Publikationer från Uppsala Universitet
Not a member yet
127603 research outputs found
Sort by
Diskretionära periodiseringar och insiderhandel : En studie på företag inom EU
Insiders är personer med särskild insyn i ett företag och besitter därmed ett informationsövertaggentemot externa marknadsaktörer. Till följd av detta informationsövertag tillskriver marknadeninsidertransaktioner ett informationsvärde. Ju större informationsövertag de besitter, desto merinformativa anser marknaden att transaktionerna är. Samtidigt kan insiders påverka detinformationsövertag de besitter genom användning av diskretionära periodiseringar.Vi utför en kombination av en eventstudie, kvintilanalys och tvärsnittsregressioner på företag inomEU och undersöker transaktioner som genomförts mellan 2016 och 2023. Vi undersöker hurdiskretionära periodiseringar påverkar marknadens kortsiktiga reaktion i samband medinsidertransaktioner och finner ett samband där skillnader i diskretionära periodiseringar ocksåinnebär skillnader i onormal avkastning. Sambandet som identifieras uppvisar indikatorer på ickelinjäritet, vilket bör undersökas vidare i kommande studier
Artikulation och frakturläkning hos ventrala dermala element hos en devonisk Ichthyostega-liknande stam tetrapod
This thesis investigates the morphology, articulation, and histological features of ventral dermal elements in a Devonian Ichthyostega-like stem tetrapod. Using high-resolution three dimensionalreconstruction and virtual thin-section histology, the gastralia are described as compact rods with a twolayered organization: a thin superficial layer and a thicker basal lamellar layer densely populated with aligned osteocyte lacunae. Unlike amniote gastralia or osteoderms, vascular canals are absent, although evidence of fracture healing indicates a capacity for biomechanical stress resistance and skeletal repair.Articulatory patterns reveal a structured, interlocking arrangement, suggesting a functional role in ventral body support, stabilization, and possibly respiration. Comparative analysis with early osteichthyan scales, temnospondyl gastralia, and amniote ventral elements places this specimen within an intermediate stage of dermal ossification and skeletal adaptation during the fish-tetrapod transition.These findings highlight the evolutionary significance of gastralia as unique dermal structures that provided mechanical support during early terrestrialization. Their eventual reduction and loss in later tetrapods may reflect shifting functional demands as locomotor and respiratory strategies evolved
Time-Step Analysis for cutHermite Finite Element Methods
Hermite finite element methods, or Hermite FEM, use a set of basis functions that parameterise derivative values as well as function values at various points over a computational domain. This allows the use of function spaces that are continuous in derivatives as well as value over the computational domain. Doing this is shown to have computational benefits, such as allowing a larger stable time-step in time-dependent problems, and solving equations with higher spatial derivatives. This is particularly useful for modifying equations with higher derivative terms to increase the order of accuracy of a time-stepping scheme. A limitation of these methods is that basis functions corresponding to derivatives in 2D or higher are more dependent on the shape of grid elements than basis functions corresponding to value. By using cut finite element methods (cutFEM), problems can be solved on domains without the exact shape of the domain being represented in the grid. This allows the benefits of Hermite FEM to be gained, without placing restrictions on the computational domain or deriving a new basis for every grid
Phenomena based on Electron hopping transport in Metal-Organic Frameworks
Electron transfer reactions are fundamental to nearly all energy transformations in nature and modern technology, yet the increasing demand for clean energy has exposed fundamental limitations in current electrode materials. Achieving efficient and tunable charge transport remains a central challenge in the development of sustainable energy conversion systems. Metal-organic frameworks (MOFs), with their modular structures and electronically isolated redox units, offer a unique platform for probing and harnessing redox conduction. This thesis investigates redox-active MOF thin films, with Zn-naphthalene diimide (Zn-NDI) as the central model system, to uncover charge transport mechanisms and their implications for energy-relevant applications. In Paper I, redox conductivity (RC) is experimentally demonstrated in Zn-NDI for the first time, showing a bell-shaped dependence on redox composition and a 10,000-fold enhancement between the least and most conductive states. The generality of this phenomenon is confirmed in MOFs with different linkers and secondary building units, establishing RC as a fundamental transport mode in MOFs. Paper II extends the study to an isoreticular Zn-XDI (X = PM (pyromellitic), N (naphthalene), P (perylene); DI = diimide) series, where structural variations in diimide linkers modulate both optical and electrochemical properties. Among these, Zn-PDI exhibits a record high coloration efficiency of 941 cm² C-¹ at 746 nm with excellent cycling stability, highlighting the impact of faradaic processes and electrochromic performance. In Paper III, mixed-linker and bilayer MOF films are constructed to investigate directional charge transport, electric rectification, and charge storage properties. The results demonstrate that redox properties of the linkers and their spatial distribution can be leveraged to control electron transport pathways in MOF thin films. Paper IV integrates Zn-NDI as a redox-active overlayer on semiconductor photocathodes, revealing transport-limited photocurrents at high light intensities. These findings underscore the importance of transport properties in the rational design of surface layers that dictate interfacial charge dynamics, extending conventional perspectives on the kinetic limiting factors of photoelectrodes. Collectively, this thesis establishes RC MOFs as a tunable scaffold for understanding and tailoring charge transport, providing design principles for next-generation devices in electrochromism, molecular rectification, energy storage, and photoelectrochemical energy conversion, thereby contributing to the development of sustainable energy technologies
Asthma diagnosis is not associated with reduced beta-cell function in children and adolescents with obesity
Postoperative Cholangitis After Pancreatoduodenectomy : A Frequent Late Complication Demanding Standardized Prevention and Treatment Protocols
Background: Postoperative cholangitis (POC), a late complication after pancreatoduodenectomy, is thought to be caused by digestive fluid reflux to the biliary tract or an anastomotic stricture. It is not clear which perioperative risk factors contribute to POC. Methods: This study included all adult patients undergoing pancreatoduodenectomy between 2008 and 2021 at Karolinska University Hospital, Stockholm, Sweden. Electronic medical records were used to identify patients with POC. Fine and Gray and logistic regression models were used to investigate perioperative risk factors for late cholangitis. Results: Of the 1002 patients in the study, 86 (9%) experienced POC, and 33 (38% of all the patients with POC) had recurrent POC at least three times during the follow-up period. Preoperative biliary drainage (PBD) was associated with a lower risk of POC (unadjusted subhazard ratio [SHR] 0.60; 95% confidence interval [CI] 0.39-0.92). The patients with Clavien-Dindo grade >= IIIa complications and those with bile leakage grade B or higher had a higher risk of POC (odds ratio [OR] 1.76 [95% CI 1.11-2.81] and OR 2.76 [95% CI 1.19-6.64], respectively). Intermittent antibiotic treatment was used for 78 (91%) of the POC patients, and 36 (42%) of the POC patients were receiving prophylactic treatment. There were no major differences in the risk of cholangitis and surgical technique, including anastomosis level and suture method. Conclusions: Postoperative cholangitis after pancreatoduodenectomy is a rather common late complication, occurring in nearly 1 of 10 patients. In this study, preoperative biliary drainage was associated with a decreased risk of cholangitis. Also, the patients with major postoperative complications, including biliary anastomotic leakage, were more likely to experience cholangitis
Major cardiovascular events in first-degree relatives of individuals with elevated plasma lipoprotein(a) : a registry-based cohort study
Background and Aims: Lipoprotein(a) [Lp(a)] levels are genetically determined and causal in the development of atherosclerotic cardiovascular disease. Whether first-degree relatives (FDRs) of individuals with elevated Lp(a) levels (>= 80th percentile) have an increased cardiovascular disease risk is unknown. Methods: Based on 41 304 indexes with a routine plasma Lp(a) measurement, 61 715 FDRs without a measured Lp(a) aged 35-69 years (49% women) were identified in the Swedish STRIREG cohort. First-degree relatives were stratified according to index Lp(a) percentile level: <50th, 50-<80th, 80-<95th, and >= 95th. In competing risk-adjusted cumulative incidence and adjusted Cox proportional hazards regressions, the association between Lp(a) strata and major adverse cardiovascular events (MACE; cardiovascular death, myocardial infarction, ischaemic stroke, coronary revascularization) was investigated. In a nested analysis of 4243 indexes with a first-degree relationship with another index, the concordance of plasma Lp(a) levels was assessed. Results: During a median follow-up of 19 (11-26) years, 2043 FDRs had a MACE. The cumulative incidences of MACE in FDR until age of 65 were 6.2%, 7.0%, 7.5%, and 8.1% by increasing index Lp(a) level strata (P < .001). Compared with FDR in the lowest Lp(a) stratum, there was a higher hazard ratio for MACE by increasing Lp(a) stratum: 1.08 (95% confidence interval, 0.97-1.19), 1.30 (1.15-1.47), and 1.28 (1.06-1.55; P-trend < .001). The Lp(a) concordance was 86% (<80th percentile) and 53% (>= 80th percentile). Conclusions: First-degree relatives of individuals with elevated Lp(a) levels have a higher incidence of MACE. Cascade screening could be a feasible strategy to identify FDR at heightened risk. [GRAPHICS
A Model-Based Approach to Evaluate Anti-Drug Antibody Impact on Drug Exposure With Biologics : A Case Example With the CD3 T-Cell Bispecific Cibisatamab
The administration of biologics can lead to immunogenic responses that trigger anti-drug antibody (ADA) formation. ADAs can decrease drug exposure. A population pharmacokinetic (popPK) model was developed to describe clinical PK data with and without ADA-driven exposure loss with CEA-directed T-cell bispecific antibody cibisatamab. The PK of cibisatamab was evaluated in two clinical studies (as a single agent and in combination with the checkpoint inhibitor atezolizumab) in patients. The popPK model was developed on cibisatamab clinical PK data using the Stochastic Approximation -Expectation Maximization (SAEM) algorithm implemented in Monolix. Cibisatamab's PK followed a two-compartment model with linear clearance decreasing over time and ADA-associated exposure loss. ADA-driven exposure loss was implemented in the model by accounting for ADA formation, reversible binding to cibisatamab, and elimination of both free ADA and the ADA-cibisatamab complex from the central compartment. The impact of ADAs on PK exposure was time-dependent in the model, with the ADA formation described as a function of time (increasing from zero, reaching its estimated maximum value, and possibly decreasing down to 94% of this maximum value in some patients). The final model included a mixture component differentiating patients with and without exposure loss due to ADA formation (75% and 25% of patients, respectively). The investigated patient demographics, dose or dosing schedule, or atezolizumab coadministration were not identified as factors influencing exposure loss due to ADAs. The developed model can be used to differentiate patients with and without ADA-driven exposure loss, as well as for a precise PK characterization in patients even with ADA formation
Deuterium and tritium density determination at JET using neutron diagnostics
The fuel ion densities of tritium (T) and deuterium (D) are important parameters for reactor control in fusion experiments. The fuel ion densities typically need to be determined for a wide range of experimental conditions, anywhere between pure deuterium plasmas with trace amounts of T to pure tritium plasmas with trace amounts of D. Given the significantly larger cross section for D+T fusion reactions, compared to D+D or T+T, introducing small amounts of tritium to a deuterium plasma (or vice versa) quickly has a large impact on the neutron emission rates and the neutron energy spectrum. For experiments that use the neutron emission rate as a figure of merit, or for simulations, the time-resolved T and D densities often need to be considered. In this paper, we present a method for determining the densities using multiple neutron diagnostics together with simulations of supra-thermal fuel ions utilizing the plasma transport code TRANSP. The method is set up utilizing a Bayesian framework to estimate the most likely distribution of T and D densities given the available data and the modeled slowing-down ion velocity distribution, using the synthetic neutron diagnostics code DRESS to calculate the expected neutron emission for the given fuel ion distributions. The method is applied to experiments conducted at the JET tokamak, involving deuterium-dominated plasmas with tritium concentrations ranging from 0 to 10%, heated by D neutral beam injection and ion cyclotron resonance frequency heating
Multimodal-LLM as A Reliable Tool for Information Extraction from Historical Documents : A Digital Humanities Approach to Swedish Patent Cards (1945-1975)
This thesis investigates whether general-purpose Multimodal Large Language Models can serve as reliable tools for historical information extraction within Digital Humanities background. Using a sample of 100 Swedish historical patent cards (1945–1975), the study evaluates both proprietary and open-source MLLMs through a pipeline adapted from prior work by Xie, Mela and Tell (2025). Beyond performance metrics, the analysis is framed by four core values in DH: accessibility, transparency and reproducibility, iterative design, and ethical responsibility. Proprietary models – especially Gemini 2.5 Pro – demonstrated strong accuracy across multiple types of text transcription and showed reliable potential in handling reasoning tasks. However, cloud-based platforms continue to raise concerns regarding the opacity of training data, the privacy of user data and the bias hidden in the model structure. In contrast, locally deployed open-source models, such as Qwen2.5-VL, offer data security and greater control over input processing. Yet, under standard consumer-grade hardware configurations, they proved limited in performance. By embedding evaluation within a real research workflow and comparing deployment scenarios, this study proposes a practical, value-driven framework for selecting and assessing AI tools in DH contexts. The findings highlight the importance of critical tool engagement and demonstrate how MLLMs, when thoughtfully implemented, can support scalable, transparent, and ethically sound digital scholarship