieves over 90% on precision, recall, and F1-score, while the quantized model exhibits reduced recall but remains usable under edge constraints. A study with 12 participants, complemented by Mini-VLAT and SUS, indicates that the dashboard supports typical staffing and promotion-planning tasks and attains an average SUS score of 69.4, slightly above the standard usability benchmark. Together, these results demonstrate that real-time, privacy-conscious crowd analytics are achievable on constrained hardware, while also highlighting trade-offs in model quantization and opportunities for further dashboard refinement.

The European Union's Corporate Sustainability Reporting Directive (CSRD) aims toenhance and standardize sustainability reporting to support the green transition. While the public sector is expected to lead in sustainability, the application of CSRD within Swedish municipalities is fragmented. Many municipally owned companies fall below reporting thresholds individually, creating a transparency gap despite the municipal group often exceeding these thresholds collectively. This discrepancy raises moral and strategic questions regarding public accountability and the effective utilization of the framework.

This thesis investigates the rationale for the proactive adoption of CSRD principles bySwedish municipalities, irrespective of formal legal requirements. It further evaluates the capacity and competence of these municipalities to utilize the CSRD framework strategically as an Environmental Management System (EMS) for governance and innovation, and analyzes the potential synergies and consequences of its implementation.

A qualitative, abductive case study approach was employed. Empirical data wasgathered through semi-structured interviews with key officials from six large Swedish municipalities and supplementary expert interviews. The data was analyzed using thematic analysis, guided by a theoretical framework integrating Porter and Van der Linde’s innovation-driving policy principles and the Coase Theorem, focusing on value chain innovation and the internalization of externalities.

The findings reveal a strong moral consensus among municipalities regarding their dutyto exceed private sector transparency standards. Proactive adoption enables the utilization of CSRD as an EMS, moving beyond mere compliance. The Double Materiality Analysis (DMA) is identified as the most valuable component for strategic prioritization. However, implementation is significantly challenged by the inertia of municipal governance structures and a high dependency on an immature consultancy market, where expertise is often developed concurrently with the clients. This dependency risks reducing CSRD to a bureaucratic "paper product" rather than a strategic management tool. Furthermore, the study identifies a critical lack of centralized support and coordination from national bodies (e.g., SKR), leading to systemic inefficiencies and redundant efforts.

Proactive adoption of CSRD by municipalities offers substantial potential for creatingsocietal value by internalizing externalities and driving sustainable innovation through mechanisms like public procurement and access to finance. To realize CSRD’s potential as an effective EMS, the municipalities must prioritize internal competence development over consultant dependency, overcome organizational inertia, and urgently address the vacuum in centralized national support and coordination.

Despite decades of scientific consensus on anthropogenic climate change, global mitigation efforts remain insufficient to curb its escalating impacts on nature and human societies. Effective climate change mitigation requires engagement across multiple levels – from policy and technological innovation, to large-scale shifts in individual behavior. Research over the past decade suggests that one reason individual behavior change has had limited impact is that people often choose actions with low objective mitigation potential. While many individuals engage in pro-environmental actions, these are often low-impact behaviors (low CO₂ reduction), whereas high-impact behaviors (high CO₂ reduction) remain largely unchanged.

This thesis operationalizes the limited adoption of high-mitigation behaviors as a manifestation of “impact neglect” – a phenomenon shaped by both low carbon numeracy (i.e., limited knowledge about the relative mitigation potential of different actions), as well as psychological and structural barriers associated with high-impact behavioral shifts. This research identified mental imagery as an underexplored intervention within the field of behavioral science for climate change mitigation. Drawing on evidence from research in memory, athletic performance, and clinical interventions, this research proposed that mental imagery may address both facets of impact neglect: by enhancing the learning and retention of objective mitigation potential (i.e., improving carbon numeracy), and by lowering psychological barriers, thereby increasing motivation to engage in high-impact pro-environmental behaviors.

The first two papers in this thesis tested the effect of simulated mental imagery on psychological determinants for climate change behaviors. Support was found for a causal relationship between mental imagery of climate change-related risk events (i.e., wildfire), affective valence, and global perceived risk of climate change. Furthermore, combining mental imagery with information about mitigation potential, in a feedback format, lead to a significantly larger increase in pro-environmental intention, when compared to only receiving the information feedback. However, mental imagery increased the impact belief (i.e., perceived mitigation potential) globally, thus not necessarily increasing carbon numeracy, whereas feedback by itself led to more correct impact beliefs.

Lastly, this research examined spontaneous mental representations of climate change – including both mental images and verbal thoughts – and tested how these representations were related to psychological determinants of climate action and policy support. Using structural topic modeling, certain types of spontaneous representations were identified as systematically associated with pro-environmental outcomes. Representations that were more emotionally charged, negatively valenced, and emphasized long-term consequences were linked to a pro-environmental profile. In contrast, representations that were vaguer, more detached, and reflected prototypical or media-reproduced content were associated with lower levels of pro-environmental engagement.

Drawing on literature from mental imagery and pro-environmental behavior, together with the empirical findings from the three studies in this thesis, the Imagery–Behavior Feedback Model was created – a conceptual framework outlining paths through which mental imagery may influence behavior. The findings presented here provide initial support for several paths of the model: first, that spontaneous mental representations were meaningfully related to pro-environmentalism; and second, that simulated mental imagery could evoke experiences that lead to updated psychological determinants of action.

Furthermore, the empirical evidence suggests several underlying mechanisms through which mental imagery may reduce impact neglect: (1) increasing the availability of relevant information, (2) enhancing perceived behavioral control, (3) strengthening the perceived value of acting, and (4) supporting risk-related deliberation. Future research should endeavor to test these mechanisms and the feasibility of the Imagery–Behavior Feedback Model.

Participation in a Megagame and a consecutive facilitated debriefing can enable large groups to co-construct awareness and knowledge about the complex relations involved in developing sustainable futures for current societies. “Switching the current” is a megagame where 20 to 100 participants collaboratively experience the challenges involved in shaping future energy consumption patterns and energy production solutions for their region, just and sustainable from a social, economic and environmental perspective. This paper explores how collective discussions can be visualized to better understand how in-game experiences are connected to real-world phenomena. Doing so, a time-snake method has been developed which visually depicts how reflective group discussions move between “game content” and “real world phenomena”, as well as between “attention to details” and “a broad systems perspective”. The comprehensive analysis and comparison of two debriefing sessions shows how participating in a Megagame can develop awareness and understanding of complex relations in shaping sustainable futures of societal energy systems. It also demonstrates how applying time-snake visualizations can help to evaluate the richness of the reflective co-construction that occurs in facilitated debriefings.

This thesis presents a comparative analysis of the European Union (EU) and the People's Republic of China (PRC) in their efforts to achieve environmental sustainability while maintaining economic growth. In an escalating global climate challenge, both regions play a crucial role in shaping international climate governance. The study examines three key research questions: (1) How are the sustainability policies of the EU and China similar and different? (2) What are the strengths and weaknesses of the EU's regulatory strategy and China's state-led green initiatives in achieving sustainability goals? (3) To what extent has economic growth in the EU and China been decoupled from carbon emissions in 2023-2024? Using qualitative case study methodology, the paper draws on policy documents, including the European Green Deal, the EU Emissions Trading Scheme, China's National Emissions Trading Scheme and China's 14th Five-Year Plan, and uses ecological modernization theory and the "What is the problem represented?" approach as analytical frameworks. The results show that the EU and China share overall climate goals but differ significantly in terms of governance techniques, institutional structure and implementation strategies. The EU prioritizes legally binding regulations, stakeholder integration and market mechanisms to stimulate sustainable transitions. In contrast, China relies on a top-down, technocratic model anchored in long-term planning, government coordination and industrial upgrading. In terms of decoupling, the EU shows evidence of both relative decoupling and signs of emerging absolute decoupling. China is achieving notable improvements in carbon intensity but remains at a weak relative decoupling. However, both regions face structural and institutional constraints that limit the pace and depth of transformation. Understanding these different governance models is therefore crucial for developing strategies that integrate economic development with environmental requirements. 

In previous work, we derived new energy and entropy stable open boundary conditions and implementation techniques for linear and nonlinear initial boundary value problems. These boundary procedures result in estimates bounded by external data only. Interestingly, these new boundary conditions generalize the well-known classical characteristic boundary conditions for linear problems to the nonlinear setting. We discuss the similarities and differences between these two boundary procedures and point out the advantages with the new procedures. In particular we show that the new boundary conditions bound solutions to both linear and nonlinear initial boundary value problems and can be implemented both strongly and weakly.

Bakgrund:

Idrottsskador är vanliga i lagidrotter som handboll och innebandy. Trots att skadepreventionsträning är effektiv är följsamheten ofta låg. Spelbaserad träning har visat positiva effekter på motivation och engagemang i andra sammanhang, men hur idrottare uppfattar spelbaserad skadeprevention är inte undersökt.

Syfte:

Syftet var att undersöka innebandy- och handbollsspelares tankar och attityder till digital spelbaserad skadepreventionsträning.

Metod:

Kvalitativ intervjustudie med semistrukturerade individuella intervjuer av sju spelare (17–24 år). Transkription och induktiv kvalitativ innehållsanalys enligt Graneheim och Lundman genomfördes; meningsenheter kondenserades, kodades och sorterades i kategorier.

Resultat:

Fyra huvudkategorier framkom som belyste påverkan på individen, inställning till användning, möjliga användningssätt och upplevda för- och nackdelar. Deltagarna uppfattade spelbaserad träning som potentiellt motiverande genom tävling, flexibilitet och synlig progression, vilket kan öka följsamhet och träningsglädje. Vissa föredrog traditionell träning eller prioriterade annan träning. Möjliga användningsområden var hemträning, komplement vid uppehåll och för yngre spelare. Fördelar var ökad variation och social gemenskap; nackdelar inkluderade risk för felaktigt utförande, prestationskrav och tekniska begränsningar.

Konklusion:

Digital spelbaserad skadeprevention ansågs kunna vara ett motiverande och flexibelt komplement, men inte en ersättning, till traditionell träning. Variation i attityder betonar behovet av individanpassning, och resultaten kan vägleda utveckling av digitala skadepreventiva verktyg. 

Bakgrund: Proprioception i handleden är central för finmotorik, stabilitet och funktion. Handledens proprioception bygger på ett samspel mellan ligamentsystem, muskler och mekanoreceptorer. Kunskapen om hur proprioceptionen skiljer sig mellan dominant och icke-dominant hand samt mellan män och kvinnor hos friska individer är begränsad.  

Syfte: Undersöka skillnader i proprioceptionen, avseende ledposition- och kraftkänsla, mellan dominant och icke-dominant hand och mellan män och kvinnor vid pronation och supination i underarmen hos friska individer. 

Metod: 28 friska individer, 11 män och 17 kvinnor i åldrarna 21–63 år rekryterades. Studien hade en kvantitativ tvärsnittsdesign för att undersöka proprioception med hjälp av två mätmetoder: Baseline Wrist Dynamometer för kraftkänsla och WristCheck för ledpositionskänsla. Varje rörelseriktning testades tre gånger och medelvärdet användes vid analys.

Resultat: Det fanns inga signifikanta skillnader mellan dominant och icke-dominant hand varken för kraftkänsla eller ledpositionskänsla. Däremot påvisades signifikanta könsskillnader där kvinnor, vid supination, påvisade sämre proprioception än män både i kraftkänsla (p <0,001) och ledpositionskänsla (p <0,018).

Konklusion: I studien uppvisade friska individer en symmetrisk proprioception mellan dominant och icke-dominant hand. Könsskillnader kunde även påvisas, där kvinnor påvisade sämre proprioception än män i kraftkänsla och ledpositionskänsla vid supination. Resultaten kan ge en indikation på normala skillnader inom proprioception gällande kraftkänsla och ledpositionskänsla, samt ett explorativt underlag för framtida forskning. Större och ålderskontrollerade studier behövs för starkare evidens.

Vid Linköpings universitet (LiU) har en översyn av matematikkurserna för trecivilingenjörsprogram (Maskinteknik, Design och produktutveckling samt Energi-miljömanagement)genomförts för att se hur matematikundervisningen kan anpassas tillprogrammens specifika behov. Denna översyn motiveras av att gemensamma kurser föralla civilingenjörsprogram kan leda till genomströmningsproblem och ett kursinnehållsom blir en kompromiss snarare än anpassat för varje program.Metoden innefattade en granskning av sex gemensamma matematikkurser och 80efterföljande teknikkurser. Examinatorer för dessa teknikkurser fick notera vilka av 35specifika matematikområden som används för antingen konceptuell förståelse ellerpraktisk problemlösning. Analysen visade att vissa matematikkurser redan var välanpassade till programmens behov och därför lämnades oförändrade. Däremot visadedet sig att delar av kurserna Envariabelanalys 2 och Flervariabelanalys antingenanvändes sällan eller inte alls i efterföljande kurser, eller bara krävdes för konceptuellförståelse.Som ett resultat av översynen har två nya, anpassade matematikkurser skapats. I denya kurserna flyttas fokus till de delar av innehållet som är viktiga för praktiskproblemlösning i de efterföljande teknikkurserna. Vissa av de moment som endast ärviktiga för konceptuell förståelse examineras nu genom digitala, automaträttadeinlämningsuppgifter i stället för genom den traditionella skriftliga tentamen. Dettaupplägg syftar till att göra tentamen mer robust och låta studenterna arbeta kontinuerligtmed kursmaterialet. Ändringarna i examinationen förväntas bidra till en bättregenomströmning och det finns också förhoppningar om att den genomförda översynenkan bidra till en ökad motivation hos studenterna, då kursinnehållet ligger i linje med denmatematik som studenterna kommer att använda i efterföljande teknikkurser. Dettaarbete visar att en nära dialog mellan matematiklärare och programledning kan leda tillen mer effektiv och ändamålsenlig ingenjörsutbildning.

Every day, security analysts investigate potential breaches in organizations’ digital infrastructure. Security Information and Event Management (SIEM) systems facilitate this by collecting and analysing security data in real time. Using analytics and automation, SIEMs help analysts detect threats and filter irrelevant information. At its core, SIEM systems rely on detection rules to analyse data. When a rule identifies a suspicious event, an alert is generated, drawing the attention of an analyst. These alerts form the basis of investigations, meaning that if an adversary evades a rule, a breach may go unnoticed.

To address this, we propose leveraging a transformer model to tokenize and embed executed PowerShell commands and PowerShell SIEM detection rules, enabling computation of cosine similarity between them. This allows for identification of evasions and determining which rules were evaded. Additionally, we introduce a regressor-based metamodel that selects the most suitable Large Language Model (LLM) from a pool and provides recommendations to improve rule coverage.

Using this approach, we found that at least 24% of the analysed rules in the 2025 SIGMA dataset were evadable. Our system detected 71% of hand-crafted evasions with a False Positive Rate (FPR) below 1% and correctly attributed 98% of the detected evasions to their corresponding detection rules, advancing the state-of-the-art system AMIDES. By selecting the most suitable LLM, the metamodel recommendations successfully updated 70% of the rules to detect previously unknown evasions. Additionally, security analysts, when presented with recommendations from our LLM-based metamodelling process reported that 31% of the recommendations could replace original rules with minor adjustments, while 47% offered key insights for manual refinement.

Contemporary planetary crises have prompted a new wave of environmental and posthumanities scholarship, aiming to address our planet and its concerns as entangled and more-than-human. How might the radical propositions of the environmental and posthumanities be taught? This article argues that arts-based methods – characterised by sensation, open-endedness, interdisciplinarity, experimentalism and situatedness – offer generative approaches for embodied and embedded learning. Through three case studies, the article elaborates different arts-based approaches that the author has found especially generative when teaching the posthumanities.

Elliptic curves are the basis for Elliptic Curve Cryptography (ECC) that utilise points on a curve for its public and private keys. A special kind of addition, point addition, adds two points on the curve to get a third point on the same curve. In order for this to be cryptographically useful the addition takes place over a finite field. Elliptic Curve Cryptography has various uses, maybe most notably in the use of blockchains. ECC has the advantage of offering higher security for shorter keys compared to for example RSA. To motivate the existence of this point relation we explore algebraic curves, points and lines at infinity as well as intersections between curves.

In this thesis, local politicians’ regulation of social work practice’s legal authority to make decisions about distributing welfare is investigated, with a particular focus on elder care. Although social workers are central actors in needs assessment and service allocation, they can only exercise legal authority to make decisions if it has been delegated to them by local politicians. In a highly decentralised welfare state such as Sweden, where municipalities are legally responsible for welfare distribution but differ widely in terms of economic, demographic, and geographic conditions, it is essential to understand the local and political contexts of welfare distribution.

This thesis is positioned at the intersection of social work and political science through the analysis of how local “hard policies” – local political policy documents that determine the scope of legal decision-making authority (DMA) – shape the legal premises for social work practice to distribute elder care. By using principal-agent theory, the interdependent relationship between politicians and social workers is made visible. Local political hard policies are understood as tools for control designed to minimise risks when politicians (principals) delegate authority to social workers (agents). Moreover, the concept of accountability is employed to explain external demands on politicians, their need for control over elder-care distribution, and their need to hold civil servants accountable for welfare distribution.

The thesis is based on four studies. The first maps the extent of social workers’ legal DMA across municipalities. The second examines contrasting types of local political policies as examples of strategies for regulating social workers’ legal discretion. The third investigates the relationship between local conditions – economic, demographic, geographic, and political – and restrictiveness in delegating DMA. The fourth analyses the implications of delegating legal DMA for accountability, depending on whether it is transferred to managers or front line social workers or retained by politicians.

The findings demonstrate that local variation in the design of hard policies determining legal DMA can significantly influence whether and how social workers can legally distribute elder care. In some municipalities, social workers possess wide-ranging legal discretion, while in others their authority is extensively restricted. These differences can be linked to local conditions and politics.

Methodologically, the thesis contributes with a macro level, nationwide mapping and comparative analysis of policy documents on delegation, combined with statistical analyses of factors explaining such variations. Theoretically, it advances social work theory by applying the principal–agent framework and highlighting the distinction between soft (optional) and hard (mandatory) policies. This adds to and enhances the legal dimension of discretion and conceptualises hard policies as core instruments of political control in decentralised welfare. Empirically, it provides systematic evidence of municipal variation in the design of legal DMA and discusses the potential implications for equality in access to services, the rule of law, accountability, and the role of educated social workers in relation to local democracy.

Ultimately, the results reveal that the way delegation is structured have implications for the legal and professional discretion of social workers, as well as potential implications for transparency and accountability – vital for democratic values. Ultimately, the thesis highlights the relation between social work and local politics, while pointing toward a contradicting structure with a complex relationship between local politicians and the public administration.  

In this article, we examine how the situation for contemporary poor EU migrants is framed by social work professionals involved in initiatives funded by the Fund for European Aid to the Most Deprived in Sweden. Through 16 semi-structured interviews with these professionals, the research questions guiding the analysis concern how they describe and frame the needs of poor EU citizens living in precarious conditions, emphasizing work as a social right as well as how they navigate the EU policy guiding their work. The analysis is informed by frame analysis as analytical approach, with a focus on identifying diagnostic and prognostic framing. The findings presented highlight the professionals’ strategies in stretching EU policy frameworks to support the category of poor EU citizens in Sweden. The results point to how professionals make use of their discretion to meet the livelihood needs of poor people by claiming their right to work.

This thesis examines the challenges of integrating LGBTQ+ inclusivity into HongKong’s mental health care system, focusing on barriers to effective psychiatric services forlesbian, gay, bisexual, transgender, and queer individuals. Guided by Queer Theory andIntersectionality, the study addresses two research questions: how the current system meetsLGBTQ+ needs and the primary challenges faced by providers in delivering inclusive care.Through semi-structured interviews with two service users and one psychiatric nurse, thematicanalysis identifies four barriers: lack of cultural competence training, heteronormative normsand cultural stigma, lack of policy implementation, and resource constraints. Policy analysis ofHong Kong’s key documents, including the Mental Health Review Report (2017) and HospitalAuthority Guidelines (2019), reveals absent provisions for LGBTQ+-specific stressors, suchas stigma and family rejection. International policy references from Taiwan, Singapore,Australia, the UK, and Canada contextualize these gaps, highlighting inclusive practices likemandated training. Findings indicate that Hong Kong’s system fails to provide affirming care,exacerbating mental health disparities, while providers face inadequate training, cultural biases,legal ambiguities, and resource shortages. The study contributes theoretically by applyingQueer Theory and Intersectionality to Hong Kong’s context, empirically by filling gaps in livedexperiences and provider perspectives, and practically by informing inclusive policy reforms.

Targeting pathological corneal neovascularization arising from infection or disease is essential to preserve corneal transparency and avoid vision loss. Aniridia-associated keratopathy (AAK), a rare genetic eye condition caused by PAX6 haploinsufficiency, leads to chronic inflammation, neovascularization, and vision loss, and has limited therapeutic options. Here we evaluated Olisens®, an antisense oligonucleotide targeting insulin receptor substrate-1 (IRS-1), in a Pax6 heterozygous mouse model of AAK. Topical delivery of Olisens for 90 days significantly suppressed pathological blood and lymph vessel ingrowth into the cornea, delaying the disease course. Transcriptomic analysis revealed sustained suppression of immune responses including dendritic cell maturation, macrophage activation, and cell proliferation. Additionally, Olisens enhanced transcripts of epithelial healing factor Lars2 in the cornea and LARS2 protein expression in the peripheral epithelium and normalized peripheral corneal thickness. While vascular regrowth occurred after stopping treatment, immune pathway suppression persisted. Our results indicate targeting IRS-1 using topical Olisens reduces inflammation and neovascularization, thereby delaying AAK progression and suggesting anti-neovascular treatment as a therapeutic strategy for AAK.

A system, named Reflex, for handling reflection documents is presented. Writing a reflection document is a mandatory last step when carrying out Master’s or Bachelor’s theses within all education programs at the Faculty of Science and Engineering at Linköping University. In the reflection document the students are expected to reflect on both the thesis work and the outcome of the entire education program. The reflection is structured according to the four sections of the CDIO Syllabus plus additional aspects around the process of the thesis work and the program as a whole. Reflection documents have been used for more than ten years, but via the introduction of Reflex in 2023 the process is entirely electronic and all texts from the reflections are stored in a searchable database. This enables efficient and systematic analyses of the texts, which can be used for evaluations of entire programs as well as selected aspects. Two examples of such analyses are presented in the paper. In addition, the thesis work process is presented, together with a description of how the system Reflex works.  

With the release of 5G, society is becoming increasingly connected. The growing number of electronic devices emitting electromagnetic signals increases the risk of unintentional interference to wireless communication systems. As 5G New Radio (NR) grows beyond its commercial role to also support mission-critical services such as public safety and military communications, it places high demands on its reliability under interference. A key component in the 5G NR downlink is the Demodulation Reference Signal (DM-RS) of the Physical Broadcast Channel (PBCH), which enables accurate channel estimation and decoding. Disruption of this signal can severely degrade PBCH performance and jeopardize overall communication reliability.

This thesis investigates the robustness of 5G in the presence of different interference signals affecting the PBCH DM-RS. The study focuses on two interference types: additive white Gaussian noise (AWGN) and continuous wave (CW) signals. System performance is primarily evaluated through the bit error rate (BER) of PBCH, reflecting the overall channel performance under these conditions.

The analysis shows that CW interference generally degrades the PBCH DM-RS more than AWGN, particularly at higher E_b/N_I values. Additionally, a CW signal applied between subcarriers causes the interference to spread across multiple subcarriers, further reducing performance. Accurate estimation of key parameters via DM-RS is crucial, since errors can prevent proper PBCH operation regardless of interference conditions.

Purpose

This paper introduces the DECODE framework, a conceptual model for understanding how memetic saturation, cognitive overload, emotional exhaustion, and relational disconnection contribute to adolescent mental health challenges in digital environments. By synthesizing insights from memetic theory, cognitive-emotional regulation, and library science, the framework positions public libraries as epistemic moderators capable of fostering digital discernment, resilience, and dialogue. The study’s aim is not to test causal hypotheses but to provide an abductively derived, theory-building structure that informs future empirical work and offers actionable strategies for librarians, educators, and policymakers navigating the adolescent mental health crisis.

Design/methodology/approach

The study adopts a critical-realist, abductive methodology that iteratively integrates descriptive survey patterns, peer-reviewed empirical findings, and theoretical constructs. Publicly available datasets (CDC’s Youth Risk Behavior Surveillance System (YRBSS) 2023; Pew Research Center’s Teens, Social Media and Mental Health 2025) are used illustratively to anchor theoretical insights, rather than as sources for causal inference. Through exploratory thematic grouping and abductive triangulation, five patterns of adolescent digital experience are mapped onto six response pillars of the DECODE framework. This abductive scaffolding process yields a provisional, testable conceptual model that both explains observed trends and generates propositions for future empirical and cross-cultural validation.

Findings

Analysis of national survey patterns and peer-reviewed evidence identified five thematic patterns: Digital Immersion and Memetic Saturation, Cognitive Overload, Emotional Exhaustion, Relational Disconnection, and Emergent Risks and Adaptive Coping. These patterns correspond directly to DECODE’s six pillars: Defend and Define, Expose Memes, Cross-Check Sources, Organize Context, Document and Disseminate, and Engage Dialogue. Together, they illustrate how algorithmically curated content environments amplify adolescent distress, while libraries and civic spaces can act as epistemic moderators. The findings suggest that resilience lies not in restricting access but in cultivating cognitive and social capacities for discernment, contextualization, and dialogue in digital environments.

Research limitations/implications

This study is conceptual rather than hypothesis-testing, using descriptive survey summaries as illustrative anchors. While this approach highlights converging patterns, it cannot establish causal pathways or quantify effect sizes. The abductive design emphasizes theory generation, leaving empirical validation to future research. Cross-sectional survey data also limit temporal insights into digital saturation’s cumulative effects. Nevertheless, the DECODE framework identifies clear mechanisms and intervention points, providing testable propositions for future studies. Future research should employ longitudinal, experimental, and cross-cultural methods to evaluate and refine the framework’s propositions, ensuring both robustness and broader generalizability across diverse youth populations.

Practical implications

The DECODE framework provides actionable strategies for libraries, educators, and civic organizations seeking to support adolescent well-being in digital contexts. Each pillar translates into practical interventions: setting digital boundaries (Defend and Define), fostering media literacy (Expose Memes), teaching verification skills (Cross-Check Sources), curating contextual knowledge (Organize Context), promoting reflective practices (Document and Disseminate), and creating spaces for dialogue (Engage Dialogue). These strategies can be integrated into library programming, classroom curricula, and community initiatives, offering scalable and evidence-informed approaches to counteracting memetic saturation and cognitive overload while strengthening resilience, critical thinking, and social connectedness among young people.

Social implications

The study addresses pressing societal challenges: adolescent mental health, digital polarization, and the erosion of civic trust in a memetic media environment. By positioning libraries as epistemic moderators, the DECODE framework underscores the importance of accessible, trusted public institutions in buffering digital harms. Implementing DECODE’s pillars can strengthen community dialogue, reduce the spread of manipulative narratives, and foster healthier patterns of information consumption. More broadly, the framework offers a civic roadmap for reintroducing collective sense-making into fragmented media ecologies, promoting not only youth well-being but also more informed, resilient, and socially connected democratic societies.

Originality/value

This article is among the first to link memetic theory, cognitive-emotional regulation, and library science into a unified conceptual framework for addressing adolescent digital distress. Unlike existing models that focus narrowly on screen time or platform regulation, DECODE foregrounds the interplay of memetic saturation, cognitive overload, and relational disconnection while highlighting libraries’ overlooked civic role. Its abductive methodology generates testable propositions and actionable pillars, offering both theoretical innovation and practical pathways for intervention. The framework’s originality lies in reimagining libraries not just as information providers but as active epistemic moderators in the digital age’s mental health crisis.

Bakgrund: Subacromial smärta är ett relativt vanligt muskuloskeletalt tillstånd som orsakar smärta för individen. Träning har visat sig vara en bra behandlingsmetod för denna patientgrupp. Med en ökande befolkningsmängd ökar kraven på att vården ska behandla allt fler patienter. Därmed är digitala behandlingsalternativ relevanta för att effektivisera vården. Fysioterapeuter behöver ställa om till mer digital vård men samtidigt kunna säkerställa en säker och bra vård. 

Syfte: Syftet var att undersöka fysioterapeuters upplevelser av att behandla patienter med subacromial smärta inom primärvården enligt Axelhjälpen.

Metod: Åtta semistrukturerade intervjuer utfördes med informanter i åldrarna 28–55 år. Informanterna skulle ha behandlat minst tre patienter enligt Axelhjälpens plattform, datan analyserades sedan med en kvalitativ innehållsanalys för att komma fram till resultatet. 

Resultat: Analysen resulterade i fyra kategorier: Fysioterapeuters upplevelser av professionell autonomi i plattformen, Fysioterapeuters upplevelser av patientcentrering i plattformen, Det digitala formatet, Faktorer som påverkar plattformens resurseffektivitet. 

Konklusion: I studien upplevde fysioterapeuterna att Axelhjälpen kan vara ett verktyg som kan användas för vissa patienter för att effektivisera vården, både för vårdgivaren och patienten. Resultatet visade dock att vissa förändringar i urval av patienter och arbetssätt med plattformen skulle kunna optimeras för att plattformen ska kunna implementeras i större utsträckning. 

Bakgrund: Långvarig smärta är ett komplext tillstånd med biologiska, psykologiska och sociala dimensioner som påverkar funktion, vardag och livskvalitet. Behandling inkluderar läkemedel, multimodal rehabilitering och psykologiska metoder. Digitala interventioner och smärtskolor erbjuder flexibel, evidensbaserad vård som kan förbättra smärtupplevelse, funktion och livskvalitet, även på distans.

Syfte: Utforska patienternas upplevelser och tankar avseende den digitala smärtskolan i ett länsövergripande projekt inom primärvården, samt undersöka vilka verktyg patienterna anammat vid egen hantering av långvarig smärta.

Metod: Kvalitativ intervjustudie med genomfördes med åtta informanter med en semistrukturerad intervjuguide. Data analyserades genom kvalitativ innehållsanalys med induktiv ansats. 

Resultat: Analysen identifierade tre huvudkategorier: användarupplevelse och tillgänglighet, lärande och förbättrad egenvård, samt förväntningar och personliga perspektiv. 

Konklusion: Studien visar att den digitala smärtskolan bidrog till beteendeförändringar, förbättrad egenvård samt ökad kunskap, med stöd av avslappningsövningar, anpassad aktivitet och patientberättelser. Den digitala formen upplevdes som flexibel och tillgänglig, även om vissa saknade socialt utbyte och vissa moment kändes stressande.  Resultatet visar att informanterna i det stora hela är nöjda med vad den digitala smärtskolan har bidragit med. Både möjlighet till en mer tillgänglig form av kunskap och stor bredd i nya insikter, verktyg och inspiration. Fortsatt utveckling av interaktivitet, socialt stöd och individanpassning kan öka engagemang och nytta. 

Combinations of electric and hydraulic machines, also known as e-pumps or electro-hydraulic energy converters, are essential for the electrification of mobile working machinery. Currently, these machines are typically combined by axial stacking, and the electric machine directly drives the hydraulic machine. Alternatively, the hydraulic machine can be radially integrated within the core of the electric machine, or a gearbox in combination with a downsized electric machine can be used. However, to the authors’ knowledge, no systematic comparison of these different concepts has been published. This paper uses analytical methods to determine the dimensions of the active parts of hydraulic machines, electric machines, and gearboxes in order to compare different design concepts based on volume, aspect ratio, total mass, copper mass, magnet mass, electromagnetic efficiency, and inertia. Axially stacked concepts can yield the highest compactness. However, they achieve this compactness at low aspect ratios, with their lengths being several times greater than their outer diameters. For balanced aspect ratios, where the outer diameter and total length of the machine are similar, the radially integrated, direct-driven concept is most compact.

In Sweden, substantial volumes of municipal solid waste are currently incinerated in combined heat and power (CHP) plants, leading to the loss of potentially recyclable materials and the generation of fossil-based carbon dioxide emissions. In response, Tekniska verken in Linköping is establishing a material recovery facility (MRF) designed to extract plastics, paper, biowaste, as well as ferrous and non-ferrous metals from residual household waste for further material recycling and biogas production.This master’s thesis aimed to assess the climate implications of the MRF within the broader waste management system. A life cycle assessment (LCA) approach was applied to encompass both the energy recovery from waste incineration and the new material flows introduced by the facility. Recovered materials were modelled to displace the production of virgin materials, while biogas generated from biowaste was assumed to replace fossil fuels in heavy transport, and digestate was assumed to substitute mineral fertilizers. As the sorting process reduces the amount of waste available for incineration, imported industrial waste and biofuel are utilised as substitute fuels.Since the facility is not yet operational, key parameters such as sorting performance, recycling rates, market absorption of recovered materials, and the composition of substitute fuels remain uncertain. Scenario-based sensitivity analyses were therefore conducted. The results show that the MRF has the potential to reduce greenhouse gas emissions significantly, particularly under conditions of high material recovery efficiency and effective substitution of virgin materials. However, the climate benefits may be marginal if market outlets for recovered fractions are lacking, recovery rates are low, and industrial waste with high plastic content is used as replacement fuel.The study concludes that the climate performance of implementing a mechanical MRF is highly dependent on the degree of material separation, recycling efficiency, market conditions for secondary materials, and the properties of the substituted fuel. To realise the system’s climate mitigation potential, it is critical to achieve high recovery rates and strengthen the market demand for recycled materials.

This essay discusses the Italian renaissance poet Gaspara Stampa with focus on the negative reception her poetry received due to her gender. The discussion concerns an anonymously written poem that slander her, which is an imitation of one of Stampa's own poems. The parallels to the internet trolls of our own time are clear, as is the misogynous values behind this kind of slander. However, Stampa and other norm breaking poets, caused problems since they did not fit the neat categories of virgin/whore, but positioned in a third androgyne space between the binary configurations of feminine and masculine domains.