Högskolan i Skövde

Background: Hybrid work is a phenomenon that has become more popular as a way of working in recent years in line with technological developments and the corona pandemic. This way of working, which involves alternating between working in the physical workplace and working remotely, such as from home, can have both positive and negative aspects. It can result in increased flexibility and a better work-life balance. Hybrid work can, on the contrary, result in decreased social cohesion, impaired communication and social isolation. 

Purpose: This study aims to investigate how managers adapt their leadership and communication strategies to promote a sense of community in a hybrid work environment, as well as to highlight how these adaptations interact with employees' actual experiences. 

Method: This study is designed as a case study of a retail company in Western Sweden. This study has applied mixed methods, which is a combination of qualitative and quantitative methods. The empirical material was collected using semi structured interviews with sex managers and a survey that was sent to the managers employees. The survey was sent to 33 employees and 28 responded, giving the study a response rate of 84,8%. The material collected from the interviews was then transcribed and analyzed through thematic analysis. Responses from the survey were analyzed through descriptive statistics. 

Results and Conclusion: The results show that the managers adapt their way of working when the team is working remotely by being more attentive during digital meetings and by focusing on informal conversations because the managers consider that communication worsens during hybrid work. Another important aspect to build and maintain social cohesion in a team is according to the managers to implement social activities such as teambuilding during physical meetings. The managers implement this because they consider that social cohesion can only be built during physical meetings and then be maintained during remote work. However, the employees consider that the managers’ adaptations do not contribute to increased social cohesion in the team, which implies that the managers’ adaptations are not effective enough in practice. Another difference in experience between the managers and employees is the opinion on motivation and engagement among the employees. The managers’ perception is that these factors are impaired during remote work while the employees, on the contrary, do not experience depreciation. This can be a result of the managers’ adaptations effectively maintaining motivation and engagement or the result of a discrepancy between the managers and employees’ perceptions.

Designing serious game visuals for multiple cultural contexts presents significant challenges, as imagery familiar to one audience may be misunderstood or carry different meanings in another. This thesis adopts the perspective that visuals in serious games are not merely decorative but function as connectors between game content and players’ cultural and perceptual contexts. It argues that visual elements must be both clear and appealing across different cultural and domain contexts, and therefore investigates how varying levels of visual fidelity influence players’ visual recognition and visual preferences in cross-cultural serious game contexts.

A mixed-methods research approach was adopted, combining a literature review, fieldwork, and multiple case studies, with qualitative and quantitative data collected in Nepal and Sweden. To support the empirical investigation, a serious game titled Happy Heart was designed and developed as both an educational game and a research instrument, enabling systematic evaluation of visual fidelity across culturally distinct contexts. In both locations, the same Happy Heart game was implemented in localized versions and playtested with local participants.

The findings indicate that higher levels of visual fidelity generally support improved recognition, increased preference, and greater familiarity with serious games in both cultural contexts. However, higher visual fidelity did not consistently result in stronger emotional responses. Participants in both Nepal and Sweden experienced difficulties identifying culturally specific food items, highlighting the role of visual literacy in shaping how visual representations are interpreted. The results further suggest that adolescents’ visual recognition and preferences are shaped more strongly by exposure to global media, such as video games, animation, and social platforms, than by everyday lived experiences. A clear preference for 3D graphics and globally recognizable game aesthetics was observed, reflecting increasing familiarity with shared visual design conventions across cultures.

Based on these findings, this thesis proposes a Visual Alignment Framework to support culturally inclusive serious game design. The framework offers structured guidance for selecting representational objects and determining appropriate levels of visual fidelity when designing for diverse audiences, thereby reducing the risk of visual miscommunication in cross-cultural contexts. This research contributes both empirically and conceptually to the field of serious games by providing cross-cultural evidence and a theoretical foundation for visual alignment in game visuals. By synthesizing insights from the Nepalese and Swedish case studies, the thesis advances understanding of how visual communication can be strategically aligned with players’ cultural and perceptual contexts to support more effective (in terms of visual recognition and visual preference) and inclusive educational game design. Future research is encouraged to further evaluate the framework across additional cultural settings and game genres.

Background: Sexual health is an important part of young people's social and personal well-being and development. Receiving education in sexual health is a prerequisite for each individual to be able to make informed decisions. The school nurse has the opportunity to inform and educate young people about sexual health during health interviews. However, the school nurse lacks the right conditions to teach the subject and collaboration with other staff at the school should be improved.

Aim: To investigate school nurses' experiences of discussing sexual health during health consultations in upper secondary school.

Method: The study is a qualitative interview study with an inductive approach. Which was conducted with an open interview guide. The material from the interviews was analyzed through a qualitative content analysis.

Findings: The results show that the health interview and the health questionnaire are an important gateway and support in conversations about sexual health. Important factors for the quality of the conversation are trust, security and relationships between school nurse and student. The results show that school nurses experience differences in the conversation based on gender and culture. In summary, the school nurses experience that the conversation is important and complex.

Conclusion: The conclusion is that school nurses need more time to talk about sexual health. More ongoing training to stay up to date in the area of sexual health. The school nurses believe they need to be trained in gender and culture in the aspect of sexual health.

Debates about underdetermination and interpretation—familiar from the philosophy of quantum mechanics—may also offer valuable insights for evaluating alternative cosmological frameworks. Bohm (1952) argued that the usual ‘Copenhagen’ interpretation of quantum theory is unfalsifiable not because it is empirically empty, but because its assumptions leave the mathematical formalism free to absorb any anomaly through ad hoc modifications. His response was not to argue that Copenhagen was false but to construct a hidden-variables alternative reproducing its predictions—showing, by existence rather than argument, that Copenhagen’s commitments were never forced by the data. We ask whether an analogous situation might be emerging in cosmology. The Dynamic Universe (DU) model of Suntola (2026) offers a candidate: from two postulates it reproduces several ΛCDM (standard model of Big Bang cosmology’s) predictions without dark energy or inflation, and with fewer parameters, while predicting a monotonic angular-size–redshift relation that recent JWST (James Webb Space Telescope) data appear to favour over ΛCDM’s non-monotonic one. We argue the Bohmian parallel is instructive but partial: DU’s equivalence with ΛCDM remains undemonstrated in domains—the CMB (Cosmic Microwave Background) spectrum, nucleosynthesis, gravitational waves—where ΛCDM’s evidence is strongest. Nonetheless, we suggest such rivals warrant continued attention precisely for what their partial success reveals about the mainstream view’s contingency.

This chapter first contrasts the classical worldview (deterministic, reductionist, and mechanistic) with that of quantum theory (indeterministic, contextual, and holistic). The working hypothesis is that quantum features may help explain aspects of consciousness that classical neuroscience struggles with. The chapter then reviews several major quantum interpretations and their connection to consciousness. First, the Copenhagen interpretation and QBism, which hold that the wave function represents knowledge or subjective beliefs, not objective reality. Second, Wigner’s view that consciousness is required to collapse the wave function, turning superpositions into definite outcomes. Third, the reverse idea by Penrose and Hameroff that collapse constitutes consciousness, in that objective gravitational collapse of quantum states (Orch OR) in neural microtubules underlies conscious moments. Fourth, the pilot wave interpretation, in which no collapse is needed; instead, in Bohm’s version, ‘active information’ in pilot waves guides particles and provides the intrinsically mental ground from which consciousness may emerge. And fifth, Everett’s many-worlds interpretation, according to which no collapse is needed and the world divides into branches at the macroscopic level in measurement-like situations; here, some researchers have linked conscious experience to superposition formation. The chapter concludes by briefly considering some related applications of quantum theory, such as quantum biology and quantum cognition.

The western honeybee, Apis mellifera, depends heavily on proper nutrition during overwintering, when colonies survive entirely on stored food resources. Many beekeepers replace natural honey with sugar-based feeds, but the effects of these diets on the honeybee gut microbiome are still not fully understood. This study investigated how different overwintering diets, including honey, sugar syrup, and a mixed honey-sugar diet, affect the gut microbiome of worker honeybees, with a special focus on the ileum, an important but understudied gut region. Worker bees were collected from three dietary treatment groups in Västra Götaland, Sweden, during early spring 2025. Ileal samples were dissected under sterile conditions, and full-length 16S rRNA gene sequencing was performed using Oxford Nanopore MinION technology to improve bacterial identification. Microbial diversity and bacterial composition were analyzed using taxonomic and statistical approaches. The results showed that core bacterial genera such as Gilliamella, Lactobacillus, and Snodgrassella were present in all dietary groups, indicating that the honeybee gut microbiome remained generally stable during overwintering. Bees fed with honey showed slightly higher microbial diversity compared with bees fed sugar or mixed diets. Some differences in the abundance of certain bacterial taxa were also observed between dietary treatments. Comparisons between the ileum and rectum revealed similar dominant bacterial groups, although the rectum contained greater bacterial abundance and variation. Overall, the study suggests that the overwintering diet can influence the balance and diversity of gut bacteria, while the core microbiome of honeybees remains relatively resilient under different feeding conditions.

Biological nitrogen fixation (BNF) allows plants to utilize atmospheric nitrogen for their nutrient needs, through a symbiotic relationship with rhizobia and has attracted considerable attention of the scientific community due to its potential to enhance sustainability in agriculture. Recent studies, focusing on model legume Medicago truncatula, imply an active role of glycoprotein GBP1 during nodulation, which is an important step in the establishment of the legume-bacteria relationship. The physiological properties of this protein remain poorly studied. Structural data indicate the existence of nine N-glycosylation sites plus one predicted site that could not be observed in the crystal structure. To investigate the importance of glycosylation in the physiological function of the protein, constructs with their glycosylation sites mutated were expressed in Nicotiana benthamiana and their structural stability and enzymatic activity were examined using nano differential scanning fluorimetry and reducing sugar assays.

Mutations affecting glycosylation resulted in a decrease in both protein stability and enzymatic activity, with the extent of these effects being highly site-dependent. This decreases across the mutants suggests that glycosylation contributes to both the structural integrity and functional efficiency of MtGBP1. Interestingly, the fully deglycosylated mutant apoGBP1 retained enzymatic activity and exhibited higher melting temperatures than variants with only single glycosylation site disrupted, shedding light in the non-additive effects of accumulation of mutations. Overall, these findings underline the importance of site-specific N-glycosylation in maintaining MtGBP1 stability and function and provide new insights into the role of site-specific glycosylation in regulatory role of glycosylation in enzyme activity.

Introduction: Body mass index (BMI) fluctuation has adverse health consequences, yet its determinants remain poorly understood. We examined genetic and environmental influences on BMI fluctuation and its associations with BMI trajectories.

Data and Methods: Data from 14 longitudinal twin cohorts, including 58 311 individuals (54% women) and 22 714 complete twin pairs (9761 monozygotic), were pooled. BMI trajectories from ages 18 to 99 years were estimated using linear mixed-effects models. BMI fluctuation was defined as the average squared residual between observed and expected BMI around individual trajectories. Genetic and environmental contributions to BMI fluctuation and its associations with baseline BMI and BMI change were assessed using structural equation modelling.

Results: Additive genetic effects explained a moderate proportion of variance in BMI fluctuation (a2 = 0.20 in men, 0.29 in women), with the remainder attributable to unique environmental effects (e2 = 0.80 and 0.71). No evidence of sex-specific genetic effects was found. Genetic contributions varied across life stages, peaking in men at ages 31–50 (a2 = 0.36) and in women at 51–65 (a2 = 0.36). Higher baseline BMI was associated with greater fluctuation, whereas greater directional BMI change was associated with less fluctuation. Higher BMI at ages 18–30 was associated with greater subsequent BMI fluctuation across later adulthood stages. Previous associations were driven by both genetic and unique environmental factors.

Conclusions: BMI fluctuation was predominantly explained by unique environmental effects, with moderate heritability that varied across life stages with no evidence of sex-specific genetic effects. Genetic and environmental factors contribute to how BMI fluctuation is associated with BMI change.

N-(1,3-Dimethylbutyl)-N′-phenyl-p-phenylenediamine-quinone, commonly called 6PPD-quinone (6PPD-Q), is an environmental transformation product formed when the antiozonant and antioxidant 6PPD, widely used in tires, reacts with ozone. 6PPD-Q enters the environment through tire wear particles and roadway runoff and has been linked to acute toxicity in aquatic and mammalian models, raising concerns about its potential effects on human health. However, its impact on human immune responses remains unclear. This study aimed to investigate the effects of 6PPD-Q on inflammasome formation and activation, as well as on inflammatory signaling, in differentiated THP-1-ASC-GFP macrophage-like cells. Specifically, this study focused on NLRP3 inflammasome-associated ASC speck formation and interleukin-1 beta (IL-1β) secretion after inflammasome activation. ASC speck formation was assessed by live-cell imaging every 30 minutes for 12 hours after exposure to 0, 1, 10, 50, 100, or 200 ng/mL 6PPD-Q, while IL-1β secretion was measured by ELISA at 1, 3, and 12 hours in cells exposed to 0, 10, 50, or 100 ng/mL 6PPD-Q. No statistically significant differences in IL-1β release were observed between 6PPD-Q-treated cells and untreated controls at any time point. Similarly, live-cell imaging did not reveal a significant dose-dependent effect of 6PPD-Q on ASC speck formation. High variability and outliers were noted in both assays, potentially limiting the detection of subtle effects. Overall, these findings suggest that, under the conditions tested, 6PPD-Q does not significantly alter IL-1β secretion or ASC speck formation in human macrophage-like cells. The observed variability underscores the need for further studies with improved experimental consistency, increased biological replication, and assessment of additional functional and mechanistic endpoints, such as caspase-1 activation, cell viability, or mitochondrial function, to clarify the immunomodulatory potential of 6PPD-Q in human cells.

Freshwater ecosystems are increasingly threatened by anthropogenic stressors such as pesticide contamination, agricultural runoff, pollution, and climate change. Freshwater unionid mussels play a critical ecological role as filter feeders and bioindicators of aquatic ecosystem health; however, genomic resources for these organisms remain limited due to the large and repetitive nature of their genomes. This study aimed to sequence and assembled the genome of Anodonta anatina using Oxford Nanopore long-read sequencing technology while evaluating the effectiveness of two high-molecular-weight (HMW) DNA extraction methods. Fresh specimens of A. anatina were collected from Lake Vingsjön, Sweden, foot and gill tissues were processed using the Monarch HMW DNA Extraction Kit and the PacBio Nanobind PanDNA protocol. DNA quality and integrity were assessed using Qubit fluorometry, NanoDrop spectrophotometry, and pulsed-field gel electrophoresis (PFGE). Sequencing libraries were prepared using the Oxford Nanopore Ligation Sequencing Kit V14 and sequenced on MinION R10.4.1 flow cells. Sequencing quality was evaluated using NanoPlot, genome assembly was performed using Flye, and assembly quality was assessed using QUAST. PFGE and NanoPlot analyses demonstrated superior preservation of high molecular weight DNA and improved long-read distributions in PacBio-derived samples compared with Monarch-derived samples. The PacBio assembly generated a draft genome of approximately 1.38 Gb with an N50 value of 34,332 bp and a maximum contig length of 381,012 bp, whereas the Monarch assembly was substantially smaller and more fragmented. Although the assemblies remained incomplete relative to the estimated 3 Gb genome size, the study demonstrates the feasibility of nanopore-based genome assembly in A. anatina and highlights the importance of extraction methodology for long-read sequencing performance. These genomic resources provide a valuable foundation for future studies in freshwater mussel genomics, conservation genetics, and aquatic ecotoxicology.

Currently, Artificial intelligence (AI) is increasingly becoming embedded as a normality in a growing amount of work positions and is reshaping how professionals perform tasks, process information and create value. While previous research has largely focused on organizational outcomes, less attention has been given to how AI is experienced within specific professional roles, where both analytical and relational work are central.

This study explores how AI reshapes the everyday work and perceived value creation of business development professionals (BDPs). Adopting a qualitative-dominant mixed-methods design, the study combines semi-structured interviews with a mixed-method survey in order to examine how AI integration influences task execution, competence requirements, and professional role expectations. The analysis is guided by the EPOCH framework and Role Theory and follows an abductive thematic approach.

Our findings suggest that AI has become deeply embedded in the daily workflows of business development professionals, functioning not only as an efficiency-enhancing tool but also increasingly as both a cognitive and strategic support system. While AI tools reduce time spent on repetitive and information-intensive tasks, the findings indicate that it simultaneously increases the importance of several human-centered capabilities such as judgment, contextual understanding, relationship-building, and communication. AI appears to shift value creation from operational execution toward interpretation, validation, and relational anchoring of AI-generated outputs.

The study further suggests that business development work is becoming increasingly defined by the ability to balance AI-supported efficiency together with human reflection, trust, and sensemaking. The findings support perspectives viewing AI primarily as augmenting rather than replacing human work.

Theoretically, the study contributes with time-relevant insights into how AI reshapes competence requirements and perceived value within a boundary-spanning role. Practically, the study highlights the growing importance of combining AI literacy with interpersonal and interpretive capabilities in contemporary business development work.

AI-powered visual inspection is increasingly used in manufacturing quality control, yet its practical implementation in small and medium-sized enterprises raises questions about data availability, production scale, and the internal expertise needed to sustain such systems. This thesis investigates how a data-centric AI vision system can be implemented in a resource-constrained manufacturing context, where labelled defect data, specialist AI knowledge, computing resources, implementation time, and long-term maintenance capacity are limited, using Skandia Elevator as the industrial case company.The study builds on a longer collaboration between the University of Skövde and Skandia Elevator, where previous work established parts of the physical inspection infrastructure and an initial proof-of-concept system. The present project focuses on three connected aspects of implementation: sustainable knowledge transfer to company personnel, the trade-off between model quality and inference speed, and the suitability of anomaly-detection model candidates for temperature-sensor image inspection. The project therefore treats the AI system not only as a technical artefact, but also as an organisational learning process that must be understandable, maintainable, and useful after the thesis project ends.This thesis explains how data-centric AI can be implemented in a manufacturing SME context. The findings show that model performance depends not only on the algorithm, but also on data quality, user understanding, and organisational ownership. In resource-constrained settings, systematic work with data collection, labelling, retraining, and knowledge transfer may be more important than using the most complex algorithm. The study provides practical guidance for manufacturing companies that want to introduce AI-based visual inspection while maintaining local competence and long-term system ownership.

Dengue virus infection is a major global health problem that can cause a wide range of clinical outcomes, from mild febrile illness to severe dengue associated with plasma leakage, haemorrhage, and organ dysfunction. The mechanisms underlying disease severity are not fully understood, particularly regarding immune and metabolic responses in different immune cell populations. Therefore, the aim of this study was to investigate transcriptional and metabolic alterations in monocytes and NK cells during dengue infection using a systems biology approach. Publicly available single-cell type RNA sequencing data from dengue patients were analyzed. A total of 2449 monocytes and 802 NK cells were included in the study. Differential gene expression analysis, pathway enrichment analysis, genome-scale metabolic modelling and flux balance analysis were performed to compare control, mild, and severe dengue conditions. Context-specific metabolic models were generated using transcriptomic profiles of the cells and analyzed using flux balance analysis. The results showed substantial transcriptional and metabolic alterations associated with dengue infection, particularly in monocytes. Severe dengue conditions showed a higher number of differentially expressed genes and stronger pathway dysregulation compared with mild infection. Several immune-related and metabolic pathways, including oxidative phosphorylation, cytokine signalling, NOD-like receptor signalling, and lipid metabolism, were significantly altered. Metabolic modelling further revealed major changes in transport reactions and lipid metabolic pathways, especially in severe monocytes. Compared with monocytes, NK cells showed more moderate transcriptional and metabolic alterations. In conclusion, this study demonstrates that dengue infection induces significant immunometabolic changes in immune cells, with monocytes showing stronger metabolic reprogramming during severe disease. The integration of transcriptomics with genome-scale metabolic modelling provided systems-level insight into host–virus interactions and highlighted potential metabolic pathways associated with dengue severity. These findings contribute to the understanding of dengue pathogenesis and may support future research into therapeutic strategies targeting host metabolism.

Staphylococcus aureus is a common opportunistic pathogen and a major cause of foodborne intoxication worldwide with high ecological plasticity and adaptability to diverse ecological niches. Its accessory genome, contains extensive virulence determinants that contribute to strong metabolic and physiological flexibility, enables it to persist under complex and adverse conditions encountered throughout the food supply chain including during food production and preservation. The aim of this study was to investigate the distribution of adhesion-associated genes with a particular focus on the fibronectin-binding protein genes (fnbA and fnbB), and their potential role in the ecological adaptation of Staphylococcus aureus across the food supply chain. Comparative genomic analyses were performed using publicly available isolate genomes and metagenome-assembled genomes. A total of 4,503 genomes were analysed and annotated using comparative pangenomics, core-genome phylogenetics and phylogeny-aware genome-wide association analyses to identify niche associated genomic patterns among 15 adhesion-associated virulence determinants with a focus on fibronectin-binding proteins. The samples were unevenly distributed among ecological niches and the dataset was biased towards milk-associated samples. Nevertheless, the pangenome contained approximately 80% accessory genes with a Heap’s law α value of 0.87, supporting an accessory-rich open pangenome. Significant associations were identified between different ecological niches and several adhesion-associated genes, including sdrC, sdrD, sdrE, icaA, icaD, icaR, ebpS, clfA clfB, fnbA and fnbB. These results provide evidence for possible candidate niche-associated genomic patterns. However, the broad distribution of fnbA and fnbB suggests that these genes are biologically important but cannot be considered simple niche-specific markers or direct markers of transmission routes. Overall, the findings suggest that adhesion association genes may contribute to the ecological flexibility of S. aureus across the food supply chain, although they do not confirm direct transmission, functional adaptation or causality. Further studies are required to validate these associations and to evaluate the potential role of fibronectin-binding proteins in the persistence and adaptation of S. aureus in food-associated environments.

This thesis was carried out in collaboration with Metsä Tissue AB. The purpose of the work has been to troubleshoot the existing problems that occur in their conveyor system and then implement one or more solutions depending on the severity of the problem. Three methods have been used as support: Inductive, DECSAR and DMAIC. The inductive method has been used at the beginning of the work to gain an understanding of the problem. DECSAR has been used to structure the troubleshooting process and DMAIC has been used as support for implementing the solution and checking and verifying. To implement the solutions, programming has been carried out in Siemens Simatic Manager (Step 7) according to the PLC standard IEC 61131–3. 

The result was a more stable and robust production, less wear and tear and lower energy consumption on the conveyor system.

De senaste åren har miljöförändringar blivit ett allt större hot mot jordens ekosystem, med ständig närvaro av mänskliga aktiviteter och en varmare planet är detta förödande för de komplexa organismer som lever i vatten- och havsmiljöer. För att på bästa sätt kunna ta beslut om åtgärder som skulle kunna förbättra den marina miljön och följa upp tidigare åtgärder behövs data att basera besluten på. I samband med utvecklingen av informationstekniken har projekt som involverar medborgare blivit ett alltmer diskuterat ämne och anses vara ett möjligt alternativ till traditionella provtagning på grund av sina ekonomiska fördelar. Marin citizen science har ett brett spann användningsområden och skulle potentiellt kunna underlätta provtagningar och observationer av den marina miljön. Under arbetet används en systematisk litteraturstudie för att samla och analysera vetenskapliga artiklar som behandlar olika typer av projekt inom marin citizen science med fokus på sportdykare. Syftet med studien är att förstå hur digitala lösningar kan möjliggöra insamling, hantering och analys av miljödata med hjälp av sportdykare samt vilka krav de måste uppfylla för att öka kunskapen om aktuell miljöpåverkan i svenska vatten- och havsmiljöer. Analysen visar att fler av de insamlade artiklarna som behandlar den marina miljön med hjälp av citizen science använder sig av olika typer av digitala verktyg och informationssystem.

The parasitic interaction between brown trout (Salmo trutta) and the freshwater pearl mussel (Margaritifera margaritifera) is complex where the two species through mutual adaptations are maximising their fitness in an evolutionary arms race. While M. margaritifera is obligatorily dependent on salmonids for its reproduction, a high grade glochidia infection can result in significant physiological costs for the host in the form of impaired respiratory capacity, reduced swimming performance and inhibited foraging. In response to this parasitic burden ectothermic organisms can utilize behavioural thermoregulation as a defence mechanism to optimize their physiological state. To determine if infected S. trutta can alter their thermal preference a study was conducted using a temperature preference chamber with a controlled temperature gradient. The aim was to isolate the importance of temperature for fish behaviour independent of external factors such as predation or flow. The results revealed a significant difference in temperature selection where infected individuals exhibited behavioural fever by seeking out warmer water compared to uninfected individuals. This indicates that at a high infection rate there may be a threshold value where the burden from the glochidia infection becomes more costly for the individual than the higher metabolic expense of remaining in warmer water. To accelerate glochidia development through heat may further increase the risk of trophic mismatch where juvenile mussels are forced to conclude their parasitic phase before optimal feeding conditions prevail in the stream. To strengthen the two species survival and dispersal capability it is vital to preserve thermal refugia and shaded riparian zones as well as to prioritize connectivity and habitat heterogeneity in stream restoration. 

This thesis examines how marketing is enacted in small indie game studios in Västra Götaland, Sweden, where formal marketing education, dedicated marketing departments, and structured marketing processes are often limited or absent. While entrepreneurial marketing literature has traditionally focused on orientations, capabilities, and behavioural traits such as innovativeness, proactiveness, and opportunity recognition, less attention has been given to how marketing is performed in everyday organizational life (Hills & Hultman, 2011; Kraus et al., 2010). To address this gap, the study adopts a practice-theoretical perspective and integrates insights from strategy-as-practice, marketing-as-practice, and entrepreneurship-as-practice.

A qualitative research design was employed, using two-week work diaries and follow-up semi-structured interviews with ten indie game studios employing between one and five members. The data were analysed abductively through thematic analysis, with attention to practices, praxis, and practitioners (Whittington, 2006; Reckwitz, 2002).

The findings indicate that marketing in indie studios is embedded within everyday work rather than organised as a separate function. Recurring practices included community engagement, content creation, platform management, analytics monitoring, networking, internal coordination, and development activities later repurposed for promotional purposes. Participants frequently engaged in marketing-related work without explicitly identifying it as marketing, suggesting that marketing knowledge was often tacit, experiential, and developed through practice.

Based on these findings, the thesis proposes the concept of Entrepreneurial Marketing as Practice (EMaP), which conceptualises marketing in SMEs as an emergent set of value-creating practices enacted under conditions of resource constraints and limited formal expertise. The study contributes theoretically by extending entrepreneurial marketing into a practice-based framework, empirically by identifying concrete marketing practices in indie studios, and methodologically by demonstrating the usefulness of diary methods for capturing everyday organizational activity.

Sepsis is a life-threatening organ dysfunction caused by a dysregulated host response to infection. As the immune system overreacts, the patient outcome can be detrimental, where every hour of delay substantially increases the mortality rate. Therefore, early diagnosis and treatment are critical for patient outcomes. The project aimed to obtain high quality and quantity bacterial DNA to be used in metagenomic sequencing for identification of pathogens and antimicrobial resistance genes, ultimately contributing to earlier diagnosis of sepsis and effective treatment. DNA extractions were performed manually and semi-robotically to compare the outcome, turnaround time, and hands-on time, considering time-sensitivity in clinical applications. Whole blood from healthy human donors were spiked with whole bacteria, bacterial DNA, and a microbial community DNA standard, where the DNA were extracted before sequencing with the MinION device. Raw sequencing data was analysed through the EPI2ME software which enabled taxonomic classification and antimicrobial resistance gene detection. Obtained results demonstrated the success of host exclusion via adaptive sampling, with high quality of data. Negative controls were expected to only contain human DNA, but the presence of different bacterial species were detected, and several unexpected species were also present throughout most DNA samples. However, expected bacterial species were still detected throughout the DNA samples, with mdf(A) gene present in two. Detection of the Escherichia coli spike-in of 10 CFU which mimics the limited amount naturally found in sepsis patients’ blood was achieved. Therefore, the project showed the potential of nanopore sequencing in detecting pathogens and antimicrobial resistance genes in blood samples, while achieving earlier sepsis diagnosis. 

VPREB3 has previously been identified as a useful diagnostic marker, however its precise biological function is still unknown. Previous research has attempted to propose a function however, due to a lack of phenotype in VpreB3-deficient mice (unpublished) and a lack of robust reagents, i.e. monoclonal antibodies recognising VPREB3, it is still unknown. Its intracellular localisation and restricted expression have made it difficult to produce large amounts of the protein to be used in the production of monoclonal antibodies. This project attempted to address the production of soluble protein by creating a vector expressing a potentially soluble VPREB3 accomplished by replacement of the protein’s signal peptide with that of VPREB1, coupled with a HaloTag. The study sought to express the modified VPREB3 coupled with the HaloTag, following different transfection approaches in HEK 293T cells, with the HaloTag acting as a proxy for expression. Fluorescent labelling using the R110Direct Ligand and fluorescent imaging of the cells detected the expression of the HaloTag, indicating HaloTag expression from the transfection vector. However, Sanger Sequencing revealed a lack of the modified sequence within the vector and as such a lack of VPREB3 expression. The analysis of the data from the images proved a significant difference in transfection efficiency between the Invitrogen-7.5 and Alternative-7 methods, at a p = 0.05 significance level. The findings in this study contribute to the methods required for establishing a secretory VPREB3 expression system, which can be used to facilitate production of monoclonal antibodies specific for the protein and thereby aid in identifying its function.

Background: Obesity is a growing public health problem and a chronic condition that can have serious consequences, including cardiovascular disease, stroke, and type 2 diabetes mellitus. Patients with obesity often experience guilt and shame and feel prematurely judged by healthcare professionals because of their condition. There are various types of care available for obesity, in which the nurses' role and health promotion efforts are central. Purpose: To examine how nurses provide support to adult patients with obesity in outpatient care. Method: The study was conducted as a general literature review based on eleven scientific articles, including eight qualitative and three quantitative studies. Results: Three themes have been identified: approaching a sensitive topic, the complex task of caring for patients with obesity, and barriers to effective obesity care. The nurse's role in encounters with patients with obesity is multifaceted and influenced by the nurse's ability to establish a trusting relationship in which communication is well adapted. Conclusion: The nurse's supportive role in the management of obesity is complex. A person-centered approach, seeing the patient as a whole, and the ability to motivate the patient are particularly beneficial in this work. 

Introduction: Parenting support is an important part of preventive public health work and aims to strengthen parents and promote children’s health and living conditions. Despite national strategies and positive effects of parenting support programs, challenges remain in reaching parents. There is limited knowledge about what motivates parents to enroll in order to take part in the support and how communication can best reach the target group.

Aim: The aim was to investigate how group leaders perceive parents’ motivation to enroll in parenting support and their perspectives on communication efforts that reach the target population and the role of digital parenting support.

Method: The study was conducted using a qualitative approach and an exploratory design. Eight certified ABC group leaders were interviewed through semi-structured individual interviews. The material was analyzed using reflexive thematic analysis based on an abductive approach, and Self-Determination Theory was used as the theoretical framework.

Results: The results showed that motivation to enroll was influenced by needs for relatedness, competence, and autonomy. Parents were motivated by the opportunity to share experiences, develop in their parenting role, and strengthen the relationship with their children. At the same time, the group leaders described how social norms, practical barriers, and organizational conditions influenced which parents they reached with the support. Resourceful parents were perceived as easier to reach with the support, whereas groups with greater needs were often more difficult to reach.

Conclusion: The study shows that motivation to enroll in parenting support cannot solely be understood as an individual process, but also needs to be viewed in relation to social, structural, and organizational conditions.

Background: Credit assessment constitutes a central part of banks’ credit decisions, as it forms the basis for evaluating the borrower’s risk and repayment capacity. An important source of information in this process is the annual report, which provides banks with insight into the company’s financial position and development. Previous studies have examined the importance of credit assessment in banks’ credit decisions, but there is limited research focusing on how Swedish banks assess real estate companies in connection with lending. This particularly concerns which factors are considered most significant and how these factors are interpreted in relation to risk and repayment capacity. This thesis aims to contribute to an increased understanding of how Swedish banks’ credit assessment of real estate companies is conducted. 

Purpose: The purpose of this thesis is to examine how banks use real estate companies’ accounting information, such as annual reports, to assess risk and future repayment capacity. Furthermore, the thesis aims to identify which accounting-related factors are considered most significant in banks’ credit assessments of real estate companies. 

Method: This thesis applies a qualitative approach, using semi-structured interviews. In addition, the thesis is based on an abductive approach, which means that the study draws on both empirical material and theoretical information. This was done to ensure that the purpose of the study is appropriately reflected in the collected material. 

Conclusion: Accounting information is a central part of banks’ credit assessments of real estate companies, as it provides insight into factors such as cash flow, indebtedness, and profitability measures. In addition, accounting information presents the collateral values that banks may use to secure their lending.  

Purpose

This study explores GAI's impact on business models of marketing agencies through the theoretical lenses of the Resource-Based View (RBV) and Dynamic Capabilities (DC) theories. 

Methodology

The study adopts a qualitative exploratory approach, using semi-structured data collection and thematic analysis. Data were collected from 11 professionals working in SME-sized marketing agencies in two countries (Sweden and Canada). 

Findings

The findings indicate that GAI's impact on agencies' business models varies. There is a common observation that GAI is not a VRIO resource that provides sustained competitive advantage on its own. Instead, agencies' unique VRIO resources, such as professional expertise and internal business processes, are used alongside GAI to differentiate in the market. The study also suggests that GAI alone cannot provide a sustained competitive advantage, and that proper BM reconfiguration (or BMI) is a significant part of GAI adoption. 

Background: Research on how individuals are affected by monetary reward systems has developed for nearly a century, with quantitative research dominating the field. While the effects of reward systems on performance and organizational outcomes have been widely studied, employees’ own experiences and perceptions of these systems have received less attention. In a work environment where increasing emphasis is placed on the human role within organizations, it is important for companies to understand how employees perceive motivation, collaboration and perceive fairness within monetary reward systems.

Purpose: The purpose of this study is to examine how motivation, collaboration, and perceived fairness are affected by monetary reward systems. This was conducted through qualitative interviews at two Swedish companies, each applying a different type of reward system.

Method: The study was conducted using a qualitative research approach. The participants consisted of employees whose compensation was influenced by monetary reward systems at two Swedish companies. Empirical data was collected through semi-structured qualitative interviews.

Results and conclusion: The positive and negative effects of reward systems identified in previous research, such as free riding, positive interdependence, and negative interdependence, were only found to a limited extent in this study. Motivation, collaboration, and perceived fairness appeared to be only marginally influenced by monetary reward system. However, regardless of which system the respondents worked with, all participants expressed a preference of retaining their current reward system. Furthermore, the respondents’ experiences of working within these reward system showed limited alignment with the predominantly quantitative international research within the field.

This study examines how accounting choices related to research and development (R&D) are influenced by buisnesscycles in Swedish listed innovation companies. The study is based on the premise that accounting for R&D under IAS 38 involves significant managerial discretion, which may enable strategic accounting choices and influence how companies are perceived by external stakeholders. Previous research shows that both earnings management and R&D investments are affected by the economic environment, however there is a lack of studies examining how these phenomena interact in practice. The purpose of this study is therefore to analyze both changes in R&D accounting choices over a business cycle and how these changes are communicated and justified. The framework is based on Positive Accounting Theory, disclosure theory, and theories of earnings management, which together are used to understand both the incentives behind accounting choices and how these are communicated. These theories are integrated into an analytical model that forms the basis for the empirical analysis. The study applies a longitudinal research design using methodological triangulation where quantitative and qualitative content analysis is used to examine communication in annual reports. The data collection is based on six Swedish listed innovation companies over the period 2018–2025. The results indicate that variations in R&D accounting choices occur across different phases of the business cycle suggesting that accounting practices may be influenced by changing economic conditions, Furthermore the analysis shows that companies vary in communication often adapting to the economic environment. The study concludes that R&D accounting choices do not solely reflect underlying economic conditions but may also be influenced by strategic incentives and communication strategies. The study therefore contributes to a deeper understanding of the relationship between accounting choices, business cycles and corporate communication and highlights the need for further research in this area. 

This study explores how generative AI search tools are used in early-stage supplier identification and shortlisting within business-to-business sourcing. The purpose of the study was to understand how sourcing professionals use and experience generative AI tools during supplier search activities. It also examines how these tools influence supplier visibility, information evaluation, and early-stage sourcing decisions. To understand this, the study explores factors such as perceived usefulness, trustworthiness, supplier visibility, and digital information structures, in sourcing environments.

The study was conducted through a qualitative research method, where eleven professionals from different industries were interviewed through semi-structured interviews. By operationalizing theoretical concepts, the study was able to explore how sourcing professionals use generative AI during supplier search and evaluation processes. Through an analysis of both theoretical and empirical material, the study identified both advantages and limitations of using generative AI in sourcing activities.

The study then concludes that generative AI tools are mainly used as supportive research and information tools rather than as final decision-makers. The main advantages identified were improved efficiency, reduced research time, faster market overviews, and structured summaries of supplier information. Generative AI is described as particularly useful during the earliest stages of supplier search when identifying supplier categories, comparing alternatives, and gathering market research. At the same time, limitations were also identified, including hallucinations, outdated information, lack of transparency, and difficulties understanding how AI systems prioritize supplier information.

The service ecosystem is an arguably new concept with minimal research on the function of feedback between actors. Furthermore, the service ecosystem concept shares many similarities with the gaming industry, but existing research on game development from a business perspective is limited even though co-creation and feedback is a central part of it. This paper explores this overlap and investigates how feedback functions, and is facilitated, as part of a service ecosystem within the gaming context by conducting a qualitative study with focus on game developers perspective. 

The study explores this topic based on semi-structured interviews conducted in Sweden. The transcripts of the interviews are analysed using the Deductive Qualitative Analysis (DQA) method. In addition to this, operationalising terms and a theoretical lens consisting of relevant concepts are used for the analysis. 

The findings show that feedback is used between players and game developers by creating a continuous feedback-loop, which allows a constant co-creation of the services value. Findings also show that feedback is facilitated in multiple ways, such as providing tools for the consumers to facilitate further resource integration, usage of social media to improve the resource integration between actors within the ecosystem, as well as catering to consumers' motivations. Additionally, emergent findings provide further insight on how co-destruction is a part of feedback and how it exists within a service ecosystem. 

Lastly, the study provides a deeper understanding to the existing literature on service ecosystems, as well as provides insight on how feedback functions within a service ecosystem.

Background: This study aimed to investigate the influence of BMI, age, and socioeconomic status on QoL among BC patients in Bangladesh.

Methods: This cross-sectional study was conducted among 200 female breast cancer patients who had completed treatment or were undergoing follow-up at the Shanti Oncology and Palliative Care Foundation, Dhaka, Bangladesh, between May 2023 and July 2024. Participants were interviewed using validated European Organization for Research and Treatment of Cancer (EORTC) questionnaires. The QLQ-C30 for general cancer-related QoL and the BR23 module specific for breast cancer. Key independent variables included variables age, family income and BMI, assessed in relation to cancer profile, global health scores, C30-BR23 functional domains, and symptom scales. Data analysis involved descriptive statistics, assessment of normality using the Shapiro-Wilk test, and inferential analyses using logistic and linear regression models to evaluate associations between independent variables with cancer profile and QoL scale.

Results: Among 200 breast cancer patients, the mean age was 49.5 years (SD 9.02), with the majority overweight (52.5%) and housewives (74%). Most had Stage II (53%) cancer, 58.5% underwent mastectomy, and 98% received chemotherapy. Global health score (GHS) was 42.44 (SD 12.46). Regression analyses identified age and family income as the most significant factors associated with QoL in this study. Older age was significantly associated with lower GHS score (β = –4.68; 95% CI: –6.68 to –2.67; p < 0.001), as well as higher insomnia (β = 6.45; 95% CI: 0.45 to 12.46; p = 0.035) and dyspnea (β = 5.64; 95% CI: 0.45 to 10.84; p = 0.033) scores. Higher family income was significantly associated with localization of BC (AOR = 1.51, 95% CI: 1.02–2.23, p = 0.038), metastasis (AOR = 2.03, 95% CI: 1.28–3.22, p = 0.003), undergoing mastectomy (AOR = 1.55, 95% CI: 1.07–2.25, p = 0.020), and receiving radiotherapy (AOR = 11.92, 95% CI: 2.34–60.63, p = 0.003). Lower family income was significantly associated with increased financial difficulties (β = –11.88; 95% CI: –16.76 to –7.00; p < 0.001), poorer future perspectives (β = –5.66; 95% CI: –10.15 to –1.16; p = 0.014), and worse breast and systemic therapy side effects (β = –5.44; 95% CI: –8.94 to –1.94; p = 0.002; β = 4.07; 95% CI: 1.23 to 6.90; p = 0.005, respectively). BMI indicated limited associations, which include increased dyspnea (β = 1.40; 95% CI: 0.15 to 2.64; p = 0.028) and decreased diarrhea scores (β = –2.11; 95% CI: –3.09 to –1.13; p < 0.001).

Conclusion: The overall QoL score was poor among Bangladeshi patients. The association of BMI with C30 and BR23 was limited. In this study, age and family income were statistically significant predictors of QoL. Further studies with larger sample sizes are recommended to better explore the association between higher BMI and QoL among BR patients.

Post-Quantum Cryptography (PQC) has emerged as a means to protect encryption from Quantum Computing (QC). Quantum Computing poses a threat towards classical cryptography because of its enhanced capability to solve complex mathematical problems quickly. To combat this, NIST has developed three standards for PQC: FIPS 203, FIPS 204, and FIPS 205. This study compares how the algorithms defined in FIPS 204 and FIPS 205, ML-DSA and SLH-DSA, compare to a baseline classical algorithm, Ed25519. 

Findings show that both ML-DSA and SLH-DSA offer slower signing times and verifying times compared to Ed25519. The algorithms also consume different amounts memory which can have an impact on future implementations. Future directions include the addition of future algorithms to improve the generability, a closer look at the ecological impacts as well as examining the component bottleneck in order to get a better understanding of how a switch towards these algorithms may impact the future. 

This study investigates how selected hyperparameter configurations affect the performance of a convolutional neural network when classifying LEGO bricks across different image datasets. The study was conducted using two LEGO brick image datasets. Dataset 1 contained rendered and photographic images and was used to evaluate different configurations of image size, batch size, learning rate, and class weights. Dataset 2 consisted of high-resolution photographs captured by the authors and was used to investigate how image size affected CNN model performance when the original image resolution was higher and more consistent. 

Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) is a novel method of targeted gene modification within the cell's genome using guide RNAs (gRNA). One of CRISPR applications is using single guide RNA (sgRNA) directing the unit to a specific genome location, inhibiting the function of the gene. This mechanism has been used recently in cancer screening, where individual genes can be targeted across defined genomic sites. Due to variations in experimental protocols and sgRNA targeted genes, there is no direct method by which to compare the results across datasets. Hence, in this study, two edgeRbased pipelines were constructed that would allow for direct comparison between two genome-wide CRISPR screening datasets of colorectal carcinoma, the Wellcome Sanger Institute’s (Sanger) and the Broad Institute's DepMap (Achilles) datasets. The Achilles results from edgeR were also compared to Model-based Analysis of Genome-wide CRISPR/Cas9 Knockout (MAGeCK) for comparison and validation of the edgeR results. The pipelines successfully identified overlapping genes across datasets associated with colorectal carcinoma, such as RAS related nuclear protein (RAN) connected to the PI3K/Akt/mTOR and Ras-Raf-MEK-ERK pathways in addition to identifying RDX and MATN2 regulatory genes in both cell lines. From the 787 overlapping genes the majority had a negative cross dataset correlation in logarithmic fold change (LFC), with only 36 genes sharing concordance in LFC. In addition, the Achilles MAGeCK MLE results identified a small number of essential genes, which failed to overlap with the essential genes identified in the edgeR pipeline. 

This thesis was conducted at Automations Teknik in Lidköping (Atlid) and describes the development of an automated control system for strength testing of concrete pipes. The purpose of the project was to create a system that enables accurate, repeatable, and efficient testing by automating the operational steps. The system was developed using Siemens TIA Portal V21 in accordance with the international standard IEC 61131-3 to achieve a standardized, modular, and scalable software structure tailored for industrial presses. To manage various test types and inputs, an HMI was integrated, allowing the operator to input the desired operating parameters.

The result is a program that handles breaking limits, test sequences, and the presentation of results without manual intervention. The implementation of the V-model throughout the project ensured stable and repeatable logic. Furthermore, this work demonstrates that automation in industrial testing contributes to increased process stability, a reduced risk of human error, and enhanced efficiency. The report also highlights challenges encountered along the way, such as hardware configuration and the simulation of external components.

Background

The transition from K2 to K3 involves increased requirements regarding competence and resources. K3 is a more complex and principles-based accounting framework that provides a fairer representation of a company’s financial position. The transition from K2 to K3 also entails increased demands for appropriate competence, time and resources. K2 is recommended for smaller companies, as it is a rules-based accounting framework that reduces the risk of future challenges requiring adjustments.

Purpose

The purpose of the study is to describe how auditors and accounting consultants experience the transition from K2 to K3. Particular focus is placed on driving factors and practical challenges. The most complex parts of the process are also highlighted in the study.

Theoretical Framework

The study is based on previous research regarding the transition from a rules-based to a principles-based accounting framework.

Method

A qualitative method has been applied in which semi-structured interviews have been conducted with both accounting consultants and auditors.

Conclusion

The study’s conclusion show that the transition from K2 to K3 is driven by legal requirements, but also by professional norms in voluntary transitions. There is no standardized model for the transition process itself, however, recurring phases emerge during transitions. These phases show that recalculation and opening balances are central elements. The transition can become complex and resource demanding. The study has identified challenges related to time consumption, competence, communication and financial aspects. Early planning, appropriate competence and clear communication are key factors for a successful transition.

Limb-girdle muscular dystrophies (LGMDs) are disorders with an important clinical heterogeneity, usually involving proximal limb muscles. One subtype, LGMD R16 (LGMD 2P), is an autosomal recessive condition caused by pathogenic variants in DAG1, with clinical presentations ranging from mild to extremely severe forms. DAG1 is responsible for producing dystroglycan, an essential complex in the muscular protein network. Following translation, dystroglycan is cleaved into alpha-dystroglycan, which undergoes glycosylation and acts as a sarcolemmal receptor for extracellular proteins, and beta-dystroglycan, which connects to dystrophin. In recent years, heterozygous nonsense or frameshift DAG1 variants have been linked with asymptomatic hyperCKemia (increased serum creatine kinase levels) or mild muscular phenotypes characterized by fatigability and myalgia. Here, we describe a Romanian family comprising four affected individuals (one father and three sons) carrying the heterozygous missense DAG1 variant NM_004393.6:c.887G>A, NP_004384.5:p.(Gly296Asp) and showing a mild–moderate muscular phenotype similar to previous cases of DAG1 haploinsufficiency. Three of the affected individuals exhibit myopathic changes in muscle biopsies (increased fiber size variability, internalized nuclei, and regenerating fibers), while two demonstrate reduced alpha-dystroglycan glycosylation in muscle tissue. Atomic force microscopy findings in myoblasts from one patient showed a significantly lower stiffness compared to controls. These findings align with prior reports and further support the pathogenicity of this variant.

Background: Delivery of interventions within student mental health services has undergone considerable digital transformation in recent years. Traditional face-to-face meetings are being substituted with autonomous digital tools with evident advantages in terms of accessibility and scalability. Despite an increasing array of digital options, there is also a growing recognition that digital tools offer limited effectiveness without some degree of human support. For example, for mental well-being, completely digitally delivered interventions show approximately half the effect sizes of interventions delivered in a traditional format. Blended forms of delivery that use both digital advantages and recognized effects of human contact are therefore promising. Hitherto, the effects of blended delivery have not been evaluated for mental well-being. Hence, investigating how digitalization in intervention delivery impacts intervention effects on mental well-being is important. This is especially relevant among emerging adults enrolled in higher education, going through a critical, transformative life phase.

Objective: This systematic review and meta-analysis will primarily aim to investigate differences in effect due to the degree of digitalization in the mode of delivery of interventions on mental well-being among higher education students.

Methods: This work will adhere to the Cochrane Collaboration methodology, and results will be reported according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. A systematic literature search will be conducted across 9 databases (Scopus, MEDLINE, PubMed, PsycINFO, ERIC, CINAHL, Web of Science, Cochrane, and International Clinical Trials Registry Platform). The population, intervention, comparator, and outcome framework will inform both the search strategy and eligibility criteria. For inclusion, studies should be randomized controlled trials investigating the effect of individually delivered interventions on positive affect or life satisfaction among mentally healthy higher education students aged 18 to 29 years. Studies will be independently screened, and data will be extracted, including the standardized mean difference as the effect measure. Risk of bias assessment will be conducted using the Cochrane Risk of Bias 2 instrument. The Hartung-Knapp-Sidik-Jonkman method for random effects meta-analysis will be applied, and study biases will be analyzed by funnel-plot assessment and Egger test. Finally, certainty of evidence for positive affect and life satisfaction will be assessed using the Grading of Recommendations Assessment, Development, and Evaluation approach. Results will be presented in a summary of findings table.

Results: The search was finalized in March 2026, generating 6603 records after duplicate removal. Study selection and data extraction were conducted during April 2026, resulting in 41 eligible studies. Risk of bias assessment, data analysis, and manuscript preparation are planned before submission for peer review in August 2026.

Conclusions: The principal findings of this study will highlight differences in effect due to the mode of delivery of interventions on mental well-being among higher education students. This will be relevant for the management of student mental health promotion services.

Trial Registration: PROSPERO CRD420251131950; https://www.crd.york.ac.uk/PROSPERO/view/CRD420251131950

International Registered Report Identifier (IRRID): PRR1-10.2196/88458

This report deals with the design and CE marking of a lifting device intended for lifting steel doors. The work covers large parts of the product development process, from identifying needs and requirements, to the design and development of a prototype model. The work also covers the CE marking process, procedures and application of directives and standards. The work has been carried out in collaboration with Daloc, which manufactures safety-rated wooden and steel doors. The aim of the work was to find a safe solution that can supplement or replace the current lifting device used in Daloc's development center. The proposed solution presented is a further development of one of the lifting devices currently used at Daloc. The solution is based on a clamp that is mounted on the door by means of a screw. An integrated spring is used to indicate that sufficient clamping force has been achieved. The report also presents the requirements that have been achieved according to applicable directives and standards and the further work required for a complete CE marking.

Climate change, pollution and human interference have led to a decrease in biodiversity and species being threatened with extinction. Freshwater mussels are more susceptible to environmental changes than most due to them being filter feeders. Freshwater mussels are widespread and important to numerous and diverse ecosystems around the world. Despite this importance, there is a concerning lack of knowledge and, research on the genomes of different freshwater mussels. The aim of this research is to increase the knowledge of the genome for the duck mussel species Anodonta anatina through performing genome assembly. The investigation into the genome of the A. anatina started with sample preparation, continued with DNA extraction, Nanopore sequencing before genome assembly was performed and evaluated. The results of the experiment were the generation of several genome assemblies through the usage of Flye, which were then evaluated by QUAST and compared to other published genome assemblies. The evaluation of the assemblies showed smaller assembly sizes than what was desired based on an estimated genome size of 1.5-2.9 Gb. The estimated genome size range was based on recent published studies on other freshwater mussel species. With the most generous assumption of a genome size at 1.5 Gb, the largest assembly created in this experiment covers less than 33% and therefore did not yield a satisfactory assembly. However, it can potentially serve as a foundation for future research to improve upon with regards to the techniques used and through learning from problems encountered during the experiment.

Background: Midlife obesity is a well-established risk factor for dementia, whereas late-life obesity has been associated with no increased risk, or even a reduced risk in some studies. However, the joint associations of obesity (body mass index ≥ 30 kg/m2) and metabolic health phenotypes (defined by the presence of hyperglycemia, hypertension and dyslipidemia) with dementia risk are less explored, particularly with regard to age- and sex-related differences. Therefore, we investigated how obesity and metabolic health phenotypes jointly associate with dementia risk and whether this risk differs between midlife (≤ 65 years) and late-life (> 65 years), and sex.

Methods: We analysed data from 11,482 participants, aged 51 to 100 years, from the US Health and Retirement Study (HRS), and 13,068 participants, aged 45 to 90 years, from the Swedish Twin Registry (STR). Cox regression models were used to estimate dementia risk in relation to metabolically healthy obesity (MHO), metabolically unhealthy obesity (MUO), and metabolically unhealthy no obesity (MUNO), relative to metabolically healthy no obesity (reference). Models were adjusted for age, sex, smoking status, and education level. Analyses were stratified by midlife and late-life and conducted in the entire sample and separately by sex.

Results: Metabolically unhealthy status in midlife and late-life indicated increased dementia risk regardless of obesity status, reaching statistical significance for midlife MUNO in females in HRS (Hazard ratio (HR): 1.62, 95% confidence intervals (CI): 1.05–2.49) and in late-life MUNO for the full sample in STR (HR: 1.13, CI: 1.02–1.25) and males in the STR (HR: 1.22, CI: 1.04–1.42). The associations between MUO and dementia risk were not statistically significant, but trends suggested midlife MUO was associated with higher dementia risk. The associations between mid and late-life MHO with dementia risk were also not statistically significant, although the associations showed trends towards lower dementia risk.

Conclusions: Being metabolically unhealthy, especially in midlife, may be associated with increased dementia risk, regardless of obesity status. Mid- and late-life MHO showed no increased risk and suggested potential inverse associations. These findings underscore the importance of evaluating dementia risk in the context of obesity, metabolic health, age and sex simultaneously.

Denna studie undersöker 18 kombinationer av konfigurationer för konsumenthårdvara i flödet för lokal Retrieval-Augmented Generation’s (RAG) påverkan på tids-, prestanda- och kvalitetsmått. Dessa konfigurationer inkluderar varierad semantisk koncentration av dokument, vektorinbäddningsmodeller och språkmodeller. Datamängden som används kommer från ett ärendehanteringssystem för ett open-source-projekt och ger insikter om teknikens användbarhet som hjälpmedel för systemutvecklare.

Frågeställningen som ämnas besvaras lyder ”Vad karaktäriserar ett effektivt och användbart lokalt RAG-flöde för ärendehantering?”

Den använda metoden är en experimentstudie där en mängd frågor, som skapats över källmaterialet, låts besvaras av samtliga kombinationer av konfigurationer. Att mäta både tidsåtgång, prestanda och kvalitetsmått skapar en nyanserad bild av effektivitets-begreppet där flera betydande aspekter tas i beaktande.

Studiens primära fynd är vikten av noggrant genomförd testning innan driftsättning och fullskalig implementation av RAG-flöden, eftersom valet av konfiguration är avgörande för effektiviteten. Semantisk koncentration i flödets datakällas dokument visar föga fördelar med att avvika från rekommenderat värde, även lokalt. Genom att välja rätt vektorinbäddningsmodell kan tid och datorkraft sparas in beroende på hur ofta ett tänkt slutgiltigt system förbehandlar data. Språkmodellen är flödets mest avgörande komponent och källdatan har omfattande påverkan på dess effektivitet. För denna studies datamängd presterar språkmodellen med minst parameterstorlek mest effektivt. Arbetet med studien har inte bara resulterat i mönster kring konfiguration, men också en kodbas som går att återanvända till tester för andra datamängder och konfigurationer för att bredda kunskapen kring lokal RAG ytterligare.

The thesis uses a quantitative experimental research method to find differences in performance, user experience, and developer experience between the front-end frameworks Blazor and React. The thesis aims to provide a comprehensive comparison between Blazor and React, both when using Server -Side (Blazor Server and React Server-Side Rendering) and Client-Side (Blazor WebAssembly and React Client-Side Rendering) methods. The experiment conducted in the thesis consisted of the development of four applications which were identical in all ways but framework and rendering methods ; Blazor WebAssembly, Blazor Server, React Client-Side, and React Server-Side. These applications were designed based on a legacy system currently in use by Volvo Digital Technology & Operations. The experiment consisted of running four different performance tests on each of the applications: Playwright with initial load, Playwright without initial load, Lighthouse, and Google DevTools performance. During the performance tests, the only outlier was Blazor WebAssembly. The performance test results were then used alongside multiple user experience guidelines to establish if the frameworks and rendering methods had an impact on user experience. Here Blazor WebAssembly was again the only outlier. The research also included gathering data from three developer communities to compare the available community support for React, Blazor, JavaScript, and .NET by visiting the communities’ tags and tallying the total posts. This was made to give an insight into the developer's experience with the two frameworks. JavaScript and React together represented 89.95% of the total posts counted. Aside from the experiment, the thesis also touches on the subject of modernization and reasons for an organization to move away from legacy systems, and concluded that legacy systems are expensive, limit organizational flexibility, and finding experts of aged architecture is harder as modern frameworks rise in popularity. 

This case study evaluates the results of data quality efforts in an integrated database at an IT service provider using the data quality framework TDQM and the holistic data management guide DAMA-DMBOK focusing on the data quality knowledge area. Using a method that combines quantitative and qualitative approaches, quantitative data quality metrics were measured using Python before and after the ETL process, and the results were compared with qualitative interviews with stakeholders using the “Inference to the Best Explanation” (IBET) method. The results indicate a dual impact of the ETL architecture: while it functions as a corrective mechanism that significantly improves accuracy (from 13.3% to 82.5% in one system), it simultaneously reduces the completeness of the CustomerName attribute from 100% to 94.3% and 96.8%, respectively. An analysis of the underlying causes shows that this deterioration is due to a lack of organizational oversight and proactive input controls during manual data entry in the source systems, causing the ETL’s deterministic mapping logic to fail. This lack of completeness directly leads to financial inaccuracies, inefficient manual workarounds, and reduced user confidence. Consequently, the study concludes that a shift from reactive corrections to robust data governance and input validation at the source level is crucial to ensuring that integrated business data remains fit for use.

Digitaliseringen av leveranskedjor har gjort informationsdelning mellan organisationer allt mer viktig för planering, samordning och effektivitet. Men med denna utveckling rör sig data över organisatoriska gränser samtidigt som informationssäkerhet blir en gemensam utmaning mellan flera aktörer. Denna studie undersöker hur informationssäkerhet hanteras i praktiken vid interorganisatorisk informationsdelning i digitalt integrerade leveranskedjor inom industrisektorn. Syftet är att uppmärksamma hur informationssäkerhet kan hanteras i praktiken när organisationer inom industrisektorn delar information i digitalt integrerade leveranskedjor.

Studien genomfördes som en kvalitativ fallstudie i Thailand och bygger på flera empiriska datakällor. Materialet består av dokumentgranskning, nio semistrukturerade intervjuer, observationer av vardagliga arbetsmoment, arkivmaterial i form av incidentrapporter och loggunderlag samt granskning av fysiska artefakter. Datamaterialet analyserades genom triangulering och med stöd av teoretiska propositioner kopplade till CIAtriaden, Shadow IT, åtkomstkontroll och identitetshantering, styrning samt interorganisatoriska säkerhetsrisker.

Resultaten visar att informationssäkerhet i den studerade leveranskedjan vilar på tydliga formella riktlinjer men att det praktiska arbetet formas av operativa behov, tidspress och nära samarbete mellan aktörer. Informationsdelning sker inte bara på avsedda system utan också via alternativa kanaler såsom chattar, telefonsamtal, manuella listor och tillfälliga delningsytor. Studien visar även att åtkomst i vissa situationer hanteras informellt, att ansvar mellan aktörer kan bli överlappande samt att säkerhetsrisker ofta uppstår i samspelet mellan organisationer än inom enskilda organisationer. En viktig slutsats är att informationssäkerhet i digitala leveranskedjor inte bara kan förstås genom tekniska kontroller, policydokument eller avtal. Den måste även förstås i relation till hur arbetet faktiskt genomförs i praktiken. Studien visar att leveranskedjans behov av snabb respons och tät samordning kan skapa situationer där säkerhetsrutiner blir svåra att upprätthålla fullt ut. Därmed uppstår en återkommande spänning mellan kontroll, effektivitet och samarbete.

Post-quantum cryptography (PQC) is becoming an increasingly important field of study as the advance of quantum computers are expected to one day threaten the public-key cryptographic schemes used today. Although technical standards for PQC are emerging, successful implementation for organisations depends on more than just replacing the cryptographic algorithms. This study aims to find the main organisational barriers that hinder the successful implementation of PQC. This was done by conducting a systematic literature review and analysing the data to create a thematic analysis set on identifying recurring organisational barriers. The results showcase that implementation is primarily hindered by legacy systems and integration challenges, followed by insufficient strategic governance and migration readiness, external dependencies, regulatory and standards uncertainty, lack of expertise and resource constraints. The findings indicate that PQC isn’t only a technical problem but a wider socio-technical problem. The study concludes that organisations require better migration planning, improved internal expertise, clearer regulatory guidance and stronger coordination with external actors for organisations to successfully implement PQC.

Introduction: Sustainability has become a central part of fashion companies’ strategies and communication. At the same time, the fast fashion-industry has been criticized for contradictions between sustainability goals and a business model based on high consumption and rapid production. Previous research shows that differences may arise between companies’ sustainability communication and how sustainability work is implemented in practice. 

Purpose: The purpose of this study is to examine how sustainability work is implemented at the store level within the fast fashion-industry and how this relates to companies’ sustainability communication. By examining the relationship between communication and practice, the study contributes to a greater understanding of the organizational tensions that may emerge when sustainability initiatives are implemented in practice. 

Method: The study is based on a mixed methods approach with a qualitative emphasis. The empirical material was collected through six semi-structured interviews, 28 survey responses, and secondary data in the form of sustainability reports and corporate communication from H&M, Zara, Lindex, and Kappahl. The material was analyzed through thematic analysis and descriptive analysis. 

Conclusion: The results show that sustainability work at the store level is mainly implemented through practical routines such as recycling and material handling, while broader sustainability goals are often perceived as difficult to translate into everyday store operations. The study also shows that sales targets are often prioritized over sustainability issues in daily practice. Several respondents experienced differences between companies’ sustainability communication and the practical sustainability work, indicating that greenwashing can also be understood as a result of organizational priorities and implementation difficulties.  

Glioma is one of the most aggressive central nervous system malignancies; yet diagnosis still depends on invasive neuroimaging and biopsy procedures. Plasma-based protein biomarkers offer a minimally invasive alternative, but identifying a reliable panel from high-dimensional proteomic data remains challenging. This study applied differential expression analysis, machine learning (ML)- based feature selection, and comparative classifier evaluation to identify a plasma protein panel capable of distinguishing glioma from other cancer types in a pan-cancer proteomics dataset (n = 1,477 cancer patients, 1,463 plasma proteins). Differential expression analysis identified 227 upregulated and 685 downregulated proteins in glioma. Unsupervised dimensionality reduction showed partial global separability, while supervised approaches revealed locally coherent proteomic structure. Gene ontology (GO) enrichment analysis identified catabolic and metabolic processes as the dominant enriched biological processes, reflecting tumor-driven metabolic reprogramming and extracellular matrix (ECM) catabolism. Consensus feature selection across Elastic Net (EN), Random Forest (RF), and Support Vector Machine (SVM), ranked GFAP (mean rank = 2.33), MMP12 (5.0), and ITIH3 (26.0) as the top discriminative proteins. Classification on an independent test set (n = 442) using the top 30 consensus proteins yielded AUC = 0.942 for linear discriminant analysis (LDA) (sensitivity = 0.860, specificity = 0.880), AUC = 0.937 for EN (sensitivity = 0.767, specificity = 0.940), and AUC = 0.925 for RF (sensitivity = 0.837, specificity = 0.960). Repeated cross-validation confirmed that most of the discriminative information was captured within 3-10 proteins and revealed systematic interactions between the feature selection method and downstream classifier performance. The three-protein panel (GFAP, MMP12, ITIH3) achieved AUC > 0.92. Formal statistical comparison through one-way ANOVA with Tukey post-hoc test confirmed that EN achieved the most stable classification performance, RF and LDA performed at equivalent levels, while k-nearest neighbours (kNN) was significantly inferior to all four. The results show that feature extraction method and classifier choice must be considered jointly in plasma proteomics-based classification pipelines.

Background: Traditional accounting knowledge has been part of accounting work since the beginning, but how its knowledge is used in everyday work has changed. Today, AI systems are developing faster than before and are in a constant development phase, which means that the skills needed in line with this are also developing, as much of the monotonous work in accounting is carried out by AI. If accounting changes, it means that accounting economists are affected by this large and rapid development.

Purpose: The fact that accounting is changing means that the way accounting economists work is also changing. It can therefore be understood that it arouses opinions, either positive or negative. The purpose is therefore to investigate how accounting economists accept the use of AI in their work.

Theoretical framework: Central concepts and previous research that are important for the study are presented here, as well as the theory used to analyze the results. The UTAUT theory is based on several central factors that are intended to explain the individual's acceptance of new technology.

Method: This study uses a qualitative method because nuanced details can be collected from the study's respondents. Data has been collected based on semi-structured interviews. This type of interview allows for follow-up questions and thus contributes to a deeper relationship and understanding of the respondents' answers. This data has then been analyzed through a thematic analysis.

Conclusion: The general conclusion to the study's question is that accountants accept AI with conditions. To the extent that control, review and support with understanding exist for the use of AI in accounting work, acceptance increases. At the same time, there is still some concern about how it will develop, but at present there is no indication that AI will replace accountants but rather exist as an aid, which increases acceptance among accountants.

As AI-based customer support and voice-based assistants become increasingly common in customer service, the need to understand users’ acceptance increases. Despite technical advancements, most users still favour human contact and traditional customer service options. In addition, there remains a knowledge gap regarding users’ attitudes during the so-called anticipation stage, that is, prior to the actual interaction. The purpose of this study is therefore to investigate which attitudes, expectations and prior experiences influence users’ perceptions of AI-based customer support before interaction. The main focus is to identify factors that foster trust and acceptance, with an additional emphasis on voice-based in-car AI support.

The study uses a triangulation approach combining both quantitative and qualitative data. Nine semi-structured interviews were conducted, along with a survey that resulted in 400 respondents. The development of the interview and survey questions, as well as the analysis, was guided by theoretical frameworks such as the Technology Acceptance Model (TAM) and SERVQUAL.

The results show a strong general preference for human contact (83.3%) compared to chatbots (9.3%). Users reported low levels of perceived trust in both chatbots and in-car AI support, while the most positively rated construct differed between the two contexts. In automotive contexts, voice-based AI is seen as a potentially useful tool for the scenario presented in the study, although concerns about cognitive distraction and road safety are significant barriers to acceptance. Users prefer a hybrid solution where AI handles routine questions while human support remains available in case of complex or emotionally charged inquiries. Increasing willingness to interact with AI-based customer support requires a higher level of trust, transparency regarding the systems abilities and a seamless transition to human agents if needed. The study resulted in a requirement specification emphasizing the importance of designing for security and contextual understanding to increase user acceptance and engagement. The requirement specification was created by applying the MoSCoW-prioritization method to all applicable findings identified in the study. 

This thesis examines whether large language models (LLMs) can translate natural language questions into SQL queries for a Manufacturing Execution System (MES). MES databases store important production data, but retrieving that data can require knowledge of the interface, database structure, or SQL. The study therefore investigates how model size and model specialisation influence effectiveness and efficiency in natural language-to-SQL generation. A controlled quasi-experiment was carried out in collaboration with Schneider Electric using four Qwen2.5 models, the same question bank, a sanitised industrial MES database, manually created retrieval-augmented generation (RAG) context, SQL validation and read-only execution. The evaluation considered both effectiveness and efficiency. Effectiveness was measured through exact match, execution success, and F1-score, while efficiency was measured through latency and token usage. The results showed that no model performed best in every area. Among the larger models, specialisation led to different strengths: one model produced more correct results regarding exact match and F1-score, while the other more often generated executable SQL. The smaller models were faster overall. The experiment results were also discussed with a Schneider Electric representative through a limited semi-structured validation interview to support the practical interpretation of the findings. 

Large Language Models (LLMs) have advanced Natural Language Processing (NLP) but remain prone to hallucinations, particularly in domain-specific applications such as legal text processing. Retrieval-Augmented Generation (RAG) addresses this limitation by grounding responses in retrieved documents. This study investigates the main and interaction effects of chunking strategies and embedding models on retrieval performance in a Naive RAG pipeline using a GDPR-based legal corpus. A controlled factorial experiment evaluated recursive, token-based and sentence-based chunking in combination with the OpenAI text-embedding- 3-small and text-embedding-3-large models using Recall@k, Mean Reciprocal Rank (MRR) and normalized Discounted Cumulative Gain (nDCG@k). The results show that both the chunking strategy and the embedding model have statistically significant effects on retrieval performance, with the chunking strategy being the dominant factor. A statistically significant interaction effect was also detected, although the effect sizes for both the main and interaction effects were small. Sentence-based chunking achieved the strongest overall performance, while token-based chunking produced the highest MRR. Inferential analyses indicate that these differences should be interpreted with caution, given their limited practical significance.