The study examines literacy instruction in Swedish as a second language for adult second-language learners in a labour-market-oriented context within the municipal employment administration, with a particular focus on literacy events and literacy practice (Barton, 2007). The instruction in question is led by three language development instructors, and the participants consist of a linguistically heterogeneous group with limited educational backgrounds, all of whom receive financial support from the social services.

Two data collection methods have been used: qualitative interviews and observations. The empirical material consists of three qualitative interviews and nine observations. One interview with each language development instructor has been recorded and transcribed. Furthermore, three observations with each instructor have been conducted, during which field notes were taken. A thematic analysis of the material has subsequently been carried out.

The results show that a variety of literacy events occur in the instruction, such as conversations, read-aloud activities, draw pictures and symbols, writing and reading short messages and texts, role-playing, and reading and explaining words and concepts. The analysis of the material demonstrates that oral interaction and the participants’ multilingual resources play a central role in literacy instruction and shape the literacy practices, where visual aids, bodily enactments, participants’ experiences, and multimodal tools are frequently used. The study’s results also indicate that the labour-market-oriented context influences the design of the literacy instruction and that the literacy practice is part of a complex context, which also affects the literacy events that occur. In the analysis, it also emerged that linguistic comprehension constitutes a challenge within the literacy practice, which in turn affects the literacy events, where repetitions and explanations of words, concepts, or texts are continually recurring. Furthermore, the empirical material illustrates the language development instructors’ search for adequate didactic methods to support second-language development in the literacy instruction under study.

The study concludes that a resource-oriented perspective predominates in the current literacy instruction, where the sociocultural theory’s view of communication and interaction as means for language learning also permeates and influences the design of literacy practices and the types of literacy events that occur. The results further indicate that the heterogeneous teaching group’s need for support in literacy development varies, and that individual needs are not always met. The study contributes knowledge about literacy instruction in Swedish as a second language for adult second-language learners in a labour-market-oriented context within the labour market administration.

As solar power capacity in Sweden expands rapidly, accurate forecasting of aggregated power production becomes critical for maintaining grid stability and efficient energy trading. This thesis develops and evaluates predictive models for hourly national solar power production, with a particular focus on quantifying the impact of using forecast meteorological data rather than observed data. To capture the distinct temporal dynamics of solar energy, a "Two-Stage Temporal Hierarchical" modeling strategy was implemented. This approach decomposes the forecasting problem into a stable, long-term seasonal baseline modeled on daily data, and dynamic short-term deviations modeled on hourly data. Three distinct architectures were evaluated for the hourly component: Multiple Linear Regression (MLR), Extreme Gradient Boosting(XGBoost) and Seasonal Autoregressive Integrated Moving Average Exogenous (SARIMAX). The models were trained and tested using aggregated production data from Eurelectric and high-resolution meteorological data from the MetCoOp Ensemble Prediction System (MEPS), with national weather conditions computed as a population-weighted average of five major Swedish cities. The results demonstrate that the SARIMAX, time series model, achieves the highest predictive accuracy (R² = 0.9132, RMSE = 0.1717 GWh, MAE = 0.0954 GWh), outperforming both the linear baseline (MLR) and the non-linear XGBoost model. A key finding of this study is that model performance remains remarkably robust when switching from observed to forecast weather data, with negligible degradation in accuracy, RMSE differences of less than 0.005 GWh. Feature importance analysis reveals that this stability is driven by the model’s heavy reliance on autoregressive features (lagged production values) and solar irradiance, which effectively buffer the predictions against meteorological forecast errors. These findings suggest that statistical time-series methods leveraging system inertia are highly effective for national-level solar power production prediction and that weather forecasts are of sufficient quality to support reliable grid planning operations in Sweden.

Estimating causal effects from observational data is frequently challenged by the presence of unobserved confounders. The ρ-GNF estimator addresses this challenge by using deep generative models to perform sensitivity analysis, parametrizing the latent confounding structure using a Gaussian copula. However, the reliability of this estimator when the true underlying latent distribution deviates from the Gaussian copula assumption remains a critical open question.

This thesis systematically evaluates the robustness of the ρ-GNF to misspecified latent structures. Utilizing a controlled simulation based strategy, we tested the model against data generating processes exhibiting non-linear (quadratic) and asymmetric tail-dependent (Clayton copula) latent structures. The analysis focused on whether the estimator could determine an "effective" sensitivity parameter that recovers the true Average Causal Effect (ACE) despite the distributional mismatch.

Our results demonstrate that the ρ-GNF is remarkably robust to structural misspecification. In both non-linear and tail-dependent scenarios, the model consistently yielded a specific sensitivity parameter that compensated for the bias, aligning the estimated ACE with the ground truth. Furthermore, the estimator exhibited high stability and monotonicity across repeated experiments. These findings suggest that the Gaussian copula serves as a flexible functional proxy for complex dependency structures, extending the practical applicability of the ρ-GNF beyond strictly Gaussian copula environments.

Semiconductors are the cornerstone of modern technology due to its use in com-putation as a transistor. Introducing point defects in semiconductors can comewith performance increases or new applications such as quantum computing.Finding point defects with interesting properties can, however, be demandingsince it requires either experiments or computer simulations which tend to beresource-heavy. Machine learning could potentially serve as a shortcut, needingless resources and time to identify interesting point defects. This study aimsto use machine learning for predicting the zero-phonon line and the formationenergy properties of point defects consisting of vacancies, substitutions, andinterstitials, or combinations of these, in the predominant 4H silicon carbide(4H-SiC). The ADAQ database were used to train the linear ridge model, thekernel ridge model, and the multi-layer perceptron neural model. These modelswere evaluated using mean absolute error paired with a variation of 10-foldcross-validation. Two descriptors were introduced to represent the point defects:the No Symmetry Descriptor and the Symmetry Descriptor. Both consist of anatom number representing each individual point defect, the distance betweenthe point defects, and the charge and the spin state of the defect cluster. TheSymmetry Descriptor also used the Wyckoff position of the point defect or thestacking sequence of the atom layer where it resides. Due to time constraintsthe neural model was not used together with the No Symmetry Descriptor. Thedata was separated into two parts, one consisting of vacancy and substitutionaldouble defects and the other of interstitial double defects. The mean absoluteerror averages of the zero-phonon line and formation energy predictions for thevacancies and the substitutions were 0.202 eV and 1.057 eV. These values were0.230 eV and 1.729 eV for the interstitials. These results show that machinelearning could replace high-throughput calculations of point defect properties ifits accuracy is increased, either through more data acquisition or by optimizingthe models better.

Background

A sensitive, well-functioning maternal interaction is vital for healthy infant development. For moderate to late preterm infants, this is even more important as this group of infants are at increased risk of facing neurodevelopmental disorders. The Early Collaborative Intervention supports parent-preterm infant interaction and includes three sessions, two in the hospital and one after discharge.

Objective

To investigate the impact of the Early Collaborative Intervention, compared with standard care, on mother-preterm infant interaction at one month corrected age.

Design

A longitudinal randomized controlled trial, reporting secondary outcomes from the first follow-up. SettingThe intervention was conducted at a pediatric center with two neonatal intensive care units with an infant and family centered approach. The intervention was evaluated in the homes of the families.

Participants

Families with preterm infants born in gestational week 30+0–35+6 (n = 143) were randomized. In this one-month follow up a total of 101 families participated, (standard care with the Early Collaborative Intervention, n = 60, standard care, n = 41).

Methods

The mother-infant interactive behavior was videotaped during a bath and later analyzed with Ainsworth’s Maternal Sensitivity Scales and the Emotional Availability Scales. The coder was masked to group randomization as well as to demographic data of the dyads.

Results

In the analysis the maternal mean scores were statistically significantly higher for the intervention-group versus the standard care group in the Availability subscale, 7.30 vs 6.29 (CI 0.01–0.86, p = 0.045, Cohen’s d 0.43), and Acceptance subscale, 8.00 vs 7.22 (CI 0.12–0.97, p = 0.012, Cohen’s d 0.55), in the Ainsworth’s Maternal Sensitivity Scales. Mean score were also statistically significantly higher for the intervention-group versus the standard care group in the Non-hostility subscale, 6.60 vs 6.12 (CI 0.11–0.97, p = 0.013, Cohen’s d 0.54), in the Emotional Availability Scales. The results suggest that these aspects of maternal interactive behavior towards her infant, are the ones most influenced by the Early Collaborative Intervention.

Conclusions

The Early Collaborative Intervention had beneficial impacts on maternal interactive behavior for those who took part in three sessions or more of the intervention program.

Registration

The project was registered in ClinicalTrials.gov with the number: NCT02034617, registered 19/12/2013, date of the first recruitment 15/01/2014.

Den här studien handlar om hur specialpedagoger i förskolan utför sitt uppdrag i praktiken, det vill säga vilka arbetsuppgifter de har och hur dessa arbetsuppgifter tar sig uttryck i yrkesvardagen. Studien har en kvalitativ ansats och datainsamlingen omfattar skuggning, intervjuer och dokumentinsamling. Analysmetoden som används är tematisk analys. Studiens teoretiska utgångspunkt är systemteori och Bronfenbrenners bioekologiska modell, kompletterad med specialpedagogiska perspektiv. I likhet med tidigare forskning visar resultaten att specialpedagogens uppdrag i förskolan är otydligt definierat. Denna studie bidrar med nya insikter om hur specialpedagogerna hanterar denna otydlighet, bland annat genom att göra egna prioriteringar och utveckla egna arbetsstrukturer och strategier för hållbarhet. Det framkommer också att specialpedagogens arbete präglas av lärande, både i relation till egen professionell utveckling och i arbetet med att stödja andras lärande. Även uppgifter relaterade till kvalitetsutveckling samt samverkan med andra yrkesgrupper är återkommande. Ett särskilt intressant resultat är att en betydande del av yrkesutövningen består av möjliggörande aktiviteter såsom relationsskapande, reflektion och organisering av arbetstid. Resultaten synliggör därmed behovet av att dessa aktiviteter ges organisatoriskt utrymme och erkänns som centrala delar av uppdraget. Detta framstår som avgörande för att specialpedagogen ska kunna fungera som en sammanlänkande funktion mellan olika systemnivåer och bidra till ett sammanhållet främjande, förebyggande och åtgärdande arbete för ökad inkludering i förskolans praktik. Studien bidrar med fördjupad kunskap om hur specialpedagogens uppdrag tar form i förskolans verksamhet. Resultaten kan därmed utgöra ett stöd i arbetet med att utveckla mer preciserade och ändamålsenliga arbetsbeskrivningar för yrkesrollen samt för att skapa organisatoriska förutsättningar för de främjande och förebyggande arbetsuppgifterna.

To better understand the consumer behaviours that drive demand for fast deliveries, this paper explores two key temporal aspects of the consumer journey: urgency and impatience. Urgency reflects a need for quick delivery, while impatience stems from a desire to receive products as soon as possible. Urgency and impatience merit attention to further understand consumer behaviour in relation to fast last-mile deliveries, as these pose challenges for logistics efficiency and environmental sustainability, particularly in the last-mile. The purpose of this paper is to explore urgency and impatience in the last-mile consumer journey, and the influence of these two concepts on the speed of last-mile deliveries. Additionally, the paper introduces the potential effects of urgency and impatience on environmental sustainability of such deliveries. The study uses diary entries from 15 Swedish e-commerce consumers over one month, followed by semi-structured interviews. Findings reveal that both urgency and impatience appear to be important denominators of consumer behaviour in the last-mile consumer journey. Further, urgency and impatience can have a direct impact on consumer choices regarding last-mile delivery method, including speed of delivery and reception of delivered orders. The paper also gives an increased understanding of how urgency and impatience can have implications for more environmentally sustainable last-mile deliveries. By taking a consumer-centric perspective on last-mile deliveries, this paper joins the recent academic conversation focusing on an increased understanding of the consumer to improve logistics operations.

Sustainable development depends not only on economic, social,and environmental objectives, but on the systemic capacity of societies to anticipate uncertainty, absorb shocks, and adapt through purposeful reform. This dissertation advances a unifying conceptual framework,the dynamic triad of readiness, risk, and reform, to explain how systemic capacity contributes to sustainable outcomes, how its effectiveness is conditioned by uncertainty, and how it is reshaped through reform across levels of analysis. Readiness denotes the accumulated institutional, technological, financial, and social capacities that enable anticipation and response. Risk captures multidimensional uncertainty and vulnerability that condition their effectiveness. Reform represents structural interventions triggered by revealed weaknesses. Their interaction forms a continuous feedback loop shaping resilience and long-term sustainability outcomes.

Adopting a multi-scalar perspective, the dissertation examines these dynamics at the micro (firm), meso (industry and market), and macro (national) levels across five empirical studies. Using diverse econometric approaches, including survival analysis, regimeswitching VAR models, threshold regressions, triple-difference designs, mediation and moderation analysis, and panel Granger causality tests; the research investigates renewable energy innovation,climate finance and clean water investment, firm survival during COVID-19, and post-crisis safety reforms in global value chains.

The findings demonstrate that readiness is multidimensional and context dependent. At the firm level, innovation capabilities significantly increase survival probabilities during systemic shocks. At the market level, climate policy uncertainty exhibits nonlinear and regime-dependent effects: risk constrains investment under high volatility but enables strategic positioning under stable conditions. At the national level, systemic readiness promotes renewable energy innovation by reducing socio-economic vulnerability and strengthening R&D investment. These effects are strongest among innovation leaders, with R&D serving as the dominant channel in middle-income countries. However, these benefits are regime dependent: overall, economic, and social readiness promote renewable energy innovation under low uncertainty, but their positive influence weakens beyond the uncertainty threshold. In contrast, governance readiness exerts a negative effect that intensifies as uncertainty increases. Bidirectional dynamics between core and enabling technologies confirm their co-evolution within national innovation systems and highlight the risk of widening global innovation disparities if readiness accumulation remains uneven. Crisis-induced reforms can strengthen institutional capacity but may also impose adjustment burdens that reshape market entry and dynamics.

Overall, the dissertation shows that sustainable development hinges on proactive capacity building, strategic risk management, and carefully designed reform processes. By integrating insights from institutional economics, sustainability transitions, and financial instability theory; the study contributes an analytical framework for understanding resilience in an increasingly uncertain world.

Water investments play an increasingly important role in sustainable finance, yet their response to climate policy uncertainty (CPU) under different market conditions remains poorly understood. This study examines the regime-dependent influence of CPU on water equity performance using monthly data for the First Trust Water ETF (FIW) and the Invesco Global Water ETF (PIO) from 2007 to 2024. A Markov regime-switching VAR framework is employed to capture nonlinear dynamics that conventional linear models may overlook. The results reveal two distinct volatility regimes with contrasting CPU effects. In low-volatility periods, CPU is associated with higher returns, indicating that climate-policy developments can signal investment opportunities when markets are stable. During high-volatility periods, CPU exerts a negative influence, consistent with rising discount rates applied to long-term water-infrastructure cash flows. Regime persistence differs across ETFs: FIW exhibits frequent, short-lived transitions, whereas PIO displays more persistent states. A complementary DCC-GARCH analysis shows that crude oil provides a relatively cost-effective hedge for water portfolios, while technology ETFs offer substantially weaker hedging performance. Overall, the findings highlight the importance of regime-sensitive portfolio strategies for investors and emphasize that policymakers should consider prevailing market conditions when communicating climate initiatives. The study demonstrates that nonlinear models are essential for uncovering climate-finance linkages that linear approaches fail to detect. 

Facial recognition systems (FRS) in forensic contexts require rigorous proficiency testing, but collecting and sharing real facial images faces increasing privacy and ethical constraints. This thesis explores synthetic facial image generation as a privacy-preserving alternative, investigating whether multiple images preserving a consistent synthetic identity can be generated across varied contexts.

A generation pipeline using Stable Diffusion XL with IP-Adapter FaceID was implemented to generate identity-consistent synthetic facial image sets. The approach was evaluated on 100 synthetic identities, each with one reference and five identity-conditioned images. Identity consistency was assessed through embedding-based metrics and human evaluation by forensic practitioners. Results showed mean identity distances of 0.389 ± 0.039 between reference and conditioned images, with verification performance achieving true accept rates of 95.60 %, 83.40 %, and 75.80 % at false accept rates of 0.10 %, 0.01 %, and 0.001 % respectively.

Perceptual realism assessment using FID scores indicated moderate similarity to real facial images, while forensic expert review revealed varying quality and identity preservation across image pairs. The findings demonstrate that embedding-based identity conditioning can generate synthetic facial image sets suitable for automated verification, while highlighting remaining limitations in fine-grained identity preservation and perceptual realism that require consideration for forensic proficiency testing applications. 

Optimal classical planning aims to find minimal-cost action sequences in deterministic,fully observable problems. Strong domain-independent admissible heuristics are crucial for optimal A* searches, and cost partitioning is one of the most powerful techniques for generating them. However, computing optimal cost partitions is challenging because it requires solving a linear program for each encountered state. There exist two strategies to make cost partitioning more practical. First there exist non-optimal cost partitioning strategies that offer weaker guidance but are easier to compute. Second, cost partitioning heuristics are often approximated by not computing one for every encountered state.

This thesis investigates how efficiently the non-approximative, say perfect, versions of non-optimal cost partitioning strategies can be computed, which helps to understand their true practical complexity and limits. We show that the perfect variant of two common cost partitioning strategies, saturated cost partitioning (SCP) and saturated post-hoc optimization (SPhO), can be computed much more efficiently than the naive strategy.

SPhO computes cost partitions by solving a linear program for each state encountered during the search, which is prohibitively expensive in practice. We introduce cover rules based on the sensitivity analysis of linear programs that enable the reuse of cost partitions across states without sacrificing heuristic quality. This drastically reduces the number of linear program solver calls while preserving optimality. We also analyze the structure of the saturated post-hoc optimization linear program, including degeneracy and non-uniqueness, and propose an algorithm to maximize cost partition reusability.

SCP does not require a linear program to be solved and is cheaper to compute. However,its quality depends heavily on the ordering of component heuristics. The best SCP heuristic would maximize over all possible orders, but this is infeasible due to factorial growth. We address this issue by representing SCP collections as term graphs,which allows for the identification and elimination of trivial redundancies. By formalizing conditions for equivalent orders, we can eliminate non-trivial redundancies and reduce the graph size by several orders of magnitude. This enables the first computation of the perfect SCP heuristic and the first comparison of existing SCP approaches with their theoretical limit.

Drones are a promising alternative to use instead of trucks in last-mile deliveries of small packages in urban environments due to their increased mobility and lower greenhouse emissions. To explore the usage of drones to deliver small packages in a 3D environment, an any-time solver for the problem was developed. The solver uses the metaheuristic adaptive large neighbourhood search to iteratively attempt to improve an existing solution that was found using prioritised planning on a scenario problem. To find delivery routes in environments of up to a size of 350 times 350 times 30 meters that is filled with randomly generated rectangular buildings, two search algorithms, Lazy-Theta* and A* with a simple drone collision avoidance algorithm were used. While considering a drone’s flight speed, current battery, carrying capacity, desired minimum distance to other drones, and delivery deadlines for each delivery, it was found that solutions in parts of fictional cities were possible to achieve and that the best performing heuristic utilised a delivery route’s delay and cost, leading to the highest improvement rate (defined as the percentage of iterations that improved the current solution) of 38.46% among all scenarios. The use of multiple heuristics performed better than using single heuristics even though the greatest improvement rate of a single heuristic is 38.10% and was achieved by the Delay heuristic in a larger scenario. It was noted that seeing improvements in a solution became more common when a scenario contained more deliveries and when the number of re-planned deliveries in each iteration was roughly equal to eight. Unlike the improvement rates, the overall improvement of a solution’s cost was small and often none regardless of scenario and chosen heuristics. In addition, the solver has performance issues which meant that up to 1800 iterations could be evaluated on small scenarios and environments with a budget of up to four minutes and in larger scenarios, a budget of up to 16 minutes could evaluate as few as 20 iterations. The results show that it is possible to use drones to deliver packages in a 3D environment but that more work is needed to obtain better solutions.

Beyond Gravity develops separation systems for launch vehicles and space crafts. One type of separation system is the Payload Adapter System (PAS), it is used to separate space crafts from launch vehicles. The two stages the separation system holds together must be able to transmit data between each other, this is done by cables and connectors. The structure on the PAS that holds these connectors is called the umbilical tower, this structure allows for movement of the connectorsto prevent them from taking damage from shocks that occur during the launch sequence. This movement ability is called float.

The aim of the project was to redesign the umbilical tower of the PAS to increase its floating capabilities, as well as optimizing the load path and material selection for the construction.

An acclaimed methodology for concept development and product design was used to guide the project forward in a structured fashion. The current design was analysed from which needs and requirements were established. This led to the development of new concepts which were ranked based on the stated requirements. Some concepts were similar to today’s design while others weremore experimental and unconventional. This approach was used in order to explore alternative perspectives and gain fresh insight into the design process. The exact dimension, overall setup and material used were determined through calculations to ensure the design’s strength and endurance.

A few different solutions were examined, but the resulting concept consisted of an interface with larger dimensions and an improved setup of washers in a material better suited for the function. The results this project presents should be regarded as an initial step toward the continued development of the umbilical tower.

The first and foremost aim with this book is to serve as a guide and reference for engineers that want to use surrogate modeling techniques when they design or develop products, systems and components.

Additionally, the book constitutes part of the literature for the courses in Design Optimization and Multidisciplinary Design Optimization at Linköping University.

It is the intention of the author that the content should serve as an easily understood introduction to surrogate modeling. The more advanced details are briefy mentioned but can be found in the scientifc papers referenced in this book.

There exist numerous books and review articles regarding surrogate modeling, e.g. Forrester, Sóbester, and Keane, 2008 and Krawczyk and Arabas, 2025, which means that all methods presented here can be found elsewhere. The two main advantages of this book are:

• Many methods are exemplifed using numerical examples
• Some practical guidelines recommended by the author are presented

Computed tomography (CT) is widely used to evaluate patients with suspected coronary artery disease (CAD). Coronary artery calcifications (CAC) scoring, of which Agatston score is the most widely used, has been utilized since the 1990s for cardiovascular risk estimation in asymptomatic patients. An AS of 0 strongly correlates with a lack of cardiovascular events over the following 5 years. In stenosis grading, conventional coronary CT angiography (CCTA) has high negative predictive value (NPV) but may overestimate stenosis severity due to limited spatial resolution and blooming artifacts. Photon-counting detector CT (PCD-CT) offers higher spatial resolution, reduced noise, and spectral imaging capabilities that enable virtual monoenergetic images (VMI), virtual non-contrast (VNC), and virtual non-calcium reconstructions, potentially improving diagnostic accuracy.

This thesis aimed to evaluate the reliability, advantages, and limitations of PCD-CT compared with conventional energy-integrating detector CT (EID-CT) in the evaluation of coronary artery disease. In study I and II the correlation and agreement of Agatston score (AS), was evaluated between the two CT systems. In study II and IV the image quality was compared between the two CT systems. Study III investigated how reconstruction parameters affect different plaque component quantification with PCD-CT using ultra high-resolution mode. In study IV the measured grade of stenosis was compared between EID-CT, PCD-CT, using standard resolution mode (SR), with invasive coronary angiography as golden standard.

Study I used data of cadaveric hearts, study II and IV, were based on clinical patient data, while study III used data of SCAPIS (Swedish CArdioPulmonary bioImage Study) study subjects.

The first study showed that the AS, measured on cadaveric hearts, correlated excellently between PCD-CT and EID-CT with high reproducibility, confirming the comparability of the two methods for calcium quantification. The study of AS included in study II confirmed this result. The second study demonstrated that both systems provide high image quality in CCTAs, but that PCD-CT offers sharper calcium delineation as well as lower radiation dose and contrast volume with preserved image quality. The third study showed that in CCTA, reconstruction parameters influence the measured volume of coronary plaques. The fourth study showed that PCD-CT, when using a standard-resolution scan mode, provides diagnostic accuracy equivalent to EID-CT for coronary stenosis assessment, but with superior image quality.

In summary, the thesis demonstrates that PCD-CT provides image quality and quantitative diagnostic results that match or surpass those of conventional CT in the diagnosis of CAD, with potential for improved image sharpness, dose reduction, and more reliable plaque characterization. The studies also highlight the need for careful parameter optimization in CCTA acquisition and reconstruction protocols to improve CAD evaluation as well as segmentation-based plaque analysis tools in PCD-CT.

Background: Coronary CT angiography (CCTA) is a key non-invasive tool for evaluating coronary artery disease (CAD). While energy-integrating detector CT (EID-CT) offers high negative predictive value (NPV), its positive predictive value (PPV) is limited in heavily calcified vessels. Photon-counting detector CT (PCD-CT), with higher spatial resolution and reduced blooming, may enhance diagnostic performance. Current PCD-CT systems provide both standard-resolution (SR) and ultra-high-resolution (UHR) modes, but the clinical impact of these modes remains under investigation.

Objectives: To compare the diagnostic accuracy and image quality of SR-PCD-CT versus EID-CT in quantifying coronary stenosis, using quantitative coronary angiography (QCA) as reference. Materials and methods: In this prospective, single-centre study, 21 patients (5 women, mean age 71.5 years) with suspected CAD underwent CCTA with both EID-CT and SR-PCD-CT prior to QCA. A total of 301 coronary segments were assessed for stenosis severity, with ≥50 % stenosis deemed significant. Image quality was graded using a 5-point scale. 

Results: No significant differences in percentage diameter stenosis (%DS) were found between imaging techniques (p = 0.20). Both EID-CT and SR-PCD-CT showed good agreement with QCA (AUC: PCD-CT 0.89, EID-CT 0.86). Specificity and NPV were high for both; sensitivity and PPV were moderate. SR-PCD-CT yielded higher image quality compared to EID-CT (p < 0.001).

Conclusions: In standard resolution mode, PCD-CT offers excellent image quality for quantifying coronary stenosis at comparable diagnostic accuracy compared to EID-CT

This study explores how Swedish children aged ten to twelve perceive their neighbourhoods as spaces for meaningful leisure. Drawing on group interviews with 50 children residing in areas with varying characteristics, the research applies the concepts of ‘children’s places’ and ‘places for children’ to analyse how children navigate and interpret their local environments. While many children expressed satisfaction with nearby facilities, such as playgrounds and sports fields, they also highlighted limitations in variety, maintenance, and age-appropriate challenges. The analysis reveals how children’s perceptions of their neighbourhoods’ range of leisure activities spans beyond the actual supply of leisure places. Access to these leisure places is affected by different sets of normative perceptions of (a) availability for the specific age group, (b) availability and maintenance and (c) availability and proximity. This study offers new insights into the everyday lives of Swedish ten-to-twelve-year-olds and their everyday lives. It builds on previous research on children’s and young teenager’s microgeographies, by highlighting the importance and hindrance of children’s sense of availability and access to the range of activities offered by the community. The contribution of this paper is mainly the insights of the complexity of this particular age group and how being in the middle years and the increased awareness that comes with that is affecting children’s microgeographies beyond what is there.

Syftet med denna kvalitativa studie är att fördjupa kunskapen om hur förskollärare använder högläsning som pedagogiskt verktyg i förskolans verksamhet samt hur högläsning integreras i det pedagogiska uppdraget. Studien tar sin utgångspunkt i ett sociokulturellt perspektiv. Det empiriska materialet består av semistrukturerade intervjuer med fyra förskollärare verksamma på olika förskolor i olika kommuner. Materialet har analyserats med hjälp av kvalitativ innehållsanalys. Resultaten visar att högläsning i förskolan genomförs på olika sätt beroende på sammanhang och förutsättningar. Förskollärarna beskriver högläsning både som en planerad undervisningsaktivitet och som ett spontant inslag i vardagen, där graden av struktur, dialog och barns delaktighet varierar.

Högläsningens pedagogiska kvalitet är i stor utsträckning beroende av förskollärarens aktiva och medvetna deltagande i lässituationen. Dialogiska inslag i form av interaktiva samtal anses vara särskilt betydelsefulla för barns förståelse och engagemang. Samtidigt påverkas genomförandet av högläsning av organisatoriska och relationella villkor, såsom stora barngrupper, personalbrist och otrygghet i barngruppen. Dessa faktorer begränsar både omfattningen och kvaliteten i högläsningen och synliggör en spänning mellan pedagogiska intentioner och förskolans vardagliga förutsättningar.

In their professional role, teachers and teacher assistants in the Swedish adapted school encounter many students with diverse abilities. In addition to teaching students with intellectual disabilities (ID) in academic subjects and social skills, they must also create conditions that enable the students to develop an awareness of their own and others’ circumstances in a democratic society. This awareness may become significant for the students’ future participation in work and other activities as adults. Research on students’ own experiences of education as individuals with disabilities, and the special educational support they have received throughout their schooling, is limited. During the quality dialogue in spring 2025 in a medium-sized Swedish municipality, it emerged that upper secondary students show low awareness of their disabilities and limited independence. Therefore, this study aims to shed light on students’ awareness and understanding of their diagnosis, its impact on everyday life, and its significance for identity formation. The study was delimited to the diagnosis of intellectual disability (ID). Six interviews with students in adapted upper secondary school were performed and analyzed using thematic analysis, resulting in five themes: “Basically normal ordinary people” – reflections on identity after receiving an ID diagnosis; “Who would we be without the diagnosis?” – perceived strengths linked to ID; “Living as usual with ID” – the impact of being different on everyday life; “Not part of the ordinary” – the feeling of exclusion; and “Insightful, willing to change, and seeing their potential” – ideas for future improvements. The results are discussed based on Nilholm’s three special education perspectives (compensatory-, critical-, and dilemma perspective) and related to previous research. The study indicates that prevailing norms and values need to be made visible and problematized to enable change. Students who are met with conscious and goal-oriented strategies show greater insight and increased awareness, as well as the ability to contribute constructive solutions. To create equal living conditions, adaptations must be based on individual experiences while preserving students’ choices and social participation.

Examensarbetets syfte är att undersöka vilka erfarenheter speciallärare, lärare och rektorer har gällande arbetet med övergången mellan förskola och anpassad grundskola. Den frågeställning som besvaras är: Vilka är speciallärarnas, lärarnas och rektorernas erfarenheter av arbetet med övergången mellan förskola och anpassad grundskola? Övergång mellan skolformer är avgörande händelser i ett barns liv. Övergången innebär bland annat att barnet ska möta nya miljöer, skapa nya relationer och erfara nya sätt att lära. I Sverige är förskolan inkluderande vilket innebär att alla barn, med eller utan en intellektuell funktionsnedsättning eller annan svårighet, går i samma förskola. I examensarbetet har en kvalitativ metod med intervjueranvänts och analysverktyget som har tillämpats är tematisk analys. För att besvara frågeställningen har intervjuer med speciallärare, lärare och rektorer på anpassade grundskolor i fem kommuner genomförts. Det teoretiska perspektivet som examensarbetet tar sin utgångspunkt i är Bronfenbrenners bioekologiska utvecklingsmodell. Resultatet visar att övergången från förskola till anpassad grundskola för barn med en intellektuell funktionsnedsättning ser olika ut i kommunerna. De teman som framkom var arbetsgång med egna lösningar, överlämningar i flera steg, iordningställa miljöer och trygga vårdnadshavare ger trygga elever. Förskoleklassen finns inte i skolformen anpassad grundskola, någrakommuner har därmed startat olika typer av resursskolor som fungerar som förskoleklasserinom den anpassade grundskolan. Exempel på rutiner och strategier som informanterna (lärarna och rektorerna) upplever som framgångsrika vad gäller övergången från förskola till anpassad grundskola är igenkänning, relationer och att låta det ta tid. En annan framgångsfaktor som informanterna lyfter är en ömsesidig kommunikation med vårdnadshavarna. Detta får till följd attspeciallärare och lärare behöver lägga tid, energi och fokus på relationsbyggande. Ett förslag till vidare forskning är att undersöka vilka övergångsaktiviteter som gynnar elever med en intellektuell funktionsnedsättning mest.

This thesis introduces Fidelity Invariance Curriculum Learning (FICL), a general reinforcement learning framework that integrates curriculum learning with multi-fidelity optimisation to reduce training costs and improve policy generalisation. Building on Teacher-Student Curriculum Learning (TSCL), FICL incorporates cost-aware task sampling and leverages both low- and high-fidelity environments. To evaluate the approach, we construct a quad-rotor drone navigation scenario where an agent must locate and reach a target using GPS-like, approximate coordinates and visual input. FICL is compared against two baselines: high-fidelity TSCL and high-fidelity non-curriculum learning. Comparisons use convergence cost, episodic return, and episode termination metrics. Experimental results show that FICL achieves faster convergence in wall-clock time and maintains policy performance compared to baselines, while also contributing to more stable training dynamics, demonstrating improved sample efficiency without compromising solution quality. Results suggest that combining curriculum learning with fidelity-aware task sampling provides a promising approach for scalable reinforcement learning in resource-constrained domains, such as autonomous navigation.

Rapid urbanisation and population growth have increased the demand for glass as a building material, but its environmental friendliness is questioned partly due to the lack of recycling. The production of flat glass is energy-intensive and contributes to significant greenhouse gas emissions, while the raw materials used are finite resources. This research article explores the challenges and opportunities for recycling flat glass in the construction sector in Sweden. Despite its ability to be endlessly recycled without degradation, flat glass is rarely recycled and often ends up in landfills. Barriers to circular handling of flat glass include economic imbalances, high transportation costs and lack of collection and sorting facilities. The research was conducted using a qualitative case study approach, including a workshop and semi-structured interviews, to identify registers of valuing flat glass recycling. The results highlight the different roles of glass, its economic and ecological values and the tensions between different registers of valuing glass. Understanding these values and tensions can contribute to more precise measures for integrating flat glass into a circular economy and valorisation of post-consumer flat glass.

The 2030 Agenda and the Paris Climate Agreement are the result of international negotiations and political compromises. In addition to establishing epoch-making goals, they also set the direction for how these goals should be achieved. Together, the agreements emphasise the importance of stewardship, good governance, and policy coherence in achieving sustainable development. This thesis is inspired by the increased calls for policy coherence to advance implementation of the 2030 Agenda and the Paris Agreement, as progress is hampered by tensions and goal conflicts. Although progress has been made in understanding policy coherence across policy areas, less is known regarding policy actors’ ideas of policy coherence and how they engage with the concept in on-going policy processes. The aim of this thesis is to critically examine the role of policy coherence for solving sustainability goal conflicts. I operationalise this aim through two is research questions: (1) how do policy actors engage with policy coherence? And: (2) to what extent can efforts to achieve policy coherence solve sustainability goal conflicts?

To answer these research questions, I study policy coherence through an interpretivist lens. I examine policy coherence as a process as well as an outcome, and as both ideational and institutional. The thesis consists of four individual papers which explore the multi-level governance dynamics and strategic challenges of implementing the SDGs in the context of goal conflicts. These papers examine how national debates shape policy, how implementation unfolds across national, regional, and local levels, how local governments integrate SDGs into strategic planning, and how goal conflicts are navigated across diverse country contexts in Australia, Colombia, Fiji, and Sweden.

This thesis shows how actors’ ideas of coherence shape what goes onto the political agenda, what is considered a legitimate policy response, and which interventions are prioritised. The results show that efforts to achieve policy coherence can have different impact depending on the nature of the goal conflict and when in the policy process the efforts are undertaken.

Strategic formulations of synergistic and coherent goals do not necessarily result in coherence in implementation. Instead, I show that policy coherence should be considered an active process of reconciling ideas with practical activities related to policy implementation. Clearer mandates and responsibility across administrative levels can mitigate systemic issues of displacing goal conflicts, both spatially and temporally. To conclude, this thesis shows the power of policy actors’ ideas on what is considered coherent, and how these ideas can end up displacing important sustainability goal conflicts, pushing climate priorities off the agenda. Thus, future research should be directed towards how to understand, examine and solve sustainability goal conflicts of the 2030 Agenda – in policy, politics and planning. Further research should also address which efforts to achieve policy coherence are useful under which circumstances, instead of describing policy coherence as a catch-all-term to achieve sustainable development.

Network visualization plays a crucial role in understanding complex relationships across different domains, from social interactions to biological networks. However, as networks grow in size and complexity, traditional visualization approaches struggle to support the simultaneous analysis of multiple measures and temporal evolution. This thesis addresses these challenges by reimplementing and extending ViNCent, a specialized tool for centrality focused network analysis, as a modern web-based system built on a client-server architecture with a TypeScript/React/D3.js frontend and Python/FastAPI/NetworkX backend, supporting multivariate and temporal network exploration and analysis. ViNCent 2.0 preserves the original system's core analytical capabilities while introducing significant extensions. The circular diagram view extends the original centrality focused encoding to support arbitrary node attributes alongside centrality measures, enabling simultaneous comparison of multiple centrality metrics and domain-specific metadata within a unified visual representation. For temporal networks, the system implements a discrete snapshot model with foresighted layouts, chronological node ordering and explicit change indicators to support temporal exploration while maintaining spatial consistency to reduce cognitive load during navigation. Evaluation through a formative user study and structured use case scenario demonstrates that ViNCent 2.0 supports exploratory analysis of multivariate and temporal networks. The user study identified concrete usability improvements that informed design refinements, while the Delhi Metro network use case illustrates how the system enables analysts to track network evolution, identify structural hubs, and relate topological importance to domain specific metadata through coordinated views and interactive filtering. However, to what extent the system supports effective analysis for domain experts remains an open question left for future work. The work contributes a functional web-based platform for network visualization that addresses the specific challenges of comparing multiple measures simultaneously and exploring temporal dynamics. By modernizing ViNCent's centrality focused approach and extending it to support broader analytical requirements, this thesis demonstrates how specialized visualization techniques can be adapted to contemporary web technologies while expanding their applicability to increasingly complex network analysis tasks.

Samhällskunskapslärare möter ständigt kontroversiella frågor i sin undervisning. Denna litteraturstudie undersöker detta ämne utifrån ett antal frågor. För det första: vilka frågor som lärare uppfattar som kontroversiella, hur undervisningen av dessa effektivt kan gå tillväga, och hur lärare kan använda dessa frågor i fostrandet av elever till aktiva demokratiska medborgare. Utifrån åtta artiklar som undersöker lärarperspektivet på ämnet, har dessa analyserats genom en tematisk analys. På resultatet appliceras en egenkonstruerad modell som bygger på teorier från Flensner och Stradling om olika förhållningssätt lärare använder i undervisning av kontroversiella frågor. Resultatet visar att lärare uppfattar frågor som kontroversiella baserat på om eleverna själva ser de som kontroversiella. Det visar även att lärare använder olika strategier i undervisning av frågorna där vissa väljer att undvika de helt, medan andra använder strategier som betonar det kontroversiella. Bland dessa lärare framhålls ofta representation för olika perspektiv som en strategi. Slutligen skildrar resultatet att många lärare ser demokratiuppdraget som ett av de främsta skälen till att undervisa om kontroversiella frågor. Hur detta kan uppnås beskrivs vara genom att lära eleverna kritisk granskning av andras och sina egna perspektiv för att sedan kunna ta ett mer medvetet ställningstagande.

This thesis investigates how meat grown in a laboratory, known as cultured meat, is presented by three European start-ups that as of 2025 are awaiting approval to launch their cultured beef and pork products on the consumer market. An analysis of the digital marketing of MosaMeat, Uncommon Bio and Re:meat concludes that their products are presented as real meat, with an emphasis on technological production methods, and as compatible with regional dietary norms. The target audience is concluded to be younger, whiter, omnivorous adults who wish to consume meat products as part of a caring and environmentally conscious lifestyle.

Schema validation is a critical component of graph data pipelines, ensuring data quality and consistency as graphs evolve and scale. In property graph databases, validation is commonly performed as a pre-validation step prior to data ingestion, which can limit scalability and increase pipeline complexity. This thesis investigates whether in-database, query-driven validation can improve scalability and maintainability while ensuring that production graphs remain schema-compliant.The work first presents a systematic catalog of schema constraints commonly found in property graph schemas, including property-level, structural and contextual constraints. These schema elements are then mapped to reusable Cypher query templates that identify violations directly within Neo4j, demonstrating that Cypher is sufficiently expressive to serve as a general-purpose validation language for property graphs beyond the scope of native database constraints.To evaluate the practical implications of this approach, a prototype query-driven validation tool is implemented and integrated into an industrial Neo4j data pipeline. An empirical evaluation compares query-driven validation with a traditional pre-validation (check-first-then-insert) strategy across multiple dataset scales. The results show that query-driven validation achieves significantly higher throughput and lower end-to-end processing time for large datasets, with performance gains increasing as dataset size grows. While pre-validation prevents invalid data from being ingested, query-driven validation introduces a trade-off by allowing temporary violations, shifting responsibility for remediation to the application layer.Overall, the findings indicate that query-driven validation can substantially improve the scalability of graph data ingestion pipelines while maintaining expressive and flexible schema enforcement. The thesis concludes by discussing limitations related to data model dependency, update handling and invalid data management while outlining directions for future work such as hybrid validation strategies and automated remediation of detected violations.

Battery health assessment is essential for safe and reliable operation of lithium-ion bat-tery systems. This thesis investigates battery degradation characterization through electro-chemical impedance spectroscopy (EIS) and equivalent circuit modeling across multipleoperating conditions.Multiple battery cells spanning 22 battery models from various manufacturers wereanalyzed, including different chemistries, size (18650 and 21700), and ageing states rangingfrom new cells to aged cells. EIS measurements were conducted at multiple state-of-chargelevels and temperatures under different application profiles.The 3RC-Warburg-RL equivalent circuit model was created for parameter extractionbased on comprehensive model comparison. This model demonstrates good fitting perfor-mance with R2 values exceeding 0.99 across all test conditions. Ten model parameters wereextracted to quantify impedance variations.Results show systematic parameter evolution with ageing. The largest time constantτ1 exhibits the most dramatic increases (656-7760%). Ohmic resistance R0 increases con-sistently as battery cell ages, though the magnitude varies between cell models. The War-burg coefficient σ increases for all cell types, indicating progressive deterioration of dif-fusion processes. Temperature analysis reveals strong parameter sensitivity to operatingconditions. Impedance increases 3-10 times at -20°C compared to 25°C for most parame-ters. Silicon anode cells show exceptional temperature sensitivity, with impedance valuessignificantly exceeding other cell types at low temperatures. SOC analysis demonstratescell-dependent parameter variations, with aged cells showing more pronounced SOC de-pendencies than new cells.The thesis establishes that EIS-based parameter extraction effectively quantifies batterydegradation across multiple dimensions. The 3RC-Warburg-RL model provides a goodframework for characterizing impedance variations due to ageing, temperature, and SOCchanges. These findings enable development of improved battery health assessment frame-works that account for operating condition dependencies.

This dissertation advances two complementary aims in the study of large language models: (i) understanding their inner workings and (ii) improving their training and evaluation. It does so through three lines of inquiry: integrating visual signals into language modeling, instruction tuning for English and a low-resource language (Swedish), and retrieval augmentation.

First, to study multimodal grounding, pretrained masked language models are exposed to tokenized video alongside aligned text, enabling analysis of how visual context influences next token prediction. Using the psycholinguistically motivated notion of imageability as an interpretable probe, the work shows that video grounding strengthens representations for concrete, highly imageable words, with the effect most consistent in a smaller model. For less imageable words, gains are mixed, and larger models exhibit increased reliance on visual context. These findings indicate that visual grounding benefits are not uniform; they depend on lexical properties and model capacity, and imageability offers a principled lens on what video–language models internalize.

Second, the thesis develops a practical path for instruction tuning in Swedish by translating existing English instruction corpora and finetuning models of varying size and pretraining exposure. Substantial zero-shot gains demonstrate that translated synthetic instructions can substitute for costly native resources. Complementing this, the work assesses automatic evaluation for instruction-following systems using Pairwise Accuracy as a meta-evaluation criterion. It finds that reliability is task- and length-dependent: ROUGE-L is a competitive, low-cost proxy for short, format-constrained outputs; BERTScore is comparatively stronger for longer, free-form answers; and LLM-as-a-judge aligns well with human judgments primarily when provided with reference answers. Cross-lingual analyses highlight that Swedish outputs exacerbate surfacematching weaknesses and no-reference biases, refining guidance on when human assessment remains necessary.

Third, the dissertation analyzes retrieval augmentation through a RETRO-style model. It shows that perplexity reductions concentrate on tokens with lexical overlap between inputs and retrieved neighbors,revealing a dominant surface-level “copy mode.” Leveraging this, surface-focused retrieval (e.g., BM25) is used to replace the dense retrieval mechanism during inference, which reduces perplexity further within this architecture, while lightweight hybrids (semantic pre-filtering with BM25 re-ranking) recover additional gains at minimal cost. The findings also demonstrate that during pretraining, performance improves sharply once input–neighbor overlap crosses a threshold; deliberately increasing overlap with targeted paraphrases can cut training time by about 40% without degrading downstream short-answer QA, though with a modest increase in eventual perplexity.

Overall, the thesis clarifies what signals large language models actually exploit and provides actionable recommendations for data curation, model selection, metric choice, and training strategies.

Retrieval-augmented language models have demonstrated performance comparable to much larger models while requiring fewer computational resources. The effectiveness of these models crucially depends on the overlap between query and retrieved context, but the optimal degree of this overlap remains unexplored. In this paper, we systematically investigate how varying levels of query–context overlap affect model performance during both training and inference. Our experiments reveal that increased overlap initially has minimal effect, but substantially improves test-time perplexity and accelerates model learning above a critical threshold. Building on these findings, we demonstrate that deliberately increasing overlap through synthetic context can enhance data efficiency and reduce training time by approximately 40% without compromising performance. We specifically generate synthetic context through paraphrasing queries. We validate our perplexity-based findings on question-answering tasks, confirming that the benefits of retrieval-augmented language modeling extend to practical applications. Our results provide empirical evidence of significant optimization potential for retrieval mechanisms in language model pretraining.

Den miljömässiga hållbarheten har blivit en alltmer central fråga för företag globalt. Miljöaspekterna förväntas ha en betydande inverkan på världsekonomin under det kommande decenniet. I takt med att medvetenheten om miljömässiga och sociala risker ökar, har behovet av tydliga riktlinjer och regelverk vuxit. Runtom i världen utvecklas strategier och metoder för att motverka, förhindra eller minimera koldioxidutsläpp. EU har påbörjat utvecklingen av metoder för att mäta miljöpåverkan, bland annat med ETS, en marknad för utsläppsrätter, och redovisning av klimatpåverkan.

Denna uppsats undersöker EU-direktivet om tillbörlig aktsamhet och dess potentiella inverkan på små och medelstora företag (SMF) i leveranskedjan. I uppsatsen analyseras hur direktivets krav påverkar SMF genom att identifiera de utmaningar som uppstår med avseende på ökade kostnader och administrativa bördor, samt de möjligheter som skapas, såsom en mer hållbar affärsverksamhet och förbättrad konkurrenskraft. Analysen bygger på en granskning av direktivet, relevant doktrin och rättsfall avseende den tyska och franska lagstiftningen om tillbörlig aktsamhet.

Analysen visar att direktivet riskerar medföra både positivt och negativt resultat för SMF. Å ena sidan kan direktivet bidra till utvecklingen av mer hållbara affärsmetoder och stärka SMF:s konkurrenskraft. Å andra sidan riskerar de ökade kostnaderna och administrativa bördorna att utgöra ett direkt hot mot företagens fortlevnad. Uppsatsen belyser därutöver en central implementeringsutmaning: risken att företag externaliserar kostnader till sina affärspartners, vilket underminerar direktivets ursprungliga syfte att skapa en mer hållbar och ansvarsfull företagsamhet.

För de företag som direkt omfattas av direktivet är det avgörande att göra en avvägning mellan de utmaningar som uppsatsen identifierat för SMF och de möjligheter som direktivet medför för att utveckla sin konkurrenskraft på en marknad där hållbarhet blir allt viktigare.

När vi nu summerar 2025 kan vi konstatera att året har varit både innehållsrikt och symboliskt viktigt för Linköpings universitet. Vi har firat vårt 50-årsjubileum – ett tillfälle att stanna upp, och samtidigt formulera vår roll i framtidens kunskapssamhälle. Jubileumsåret har präglats av stolthet över det universitet vi har byggt tillsammans, men också av en stark övertygelse om att våra mest betydelsefulla bidrag ligger framför oss.

Sedan starten har LiU kännetecknats av mod att tänka nytt, bryta traditionella ämnesgränser och arbeta tvärvetenskapligt, nära samhälle och näringsliv. Den andan är fortsatt tydlig och har under jubileumsåret kommit till uttryck i möten mellan studenter, medarbetare, alumner och de aktörer som vi formar framtiden tillsammans med.

Även verksamhetsmässigt har 2025 varit ett av LiU:s starkaste år. Våra utbildningar fortsätter att attrahera nya studenter och vara relevanta för arbetsmarknaden och samhällets behov. För sjätte året i rad ökar studentantalet, 2025 hade vi fler än 47 500 studenter och forskarstuderande. Identiteten som ett campus- och programuniversitet är stark och våra studenter ligger i topp när det gäller att fort få arbete efter studierna. Utvecklingen av generativ AI påverkar högre utbildning och LiU AI-säkrar därför all utbildning. En kraftsamling för att höja lärares AI-litteracitet samt utveckla kursinnehåll och examinationsformer. Allt för att rusta våra studenter för framtidens arbetsmarknad där källkritik, analytisk förmåga och ansvarsfull användning av AI kommer att krävas oavsett profession.

Våra forskare fortsätter att bidra med resultat som håller hög internationell kvalitet. Våra forskningsmiljöer utvecklas i nära samverkan med både akademiska partners och aktörer inom industri och offentlig sektor. Detta skapar förutsättningar för forskning som inte bara bidrar till vetenskaplig utveckling utan också till samhällsnytta. Under året har universitetet genomfört en omfattande forskningsutvärdering, LiRE25. Utvärderingen bekräftar den position som LiU har i den internationella forskningsfronten inom flera strategiskt viktiga områden, samtidigt som den ger ett värdefullt underlag för fortsatt kvalitetsutveckling och prioriteringar.

Under året har universitetet beslutat om en färdplan för att stärka vår position i Europa. Ambitionen är tydlig – genom att konsekvent sträva efter excellens i utbildning, forskning och samverkan ska LiU vara ett ledande europeiskt universitet år 2030. Därför har vi etablerat ett Brysselkontor för att stärka närvaron i det europeiska samarbetet, bidra till strategisk omvärldsbevakning och skapa bättre förutsättningar för LiU:s deltagande i europeiska forsknings- och utbildningssamarbeten.

Med det sagt verkar vi i en tid av geopolitisk osäkerhet, ökade handelspolitiska spänningar och en tilltagande deglobalisering som påverkar akademin i allra högsta grad. Akademisk frihet och lärosätens autonomi utmanas i flera delar av världen. Då blir det europeiska samarbetet centralt för att värna öppet kunskapssökande, internationellt samarbete och långsiktiga forskningssamarbeten. Genom våra europeiska partnerskap och allianser bidrar LiU aktivt till att stärka Europas kunskaps- och innovationskraft.

En avgörande framgångsfaktor för LiU är våra medarbetare och studenter. Det engagemang, den kompetens och den kreativitet som präglar verksamheten är avgörande för våra resultat och vår utveckling. Under jubileumsåret har detta blivit särskilt tydligt genom det stora deltagande och den stolthet som visats i olika aktiviteter och initiativ.

När vi nu blickar framåt är vår ambition tydlig. LiU ska fortsätta att vara ett universitet som kombinerar vetenskaplig excellens med samhällsrelevans, som utvecklar utbildningar och forskning i nära dialog med omvärlden och som aktivt bidrar till att möta de stora samhällsutmaningarna. Vår historia visar att förnyelse är en del av vår identitet – och den kommer att vara avgörande också under de kommande åren.

Avslutningsvis vill vi rikta ett stort tack till alla medarbetare, studenter, samarbetspartners och finansiärer som bidragit till universitetets framgångar. Tillsammans skapar vi förutsättningarna för fortsatt utveckling och för att Linköpings universitet även i framtiden ska vara en stark kraft för kunskap, innovation och samhällsutveckling.

Jan-Ingvar Jönsson, rektor Karin Axelsson, prorektor

This thesis studies the applicability of a simple mass-spring system together with implicit time integration schemes Forward Euler and Runge-Kutta 4 for the purpose of simulating physical behaviour of cloth. The two focus areas of the study are physical plausibility and numerical stability under variation of parameters in the model. The results are obtained for three scenarios simulated in Python, and come in the form of animations as well as plots that illustrate the stability and computation times for a set of time steps. The results show that Runge-Kutta 4 has a larger region of stability than Forward Euler in the tested scenarios while Forward Euler is more computationally efficient for any given time step, and that the mass-spring model combined with Runge-Kutta 4 can produce stable and sufficiently visually accurate solutions. Additionally, it is concluded that computation times are very long, something which motivates the use of implicit time integration schemes utilised in other studies on the subject.

Global collaboration remains vital, but rising travel emissions conflict with LiU’s climate commitments. To stay innovative and responsible, the Department of Thematic Studies needs a more ambitious strategy. This text presents a pathway toward a climate smart travel policy that supports both research and sustainability.

Kommunerna står inför en växande välfärdsutmaning: behoven av stöd ökar, men resurserna gör det inte. Ofrivillig ensamhet bland äldre har blivit en central samhällsfråga med konsekvenser för hälsa, livskvalitet och social sammanhållning. I denna kontext framträder civilsamhället som en strategisk resurs. Frivilligorganisationer och volontärer erbjuder flexibilitet, lokal kunskap och sociala relationer som kompletterar kommunens insatser. Staten har uppmärksammat frågan genom riktade satsningar och en nationell strategi mot ensamhet, vilket skapar nya förväntningar på samverkan mellan offentliga aktörer och civilsamhället.

Syftet med rapporten är att analysera kommunala samverkansstrategier med civilsamhället för att motverka och bryta ofrivillig ensamhet bland äldre medborgare. I rapporten utforskas hur politiker, tjänstepersoner och volontärer uppfattar och upplever betydelsen av frivilligorganisationer och dess volontärer i arbetet med att motverka och bryta ofrivillig ensamhet. Därutöver undersöks mötesplatser och anordnande av aktiviteter samt olika samverkansformer mellan frivilligorganisation och kommun.

Datainsamlingen bygger på 42 intervjuer med politiker, tjänstepersoner och volontärer i tre kommuner. Intervjuerna ger en bred bild av erfarenheter och strategier i praktiken.

Resultaten visar att civilsamhället har en mångsidig roll i arbetet mot ensamhet. Frivilligorganisationer kompletterar kommunens insatser genom att skapa mötesplatser, ordna aktiviteter och främja ideellt engagemang.

Mötesplatserna är centrala för att bryta isolering och skapa sociala nätverk.

De kan vara kommunalt koordinerade träffpunkter, lokala samlingsplatser i bostadsområden eller mötesrum i servicehus. Aktiviteterna varierar från kulturarrangemang och samtalsgrupper till individuella insatser som ledsagning och personligt stöd. Volontärernas engagemang gör det möjligt att skapa relationer som skiljer sig från formella omsorgsrelationer och som ofta upplevs som mer personliga och meningsfulla.

Analysen visar tre samverkansformer som är särskilt framträdande:

• Kommunen som facilitator: Här skapar kommunen förutsättningar för civilsamhällets arbete genom ekonomiskt stöd, lokaler och idéburet offentligt partnerskap (IOP). Civilsamhället ges stor frihet att utforma verksamheten, vilket stärker engagemanget och öppnar för nya idéer. Strategin bygger på tillit och bottom-up-principer, men kräver tydliga gränser och rutiner för uppföljning.
• Kommunen som katalysator: I denna modell tar kommunen en aktiv roll i att initiera projekt, finansiera koordinatorer och bygga nätverk. Målet är att skapa momentum och engagemang i civilsamhället. Kommunen fungerar som drivande kraft i uppstartsfasen, men ambitionen är att civilsamhället ska ta över ansvaret på sikt. Utmaningen ligger i att balansera styrning med frivilligorganisationernas självständighet.
• Kommunen som koordinator: Denna form är vanlig i mindre kommuner och bygger på lokal identitet och småskalighet. Kommunen samordnar aktiviteter via mötesplatser och aktivitetsledare, ofta i nära samarbete med frivilligorganisationer och lokala företag. Strategin skapar kontinuitet och stärker gemenskapen, men kräver tydliga kontaktvägar och rutiner för samverkan med äldreomsorgen.

Gemensamt för alla former är behovet av materiella resurser, tillitsbaserade relationer och strukturer för långsiktighet. Utmaningar som framträderär otydliga gränser mellan kommunalt ansvar och frivilliginsatser, risk föröverbyråkratisering och svag kontinuitet i projektbaserade modeller.

Utifrån resultaten i studien ges ett antal rekommendationer för hur kommuner och frivilligorganisationer kan stärka sin samverkan.

• Kommuner bör aktivt överväga en mix av facilitator, katalysator och koordinator som arbetssätt och anpassa dem efter civilsamhällets kapacitet, organisationskultur och lokala behov.
• Anpassa samverkansstrategi efter lokal kontext och civilsamhällets kapacitet. Kommuner bör utgå från civilsamhällets resurser, lokala nätverk och behovsbild när de väljer hur samverkan ska organiseras.
• Säkerställ långsiktiga och materiella förutsättningar för civilsamhället. Kommuner bör erbjuda lokaler, ekonomiskt stöd och stödstrukturer som underlättar nätverkande och skapar tillit mellan parterna.
• Projektbaserade samverkansformer kräver långsiktig planering för hur volontärer ska fortsätta arbetet när projekttiden är slut.
• Stärk och synliggör volontärernas insatser. Kommuner bör aktivt arbeta med att uppmärksamma och bekräfta volontärer för att stärka engagemang och motivation.
• Nyttja frivilligorganisationen som dialogpart för att ta vara på deras lokala kunskap om behov och hur de kan tillgodoses samt hur insatser kan samordnas.
• Tydliggör gränsdragningar för kommunens lagstadgade ansvar, samt erbjud administrativa förenklingar, finansieringsmöjligheter och mindre byråkratiska former för stöd som möjliggör snabba och anpassade civilsamhällsinsatser.
• Främja inkluderande och kostnadsfria aktiviteter som utgår från äldres egna intressen och gör dem till medskapare.
• Utgå från att äldre är en heterogen grupp. Äldre ensamma är en heterogen grupp med olika förutsättningar, förväntningar och behov. I samverkan med civilsamhället är det viktigt att skapa förståelse för vilka organisationer som kan möta de olika behoven hos äldre personer i ofrivillig ensamhet.

Sammanfattningsvis visar studien att civilsamhället är en nyckelaktör i välfärdsmixen. För att möta den demografiska utvecklingen och minska ensamhet krävs strategiska satsningar som kombinerar materiella resurser,tillit och erkännande av frivilligorganisationernas unika roll. En väl avvägd mix av facilitator-, katalysator- och koordinatorstrategier kan skapa hållbara lösningar för framtidens lokala välfärd.

Safety and efficiency are cornerstones of modern-day commercial aviation. Rising traffic density and increasingly volatile, localized weather make real-time, en route decision-making critical for efficiency, safety, and resilience. Peer-to-peer (P2P) communication through a System of Systems approach can enable aircraft to share situational data and coordinate locally, thereby reducing exposure to hazards while maintaining stable schedules. This thesis builds and evaluates a reproducible SoS model to quantify how real-time aircraft-to-aircraft communication affects route deviation, delay, and system resilience under localized weather. It analyzes the operational effect of airborne P2P coordination with transparent, lightweight models focused on en route behavior in dense corridor airspace. The work adopted a System of Systems perspective and implemented a lightweight Python model that combines Agent-Based Modeling for aircraft behavior, Discrete Event elements for alerts and rerouting decisions, and aggregate views that echo System Dynamics through KPI trajectories. The constant metrics were then applied to compare three scenarios of P2P communication (non-existent, instant, and delayed), illustrating how localized events scale into system-wide efficiency and safety effects. Across application cases and mesh configurations, the results were consistent: the share of aircraft affected by adverse weather decreased notably, with minimal or no increases in delay, thereby improving safety while maintaining efficiency. The effect is strongest in the instant P2P communication scenario, yet closely followed by the realistic delayed P2P scenario. The main model application case in Croatian airspace showed a decrease of aircraft affected by the adverse weather of 27%, while the mean delay remained within 3 seconds. The findings of this thesis underline this indeed is a promising area for investing in further development. While this scenario highlights one concrete manifestation of the communication problem, the overarching aim is to assess the systemic value and the consequences of communication structures themselves, using simulation to test their impact under varying conditions. This approach allows the work to contribute not only to a specific operational case but also to the more general question of how communication capabilities shape the behavior and performance of interconnected systems.

This thesis investigates the possibility of deriving a controller tuning procedure based on black-box modelling with closed-loop system identification. As many users operate their systems 24/7, they are neither able nor willing to shut down for retuning procedures as operating conditions change. Thereby, the ability to adjust control parameters in real-time becomes increasingly important. To generate informative data, input excitation signals like steps, chirps and telegraphs were applied. The direct approach to closed-loop system identification was used to estimate black-box models of the ARX, ARMAX and BJ structures and methods like observing pole/zero diagrams and coefficient uncertainties were used to validate models. First order with delay models were approximated from the best performing black-box models and were used to compute new control parameters. Performance metrics like the Harris index and squared control error were calculated before and after retuning to evaluate the effectiveness of the retuning. The telegraph input excitation signal generated informative data both in noisy simulations as well as on a physical test unit. From evaluation of the black-box structures, the BJ structure consistently performed the best, accurately capturing the system dynamics. Utilising the noise models from this structure to calculate the Harris index indicated better controller performance after applying the new control parameters. With these results in mind, a tuning framework could be built. These results imply that the tuning procedure suggested by this thesis is well founded. It can deliver new control parameters that help the system track the operating setpoint. Future work should emphasise automating the tuning procedure as well as calculating the Harris index over time to indicate when a retuning is necessary.

A nanoparticle is defined as a particle which is less than 100 nm at least in one dimension. Nanotechnology enables the integration of entities with complementary or tailored properties into a single functional unit. In this thesis, we demonstrate how nanoprobes can be rationally designed to target and image biomedical structures and sense intracellular pH.

Cerium is biocompatible element and cerium oxide (CeO_x) nanoparticles are widely used in biomedical applications; however, they do not inherently generate a magnetic resonance (MR) signal. Gadolinium provides excellent contrast in magnetic resonance imaging (MRI) and is extensively used in clinical practice. Both cerium and gadolinium-based materials can also provide contrast in computed tomography (CT).

In this work, we aim to develop the next-generation dual-mode contrast agents for combined MRI and CT imaging by incorporating gadolinium into the cerium oxide lattice. In paper I we focus on the nanoparticle core and provide thorough characterization of a cerium oxide nanoparticle doped with gadolinium. We synthesized cerium oxide nanoparticles containing 5-20% gadolinium within the crystal lattice and evaluated their ability to enhance MRI contrast via relaxivity measurements. The resulting nanoparticles exhibit higher relaxivity than commercially used contrast agents.

The aim of paper II is to develop a nanoprobe capable of tracking the pH fluctuations that naturally occur in lysosomes. We synthesize SiO_x shell nanoparticles loaded with pH sensitive fluorophores. The emission wavelength of the first fluorophore falls within the excitation range of the second fluorophore, enabling Förster resonance energy transfer. The second fluorophore has a ring structure that will open at low pH, this makes the compound fluorescent. By measuring the ratio between the two emission maxima we can determine the pH. The cellular uptake of the pH sensitive nanoprobes is significantly increased using a cyclic disulfide.

In paper I, we show that there is great potential for cerium oxide nanoparticles with integrated gadolinium to act as contrast agent in both CT and MRI. Our aim in paper III is to further develop nanomaterials with enhanced targeting capabilities for contrast agent applications. Here we use the CeO_x core with 5% Gd to create epidermal growth factor receptor targeting nanoprobes. Poly acrylic acid (PAA) is used as a capping agent to provide colloidal stability and biocompatibility. The PAA is prefunctionalized with a fluorophore and a clickable moiety. The nanoparticles were then conjugated to the monoclonal antibody cetuximab via click chemistry. The nanoprobes were evaluated with respect to core and coating characteristics as well as targeting efficacy and cellular uptake.

The aim of paper IV is to explore a novel technique that could enable future upscaling of the nanoparticle synthesis. We use the same nanoparticle formulation as in paper I and further coat them using plasma enhanced chemical vapor deposition (PE-CVD). This creates multi-core nanoparticles with an organic coat. The obtained particles are thoroughly characterized and further functionalized using hydrazide and click chemistry. The polymeric thin film obtained using PE-CVD is characterized with respect to position in reaction chamber, electrode size and shape using Au surfaces. We show that using a small bottom electrode and positioning the samples close to the reaction chamber opening enables the incorporation of ketones in the organic matrix.

Thrombin is a multifunctional enzyme involved in the coagulation of blood. One role of thrombin is to bind and activate blood platelets to induce blood coagulation. As thrombin has a central role in blood coagulation it is a common target for anti-coagulant treatment. Paper V investigates the effect of direct thrombin inhibitor dabigatran on thrombin platelet binding. Thrombin contains two exosites that mediate interactions with multiple targets, including platelets. Herein, we show that although dabigatran attenuates thrombin binding to platelets, the binding affinities of exosites I and II are unaffected. Demonstrating the complexity of thrombin binding and that the exosites likely require synergistic binding for the thrombin-platelet interaction.

Characterization in this thesis is performed using dynamic light scattering (DLS), infrared spectroscopy (IR), transmission electron microscopy (TEM), scanning transmission electron microscopy (STEM), scanning electron microscopy (SEM), atomic force microscopy (AFM), X-ray photoelectron spectroscopy (XPS), X-ray diffraction (XRD), fluorescence spectroscopy and microscopy and surface plasmon resonance (SPR).

In summary, this thesis explores a series of nanoparticle syntheses and nanoprobe formulations to develop tools for targeted biomedical imaging.

This thesis explores the intersection of team development, knowledge management, and performance assessment within tactical and cyber control teams, focusing on how data-driven approaches can facilitate experience sharing and improve operational outcomes. In high-stakes environments such as military and rescue operations, teams must rapidly adapt to evolving situations, often acting on incomplete information and under time pressure. Society depends on them being successful, so how can we ensure that our tactical teams are equipped to handle these challenges? This thesis aims to address this question by exploring the role of data-driven approaches to team performance assessment for enhanced training and knowledge sharing.

The research is based on findings from studies encompassing a literature review on media-rich externalized knowledge representations, a large-scale cyber defence exercise, and a controlled virtual team experiment. The Baltic Cyber Shield exercise provided an opportunity to observe team dynamics, strategy development, and performance measurement in a realistic cyber defence scenario. Collaborative decision-making in defensive cyber operations was at the time a little-studied area, and the exercise offered unique insights into how ad-hoc teams operate under pressure in a complex environment. The second experiment, conducted at Hartnell College, enabled a detailed comparison between self-assessment and observer ratings in virtual teams, highlighting the challenges of evaluating team effectiveness. The virtual team setting at this experiment allowed for collection of rich data on team interactions and performance, enabling another study on the feasibility of automation for team performance assessment in controlled environments.

Collected data included surveys, observer reports, system logs, and automated performance metrics. These data were analysed using statistical methods and regression analysis to identify relationships between team behaviours, decision-making processes, and outcomes. The studies reveal that data-driven analysis not only supports experience sharing but can also be used to assist in performance assessment, offering a complementary approach to traditional self-ratings and observation-based assessments.

Key findings indicate that data-driven approaches can indeed accelerate team development, e.g. by strengthening the feedback loop between team members and enhancing their ability to learn from each other. Integrating automated performance metrics with observer and self-assessment data enables a more comprehensive analysis and understanding of team performance and effectiveness. The research highlights that diverse data sources can reveal complementary strengths and limitations, enabling more accurate identification of areas for improvement and supporting targeted interventions in team development.

Centralized collection and processing of healthcare data across national borders pose significant challenges, including privacy concerns, data heterogeneity, and legal barriers. To study some of these challenges, we formed an interdisciplinary consortium to develop a federated health data network, comprised of six institutions across five countries, to facilitate Nordic-Baltic cooperation on secondary use of health data. The objective of this report is to offer early insights into our experiences developing this network. We employed a mixed-methods approach, combining both experimental design and implementation science to assess the factors influencing the implementation of our network. Technically, our experiments indicate that the network functions without significant performance degradation compared to centralized simulation. While use of interdisciplinary approaches holds a potential to solve challenges associated with establishing such collaborative networks, our findings turn the spotlight on the uncertain regulatory landscape playing catch up and the significant operational costs.

New research highlights the importance of making more productive connections between museum work and the environment to take action and work more sustainably. This paper provides a novel entry on how to understand museum things as consisting of bundles of ecological relations. This new way of understanding how things exist provides an important method for tracing how museums impact environmental change. In our pilot study Museum Ecologies, we have worked with digital collection databases as an example to show how museum collections are interlinked with climate change and climate change transitions. The research elaborates and expands upon the museum pedagogic method, Tingenes Metode, to explore how this can be developed beyond its anthropocentric notions, using environmental humanities perspectives, and tools from a variety of thinkers within critical feminist post-human and new materialist theories. This paper investigates how museums and cultural heritages, observed, and understood as ecological agents, can guide us as we struggle to find new ways to live in the Anthropocene.

Background:

The ultrasound-guided attenuation parameter is well established for hepatic steatosis detection in metabolic dysfunction-associated steatotic liver disease. The diagnostic performance of ultrasound-guided attenuation parameter was evaluated using different numbers of measurements at different lateral locations to detect hepatic steatosis ⩾ S1 in male and female patients with metabolic dysfunction-associated steatotic liver disease.

Methods:

A metabolic dysfunction-associated steatotic liver disease cohort was prospectively enrolled in autumn of 2022. Ultrasound-guided attenuation parameter values obtained through one to five measurements, performed at single and multiple locations, were compared with proton density fat fraction. Presence of hepatic steatosis (i.e. ⩾ S1) with ultrasound-guided attenuation parameter was defined as a proton density fat fraction of ⩾ 5%. Diagnostic performance was evaluated based on the area under the receiver operating characteristic curve.

Results:

Included 60 participants with an even sex distribution. Ultrasound-guided attenuation parameter diagnostic performance to detect hepatic steatosis ⩾ S1 did not significantly differ according to the number of measurements (from 1 to 5), different lateral locations, or patient sex. Ultrasound-guided attenuation parameter performed using five measurements in one location exhibited a receiver operating characteristic curve of 0.87 (95% confidence interval: 0.78, 0.97), and a threshold of 0.53 dB/cm/MHz, yielding 90% sensitivity and 65% specificity. Three measurements in multiple lateral locations exhibited a receiver operating characteristic curve of 0.91 (95% confidence interval: 0.84, 0.98), with a threshold of 0.58 dB/cm/MHz, yielding 95% sensitivity and 75% specificity.

Conclusion:

Ultrasound-guided attenuation parameter diagnostic performance to detect hepatic steatosis ⩾ S1 in metabolic dysfunction-associated steatotic liver disease is similar with regions of interest in single versus multiple lateral locations. Three measurements in multiple lateral locations appear sufficient to detect hepatic steatosis, which must be evaluated for all hepatic steatosis stages.

Weight functions play an important role in estimating the capacity of sets. This thesis examines radial weight functions $w$ in $\mathbb{R}^n$ and how algebraic modifications of these weights affect their properties, especially in relation to admissibility and exponent sets. A two-parameter modification of the form $\widetilde{w}(r) = r^s w(r^t)$ is introduced, and its influence on the associated measure and exponent sets is analyzed. A corollary is proven which shows that, under certain circumstances, a modified weight $\widetilde{w}$ is $1$-admissible. The relationship between the exponent sets associated with $w$ and those associated with the modified weight $\widetilde{w}$ is also studied, and the mapping of exponents $q$ (for $w$) to $\widetilde{q} = (q-n)t+s+n$ (for $\widetilde{w}$) is given by a lemma. This demonstrates how the choice of parameters $s$ and $t$ determines the difference between exponent sets. Two new examples are constructed to show how specific exponent set combinations can be obtained, using previously developed methods. The corresponding weights of these two examples are extracted and are shown to be $1$-admissible. Altogether, the results obtained in this thesis show that radial weight modifications can be used to design weights with desired properties, contributing to a wider understanding of weighted measures in $\mathbb{R}^n$.

This study was conducted to look at how religious Muslim representatives perceive interreligious dialogue, whether it makes sense and how the Muslim subject emerges through the informants' statements. Semi-structured interviews were conducted with five informants. The thematic analysis went through three phases and was finally implemented together with Foucault's theories of subject, power-knowledge and historical kinship. To provide an additional dimension to how the Muslim subject experiences Islamophobia through the informants' statements, Miranda Fricker's concept of epistemological (in)justice is used. The analysis showed that open conversations, conditions for cooperation and friction are a way to create trust for the informants. It was also meaningful to influence outside the interreligious dialogue and to highlight different perspectives, which allows you to see people behind the prejudices you may have about the other. By listening to the plea expressed through the need Muslims face when they must answer questions about political Islam, Islamism, or when all fingers are pointed at them, the study suggests that cultural and institutional change is possible. 

Sub-terahertz (Sub-THz) communication is considered as a key enabler for sixth-generation (6G) and beyond communication systems, owing to its abundant spectrum resources that support extreme data rates, high spatial resolution, and low latency. However, the unfavorable propagation characteristics and significant hardware impairments pose substantial challenges to practical deployment, often resulting in high implementation costs.

Current commercial communication systems predominantly rely on colocated multiple-input multiple-output (MIMO) architectures to provide high spectral efficiency and reliable service. At sub-THz frequencies, although the severe free-space path loss can be largely compensated by large antenna arrays and highly directional beamforming, the susceptibility to blockage remains a fundamental limitation of co-located MIMO systems. Due to the strong dependence on a dominant line-of-sight path and the co-located nature of the antenna elements, obstruction by human bodies, objects, or environmental dynamics can simultaneously degrade all antenna links, resulting in abrupt signal attenuation and intermittent connectivity. Moreover, the pronounced hardware impairments at sub-THz frequencies, such as power amplifier (PA) inefficiency, phase noise, and limited data-converter resolution,increase system complexity and implementation cost, thereby challenging the practical deployment of co-located MIMO architectures in mobile sub-THz communications.

Recent advancements in polymer microwave fiber (PMF) technology have created significant opportunities for robust, low-cost, and high-speed sub-THz radio-over-fiber (RoF) communications. As an alternative, novel RoF architectures can facilitate cost-effective realization of sub-THz systems.

Recognizing these potential benefits, this thesis explores a novel RoF structure that interconnects multiple radio units (RUs), booster units (BUs), and a central unit (CU) in cascade via PMF, envisioning its application in indoor scenarios. This structure creates several research opportunities when considering cascaded distortion effects introduced by non-linear PAs and the propagation channel over the fiber.

Within this context, the contributions of this thesis are twofold.

i) Uplink positioning: We propose uplink positioning algorithms that exploit the cumulative effects of cascaded non-linear PAs and dispersive PMFs. Specifically, we develop maximum-likelihood and non-linear least-squares estimators to determine the entry RU and the time-of-arrival between the RoF system and the user equipment, where identifying the entry RU corresponds to estimating the signal propagation distance along the RoF stripe. For the special case of linear PAs, we derive the Cramér–Rao lower bound (CRLB)to benchmark estimator performance. Our simulation results demonstrate that the proposed estimators remain effective even under cascaded non-linear distortions, and that the architecture enables cost-efficient, high-resolution indoor positioning. In the numerical evaluations, experimentally measured PMF characteristics for high-density polyethylene fibers are also incorporated.

ii) Waveform selection: We introduce a set of candidate waveforms and compare them using multiple performance metrics to assess their robustness against hardware impairments. The considered waveforms are categorized into three operational regimes, each associated with a distinct system configuration. Several waveforms are then generated and evaluated within the considered cascade-structured RoF Sub-THz system. Importantly, we provide new insights and an in-depth comparative analysis of the dominant characteristics of each waveform, highlighting the most suitable option for the envisioned architecture and offering a detailed complexity assessment.

Forced displacement is at a historical high point, with over 123 million people estimated as being forcibly displaced in 2024. Adolescents make up the majority of this population. Although Sweden has recently seen the lowest number of asylum applications since the 1990s, it has for a long time been an important resettlement country for those forcibly displaced, and many adolescents with refugee backgrounds who fled to Sweden in the last decade still reside there. Beyond sharing the experience of having to flee, adolescents with refugee backgrounds are a diverse population, with experiences before, during, and after migration differing significantly.

Research on violence and its consequences among adolescents with refugee backgrounds is growing and shows that violence exposure and poor mental health is prevalent within the population. However, several research gaps still exist. First, prevalence rates for mental disorders are noticeably heterogeneous between studies, even after methodological differences are accounted for, and research rarely captures mental health as a unified concept consisting of functioning, well-being, and symptoms. Second, complex posttraumatic stress disorder (CPTSD) remains understudied in terms of validity, prevalence, and risk factors, despite indications that the diagnosis may be of particular relevance for the population. Third, data on violence exposure often overlook specific forms of violence, such as child maltreatment, and studies rarely report exposure across different migration phases. Fourth, the potential differential impact of violence exposure in different migration phases, and the influence of polyvictimisation, on mental health are poorly understood.

This thesis aimed to bridge these gaps and provide a better understanding of patterns of violence exposure and mental health among adolescents with refugee backgrounds. To do so, a systematic review mapped the use of person-centred statistical methods in research on violence exposure and mental health within the population (Study I). Furthermore, in a Swedish community sample of adolescents with refugee backgrounds (N = 296), person-centred statistical analysis was used to explore mental health subgroups (Study II) and to examine the prevalence, validity of, and risk factors associated with CPTSD (Study III). Finally, structural equation modelling along with descriptive statistics was used to investigate the prevalence of violence exposure across migration phases and their associations with mental health outcomes (Study IV).

The findings showed that mental health subgroups based on general functioning, well-being, and trauma symptoms can be a theoretically meaningful and clinically useful way of describing heterogeneity in mental health. Additionally, the findings demonstrated a high prevalence of CPTSD and provided further validity of the diagnosis. Child maltreatment and polyvictimisation were identified as significant risk factors for poor mental health. Furthermore, the results revealed a high prevalence of violence exposure across all phases of migration, with certain types and forms of violence, as well as polyvictimisation, being particularly prevalent. Additionally, the timing of violence exposure was found to be associated with mental health outcomes: early exposure was related to trauma symptoms while later exposure was more associated with general functioning and well-being. Additional factors beyond violence also played a significant role, with asylum and unaccompanied status found to be equally as strongly associated with mental health outcomes as violence exposure.

These findings offer nuance in describing adolescents with refugee backgrounds, challenging narratives of vulnerability. Mental health subgroups can inform professionals about the type of intervention to offer and where to deliver it, and may play a role in designing interventions or evaluating the effectiveness of existing ones. The high prevalence of CPTSD emphasises the importance of addressing barriers to care for adolescents with refugee backgrounds and developing clinical guidelines for the diagnosis. Finally, improving public policies to increase security and safety may be essential to address both violence exposure and other stressors and, subsequently, improve mental health among adolescents with refugee backgrounds.  

In Kazuo Ishiguro's novels Never Let Me Go (2005) and Klara and the Sun (2021), themes such as subjectivity, identity, autonomy and relationality are explored through the novels' protagonists, Kathy and Klara respectively. This paper aims to investigate these themes in relation to the characters Kathy and Klara. The paper is based on the following research questions: in which ways does the narrative construct and affirm the subjectivity and personhood of the characters, based on the concepts of subjectivity, embodiment, and interpersonality? and in which ways does the narrative counteract and deny this possible subjectivity and personhood, based on the aforementioned concepts? Using texts by posthumanist scholars such as Braidotti, Hayles, and Haraway as its theoretical framework, this paper employs a thematical method to analyze Ishiguro's novels, as well as incorporating a special focus on relationality throughout the analysis. As previous research as well as this study indicate, the question of the characters' is complicated and thwarted by conscious and unconscious processes of othering, objectification, indoctrination, and isolation. The findings of this paper show that subjectivity to a large degree appears to be something socially constructed and dependent on one's relations to one's surroundings instead of a result of an innate essence, or rather, that this essence is created by others' belief that it exists. This perspective encourages a larger moral debate regarding on what basis society defines subjectivity and personhood in relation to genetic engineering and artificial intelligence, and encourages an examination of the values behind these scientific fields.