Publikationer från Uppsala Universitet
Not a member yet
127603 research outputs found
Sort by
A µs time synchronization for industrial embedded applications : Between Fantasy and Reality
No full textTime synchronization plays a major role in modern distributed systems to collectively work on a common task. Local clocks in industrial embedded devices are unreliable for maintaining highly accurate time synchronization between devices. It raises the requirement to use a protocol-based time synchronization method. Over time, a few time synchronization protocols were introduced to address different time synchronization requirements from the industry. This study systematically evaluates the influence of critical parameters on the performance of the IEEE 802.1AS Generalized Precision Time Protocol (gPTP) in resource-constrained embedded microcontroller environments. The Precision Time Protocol (PTP) defined in IEEE 1588-2019 has been designed to keep the time accuracy of device clocks over a communication network on a microsecond level. gPTP inherits PTP and ensures sub-microsecond time accuracy. A carefully selected test suite evaluates individual and collective impact on gPTP time synchronization from different network topologies, high CPU utilization, competing network traffic, and use of different Ethernet types. Having the gPTP requirements, the selected hardware performs well in time synchronization and interoperates with gPTP stacks and hardware from other vendors
Investigating the Sim-to-RealGeneralizability of YOLOObject Detection Models
No full textIn machine learning, access to real-world data can be a limiting factor, creating the need to understand the full implications of training a machine learning model on synthetic data, and deploying it in a real setting. One component of this issue is the sim-to-real generalizability of the model, characterized by the sim-to-real gap, a frequently encountered performance drop when testing on real versus synthetic data. This work investigates the YOLO family of object detection models on their ability to generalize across domains regarding model iteration, size, and release date. Our experiments show that the models display a sim-to-real gap while the influence of size and model recency on performance is not apparent on our visually simple dataset. We furthermore carefully examine several factors that partly explain how the gap arises and also investigate the connection between generalizability and performance
Phonon-mediated high-temperature superconductivity in clathrate superhydride ThCeH18 under pressure
No full textPredicting high-temperature superconductivity in hydrogen-rich dense metallic states under pressure remains a significant challenge. The discovery of new phases of metal hydrides is vital for advancing this area of research. In our study, we present the superconducting phases of ThCeH18 using a search strategy based on potential energy surfaces and evolutionary algorithms under pressure. Consequently, ThCeH18 is thermodynamically stable in a crystalline hexagonal unit cell with the space group \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}\end{document}, similar to CeLaH18 and LaThH18. The critical importance of hydrogen atoms is demonstrated, showcasing their significant influence on the evolution of Tc. Isotropic superconductivity in ThCeH18, driven by electron-phonon interactions and analyzed using the stochastic self-consistent harmonic approximation (SSCHA) to account for anharmonicity, reveals a strong anharmonic correction and strong electron-phonon coupling, resulting in high-temperature superconductivity with a Tc of approximately 136 K at 50 GPa.This approach provides a promising direction for identifying new classes of superconducting materials and offers valuable insights into enhancing superconductivity through structural prediction
Droplet microfluidics-based detection of rare antibiotic-resistant subpopulations in Escherichia coli from bloodstream infections
No full textPopulation heterogeneity in bacterial phenotypes, such as antibiotic resistance, is increasingly recognized as a medical concern. Heteroresistance occurs when a predominantly susceptible bacterial population harbors a rare resistant subpopulation. During antibiotic exposure, these resistant bacteria can be selected and lead to treatment failure. Standard antibiotic susceptibility testing methods often fail to reliably detect these subpopulations due to their low frequency, highlighting the need for improved diagnostic approaches. Here, we present a droplet microfluidics method where bacteria are encapsulated in droplets containing growth medium and antibiotics. The growth of rare resistant cells is detected by observing droplet shrinkage under microscopy. We validated this method for three clinically important antibiotics in Escherichia coli isolates obtained from bloodstream infections and showed that it can detect resistant subpopulations as infrequent as 10-6 using only 200 to 300 droplets. In addition, we designed a multiplex microfluidic chip to increase the throughput of the assay
Evaluation of Bridge Capture technology for mutation profiling in liquid biopsies of metastatic colorectal cancer patients
No full textColorectal cancer (CRC) is the second leading cause of cancer-related deaths, often presenting at an advanced stage with significant molecular heterogeneity. This is the first study to evaluate the performance of a novel next-generation sequencing (NGS)-based Bridge Capture technology for mutation profiling and minimal residual disease detection in circulating tumor (ct)DNA from metastatic colorectal cancer (mCRC) patients. Its performance was compared to those of droplet digital PCR (ddPCR), Ion AmpliSeq Cancer Hotspot Panel v2, and Idylla ctKRAS Mutation Assay. Eighty serial plasma samples from ten mCRC patients were analyzed by Bridge Capture and ddPCR, demonstrating a very strong correlation in variant allele frequency (VAF) values (rs = 0.86). The concordance of Bridge Capture with ddPCR (kappa = 0.70) and Idylla (kappa = 0.79) showed substantial agreement. A subset of samples (n = 10) was analyzed using the Ion AmpliSeq NGS-panel and both methods identified 15 driver mutations with strong correlation of VAF values (rs = 0.74). Additionally, Bridge Capture identified several oncogenic mutations beyond those detected by Ion AmpliSeq, highlighting its comprehensive profiling capability. The scalability of Bridge Capture was validated using an expanded panel and synthetic DNA targets, showing a strong linear correlation between observed and expected VAF values. This study demonstrates the scalability and accuracy of the Bridge Capture platform, and its potential to enhance mutation detection and clinical decision-making using ctDNA samples from patients with mCRC
Policy Instruments to Support Battery Energy Storage Systems: A Comparative Analysis of Two Regional Markets in Australia and the United States
No full textBattery Energy Storage Systems (BESS) are becoming increasingly central to electricity system decarbonisation. However, their deployment depends not only on technical feasibility or economic models, but on the alignment of regulatory, market, and technological conditions. This thesis applies the Policy-Market-Technology Triangle to examine how these dimensions interact to shape BESS deployment in California (CAISO) and New South Wales (NSW). Using a comparative case study method, the research draws on regulatory, market, and technical factors identified through a literature review and document analysis. Findings show that CAISO’s mature market structure, supported by policies such as FERC Order 841 and the Resource Adequacy program, enables wide BESS participation across multiple services. In contrast, NSW relies on state-led tools like Long-Term Energy Service Agreements (LTESAs) to manage risk and drive deployment in an emerging market context. Cross-dimensional analysis reveals that policy design often compensates for or reinforces market conditions, while technical limitations, such as lifecycle degradation and dispatch constraints, remain overlooked. Where these dimensions evolve in coordination, BESS deployment is more scalable and sustainable; where misalignments persist, deployment risks inefficiency or long-term constraints. The thesis concludes by recommending the integration of lifecycle costs into planning, clearer end-of-life policy, and reforms to enable multi-service market access. The triangle framework ultimately proves useful not only for mapping system conditions, but for identifying gaps in coordination that may affect long-term outcomes
Vid gränsen mellan väst och öst : En jämförande fallstudie av Litauens och Ungerns energisäkerhetsstrategier under ökad geopolitisk osäkerhet
No full textStudien jämför Litauens och Ungerns energisäkerhetsstrategier mot bakgrund av Europas beroende av rysk energi och ökade geopolitiska spänningar. Genom en kvalitativ jämförande fallstudie analyseras hur historiska erfarenheter, geografiskt läge och politiska prioriteringar påverkar respektive lands vägval. Litauen har stegvis brutit sitt beroende av Ryssland genom satsningar på LNG-terminaler, el- och gasnätsintegration med EU samt utbyggnad av förnybar energi. Ungern har istället valt att fördjupa samarbetet med Ryssland genom långsiktiga gasavtal och utbyggnad av kärnkraft i syfte att garantera prisvärd och stabil energitillgång. Studien visar att de två ländernas strategier speglar olika syn på risk, suveränitet och geopolitisk orientering. Medan Litauen ser energipolitik som en del av nationell säkerhet och europeisk integration, betonar Ungern en ekonomisk pragmatisk hållning och energisamarbete med öst. Resultatet belyser hur energisäkerhet formas av historiska och politiska faktorer snarare än enbart tekniska lösningar
Hotels, tents and sacred houses : spaces for human accommodation in Greek sanctuaries
No full textP19-0384:1_RJ / Den oheliga helgedomen? Att definiera temenos som ett rum för samspel mellan gudar och människor i antikens Grekland 600 f.Kr.-200 e.Kr. / project:716
The quantity and quality of B-cell immunity against SARS-CoV-2 in children with cancer and hematological diseases
No full textBackground Our understanding of protective immunity after natural viral infections in children with cancer and hematological diseases is restricted. Current cancer treatments cause significant immunosuppression, affecting both innate and adaptive immunity which leads to reduced B-cell and antibody responses. The aim of this study was to characterize SARS-CoV-2 immune response in children with cancer or hematological disease.Methods A single-center study was conducted from June 2020 to June 2023, including 135 patients and 14 healthy siblings. Blood samples were obtained for serological analysis and cell-based assays. SARS-CoV-2 IgG and IgA responses were quantified using suspension multiplex immunoassay (SMIA) and enzyme-linked immunosorbent assay (IgG ELISA) while neutralizing antibody (nAb) responses were assessed by plaque reduction neutralization tests (PRNT). The memory B-cell (MBC) population was evaluated through flow cytometry and MBC responses through FluoroSpot, respectively.Results In total, 78 patients seroconverted in response to SARS-Co-V-2 but neither immunosuppression nor cancer diagnosis significantly affected seroconversion. SARS-CoV-2 IgG and IgA levels correlated positively with increasing age, and IgA seroconversion was significantly associated with the presence of nAbs. Antigen-specific MBC responses against both spike and receptor-binding domain (RBD) were elevated in older children, while children on immunosuppression had significantly lower RBD IgG-secreting cells.Conclusion Our results show that most pediatric oncological and hematological patients can mount a broad antibody response upon SARS-CoV-2 natural infection or vaccination, although there is a variability in their responses influenced by increasing age. MBC responses in children with immunosuppression were blunted with fewer RBD IgG-secreting cells. Essentially, our findings underscore that young children with severe treatment-related immunosuppression are at risk for less effective B-cell responses upon viral infection.De två första författarna delar förstaförfattarskapet.De två sista författarna delar sistaförfattarskapet.</p
Jag dokumenterar, alltså finns jag. : Hinder vid tillämpning av AI-stödd dokumentation inom konsultbranschen.
No full textDigitaliseringens hastighet har gjort välstrukturerad projektdokumentation till en konkurrensfråga i kunskapsintensiva konsultbolag, men bristande dokumentation leder ofta till en kostsam ”dokumentationsskuld”. Denna masteruppsats undersöker vilka hinder som uppstår när svenska konsultföretag försöker införa AI-baserade verktyg som ska minska denna skuld. Med utgångspunkt i TOE-ramverket (Technology, Organization, Environment) genomfördes en kvalitativ fallstudie där fem stora konsultbolag (> 250 anställda) intervjuades via semistrukturerade samtal med AI-ansvariga. Resultatet visar att implementering hämmas av en kombination av tekniska, miljömässiga och - särskilt - organisatoriska faktorer: ledningens AI-visioner omsätts sällan i resurssatta program; fragmenterade datamiljöer och varierande kund-IT försvårar integration; medarbetarnas entusiasm blandas med osäkerhet kring AI-genererad informationskvalitet; höga säkerhets- och sekretesskrav begränsar molnbaserade lösningar; och svårigheter att kvantifiera avkastningen gör investeringarna känsliga för kostnadspress. Av dessa framstår de organisatoriska hindren - ledningsmandat, kultur, kompetensutveckling och ansvarsfördelning - som mest avgörande för om AI-baserade dokumentationsverktyg faktiskt används och skalar. Tillsammans skapar dessa hinder en ond cirkel där tekniken blir pilotprojekt snarare än etablerat arbetssätt, vilket riskerar att förstärka snarare än minska dokumentationsskulden.The accelerating pace of digitalisation has made well-structured project documentation a competitive issue in knowledge-intensive consulting firms, yet deficient documentation often results in costly “documentation debt.” This master’s thesis investigates the obstacles Swedish consulting firms face when attempting to deploy AI-based tools meant to reduce that debt. Grounded in the Technology-Organization-Environment (TOE) framework, we carried out a qualitative case study that involved semi-structured interviews with AI leads at five large consulting firms (> 250 employees). The findings show that adoption is hampered by a mix of technical, environmental, and - above all - organizational factors: management’s AI visions are rarely translated into funded programmes; fragmented data landscapes and heterogeneous client IT environments impede seamless integration; employee enthusiasm coexists with uncertainty about the quality of AI-generated information; stringent security and confidentiality requirements constrain cloud-based solutions; and difficulties quantifying return on investment make expenditure vulnerable to cost pressures. Of these, organizational barriers - leadership mandate, culture, competence development and clear allocation of responsibility - prove most decisive for whether AI tools are actually used and scaled. Collectively, these barriers create a vicious cycle in which the technology remains stuck at pilot stage instead of becoming established practice, risking the reinforcement rather than reduction of documentation debt