154092 research outputs found
Sort by
Compound events in Germany in 2018:drivers and case studies
Get PDFEurope frequently experiences a wide range of extreme events and natural hazards, including heatwaves, extreme precipitation, droughts, cold spells, windstorms, and storm surges. Many of these events do not occur as single extreme events but rather show a multivariate character, known as compound events. We investigate the interactions between extreme weather events, their characteristics, and changes in their intensity and frequency, as well as uncertainties in the past, present, and future. We also explore their impacts on various socio-economic sectors in Germany and central Europe. This contribution highlights several case studies with special focus on 2018, a year marked by an exceptional sequence of compound events across large parts of Europe, resulting in severe impacts on human lives, ecosystems, and infrastructure. We provide new insights into the drivers of spatially and temporally compound events, such as heat and drought, and heavy precipitation combined with extreme winds, and their adverse effects on ecosystems and society, using large-scale atmospheric patterns. We also examine the interannual influence of droughts on surface water and the impact of water scarcity and heatwaves on agriculture and forests. We assess projected changes in compound events at different current and future global surface temperature levels, demonstrating the need for improved quantification of future extreme events to support adaptation planning. Finally, we address research gaps and future directions, stressing the importance of defining composite events primarily in terms of their impacts prior to their statistical characterisation
Measuring Security and Resilience in Cloud Outsourcing for Data-Driven Risk Management
Get PDFOver the past decade, our reliance on the Internet has grown exponentially, driving the need for faster, more reliable, and better-performing online services to support our daily lives. Cloud computing has emerged as a solution, offering organizations affordable, flexible, and reliable IT services. However, while cloud services deliver significant advantages, their associated risks—both technological and economic—are becoming increasingly complex and sophisticated.This thesis employs two complementary approaches to investigate cloud outsourcing risks and risk management strategies. First, we conduct a systematic literature review to analyze the state-of-the-art in academic research, identifying key risks and risk management techniques available to cloud consumers. Second, we use empirical Internet measurement data to examine how these risks and strategies manifest in real-world cloud environments.Our risk assessment focuses on two major cyber threats: malware infections and DDoS attacks. We quantify cloud consumers’ exposure to these risks by evaluating the effectiveness of cloud-based malware detection services and analyzing the role of popularity and industry sector in DDoS victimization. Regarding risk management strategies, we examine both reactive and proactive approaches. We analyze how organizations respond to large-scale DDoS incidents affecting cloud providers, such as the Dyn DDoS incident in 2016, and assess how cloud consumers adjust their infrastructure proactively in anticipation of potential disruptions, such as those arising from the Russia-Ukraine conflict. By bridging the gap between theoretical insights from academic literature and empirical data from real-world Internet measurements, this research provides a comprehensive perspective on cloud outsourcing risks. Our findings offer actionable recommendations to help organizations improve their risk assessment practices and develop more effective cloud security strategies
A local and historical perspective on disaster risk reduction:Tunja, Colombia case study
Get PDFThis autoethnographic study presents a historical perspective on disaster risk reduction (DRR) at the local level, with the case study of the city of Tunja in Colombia. We analyze the impact that colonization, the independence period, and recent history have had on the creation and reduction of disaster risks in this city. We offer a holistic perspective that shows the interactions of the impact of inequality on Indigenous populations, lack of urban planning, deforestation and the planting of invasive plant species, among other factors, which in combination with natural hazards, such as heavy rainfall, increase disaster risks. We conclude that although the Sendai Framework for Disaster Risk Reduction 2015–2030 is a fundamental instrument to promote risk reduction, in the local context of Tunja, the framework as such is not seen as a guide or parameter. The Colombian Disaster Risk Management Law is the main guide to advance risk reduction. This study demonstrates how DRR is not an isolated process, but a process that encompasses the general well-being of the population. We demonstrate from our lived perspective how access to public education and school feeding, as well as other social protection measures, increase the resilience of the population, making them better able to cope with adversity due to different hazards. This local perspective, with a historical review of a small city in the middle of the Andes, demonstrates the importance of continuing to prioritize and invest in measures that contribute to the population’s well-being as a way to reduce disaster risks, including adapting to our changing climate
Predicting clinical outcomes in a blended care intervention for early psychosis:Acceptance and Commitment Therapy in Daily-Life (ACT-DL)
Get PDFACT in Daily Life (ACT-DL) is a blended-care Ecological Momentary Intervention that extends ACT into the daily life of individuals, improving psychotic distress, negative symptoms, and global functioning. However, it remains unclear whether ACT-DL works equally for everyone. We investigated whether moderators (i.e., sociodemographic information, personality, and trauma history) determine clinical outcomes in individuals with early psychosis receiving ACT-DL. Seventy-one participants from the INTERACT trial, using ACT-DL, were analyzed. Outcomes included psychotic distress, negative symptoms, global functioning, and psychological flexibility. Using multivariate-multilevel models, we evaluated the effects of sociodemographics, personality, and childhood trauma across baseline, post-intervention, and six- and 12-month follow-ups. Sociodemographic characteristics and personality predicted clinical outcomes. Higher education demonstrated more substantial improvement in global functioning at 6- (B = 7.43, p = 0.04) and 12-FU (B = 10.74, p = 0.002) compared to lower education. Higher extraversion showed less improvement in negative symptoms at 12-FU (B = 1.24, p = 0.01) and more improvement in global functioning at post-intervention (B = 0.39, p = 0.046) and 6-FU (B = 1.40, p = 0.02) compared to lower extraversion. Higher negative affectivity showed more improvement in negative symptoms at 12-FU (B = −1.59, p = 0.001) and higher psychological flexibility at 12-FU (B = 8.38, p = 0.001) compared to lower negative affectivity. Our findings suggest that while ACT-DL improves clinical outcomes in individuals with early psychosis, the improvement rate is dissimilar for individuals and predictable by baseline characteristics. If replicated, these findings enable precision medicine approaches in allocating ACT-DL for early psychosis.</p
Digital inequality and digital skills: Examining barriers and solutions in Indonesia's mobile banking adoption:Addressing digital disparities to creating opportunities through the development of digital skills, supporting broader adoption and usage of mobile banking services across Indonesia
Get PDFThis dissertation explores why many Indonesians, particularly in rural and lower-income communities, struggle to adopt mobile banking despite having access to digital technology. While mobile banking has the potential to increase financial inclusion, many people face barriers such as low digital literacy, lack of trust in digital financial services, inadequate infrastructure, and concerns about security and fraud. These challenges contribute to digital inequality, where only certain groups benefit from technological advancements while others remain excluded.Through extensive survey analysis, this research identifies three main problems in mobile banking adoption in Indonesia. First, many users lack the digital skills needed to use mobile banking effectively, leading to frustration and reluctance. Second, distrust in mobile banking services and concerns about fraud discourage potential users from transitioning to digital financial platforms. Third, unequal access to stable internet connections and affordable smartphones further widens the gap, particularly in remote areas.To address these issues, this study proposes several practical solutions. First, financial institutions and policymakers should invest in digital literacy programs to help users navigate mobile banking safely and confidently. Second, banks must enhance security features and improve fraud protection measures, while also increasing transparency to build user trust. Third, expanding internet infrastructure and ensuring affordability of mobile devices will help reduce barriers to access, making mobile banking a viable option for a broader population.This dissertation contributes to the field of digital inclusion and financial technology by offering realistic and actionable recommendations to bridge the digital divide. By addressing digital inequality in mobile banking adoption, the findings of this research aim to support the development of a more inclusive financial ecosystem in Indonesia.<br/
PDE-DKL:PDE-constrained deep kernel learning in high dimensionality
Get PDFMany physics-informed machine learning methods for PDE-based problems rely on Gaussian processes (GPs) or neural networks (NNs). However, both face limitations when data are scarce and the dimensionality is high. Although GPs are known for their robust uncertainty quantification in low-dimensional settings, their computational complexity becomes prohibitive as the dimensionality increases. In contrast, while conventional NNs can accommodate high-dimensional input, they often require extensive training data and do not offer uncertainty quantification. To address these challenges, we propose a PDE-constrained Deep Kernel Learning (PDE-DKL) framework that combines DL and GPs under explicit PDE constraints. Specifically, NNs learn a low-dimensional latent representation of the high-dimensional PDE problem, reducing the complexity of the problem. GPs then perform kernel regression subject to the governing PDEs, ensuring accurate solutions and principled uncertainty quantification, even when available data are limited. This synergy unifies the strengths of both NNs and GPs, yielding high accuracy, robust uncertainty estimates, and computational efficiency for high-dimensional PDEs. Numerical experiments demonstrate that PDE-DKL achieves high accuracy with reduced data requirements. They highlight its potential as a practical, reliable, and scalable solver for complex PDE-based applications in science and engineering
Towards mapping ecosystem resilience from space:canopy defensive properties in European temperate forest revealed with spaceborne imaging spectroscopy
Get PDFFoliar functional traits are dynamic plant properties that vary across space and time, serving as principal tools for monitoring plant physiology and terrestrial ecosystem processes. Phenolics are the most crucial secondary metabolites that play key roles in plant defence against biotic and abiotic stressors, leaf decomposition, as well as consequent influence on nutrient cycling and soil microbial composition. However, spatially continuous information on canopy phenolic remains poorly characterized at the landscape level. Current and proposed spaceborne imaging spectrometers offer unique opportunities to map foliar phenolics quantitatively through space and time. Our recent work (Xie et al, 2024) demonstrated that foliar phenolics can be accurately estimated across temperate tree species using leaf spectroscopy. In this study, we leveraged imaging spectroscopy data from PRecursore IperSpettrale della Missione Applicativa (PRISMA) mission to predict and map foliar phenolic variations at canopy scale in a mixed European temperate forest. Two data-driven approaches, namely partial least square regression and Gaussian processes regression, were applied to link lab-measured phenolic concentration with PRISMA plot-level spectra (400–2400 nm). The performance statistics indicated reasonable precision and accuracy of the model results. Maps derived from the best-performing model (based on cross-validated nRMSE) provided a wall-to-wall assessment of canopy phenolics, capturing both inter and intra-species variations across the landscape. Further, we compared the phenol map with the distribution of leaf mass per area and canopy nitrogen. The results indicated that the synergy patterns across the three functional traits were consistent with the known leaf economic spectrum. These findings highlight the potential of spaceborne imaging spectroscopy to characterize spatial and temporal dynamics of ecologically important plant phenolics. Our study also paves the way for improved global monitoring of ecosystem integrity and plant responses to environmental stress and climate change, particularly with the anticipated launch of hyperspectral missions like ESA’s CHIME and NASA’s SBG
Deep learning methods for multiple building use and urban livability:Evaluation from multimodal geospatial data
Get PDFIn the era of big data, an abundance of geospatial data are now widely available. Remote sensing (RS) images, digital surface models (DSM), night light remote sensing (NLRS) images, street view images (SVI), and point of interest (POI) data provide valuable insights into the spatial and social characteristics of cities, making these data important for urban research. The advancement of deep learning methods offers a powerful means to extract and analyze information from these diverse data sources. This thesis explores the application of deep learning algorithms in urban research by leveraging multiple geospatial datasets.Building use information as a subset of land use mapping is necessary for urban planning, city digital twins, and informed policy formulation. In chapter 2, unlike mainstream research on land use classification, my research has taken mixed-use scenarios into account. To enhance classification accuracy, the proposed strategy focuses on more effectively extracting and leveraging the features contained in various data modalities. It proposes a multimodal transformer-based deep learning method for building use classification. While mainstream research typically employs decision fusion strategies, the proposed strategy adopts feature fusion strategy to integrate multiple modalities. Specifically, a pretrained DenseNet is used as the backbone for extracting features from images and Bidirectional Encoder Representations from Transformers (BERT) for extracting features from the text. An attention mechanism is employed during the classification phase to assign appropriate weights to different features. The proposed multimodal transformer-based feature fusion network is tested across four Chinese cities. The results demonstrate that it effectively predicts both broad and mixed building use, significantly improving classification accuracy. This research highlights the potential of feature fusion strategies for integrating RS images and POI data in urban building use classification.Existing building use classification methods often focus primarily on broad categories, leaving a significant gap in the classification of buildings’ detailed uses. To address this gap and test the performance of the feature fusion-based method in different regions of the world, chapter 3 is expanded to include DSM and SVI in addition to the RS images and POI data used in the previous study. I employed a multi-label classification strategy dealing with the large number of labels caused by such combinations. An ablation study investigated the synergy between different modalities and examined the attention given to each modality. A novel multi-label multimodal transformer-based feature fusion network was used to effectively extract and integrate features from various modalities, enabling the simultaneous prediction of hierarchical building uses, containing both detailed and its corresponding broad use categories. The model effectively learns the relationships between broad and detailed use categories, including hierarchical consistency, supplementation, and exclusivity. The proposed method’s performance was evaluated in three Dutch cities. For test dataset, it achieved a weighted average 1 scores (WAF) of 91% for broad categories, 77% for detailed categories, and 84% for all hierarchical categories, and macro average 1 scores (MAF) of 81%, 48%, and 56%, respectively. This research thus demonstrates that RS data serve as the cornerstone for hierarchical building use classification, while DSM and POI data provide valuable supplementary information. SVI data, however, may introduce noise.Understanding how building characteristics influence urban livability is important for architects and urban planners in urban designing that promote functionality, sustainability, and community well-being. This includes creating spaces that optimize natural light, energy efficiency, and accessibility, while also considering the social and environmental impact on the surrounding urban fabric. In chapter 4, to address this question, a random forest regression was employed to model urban livability based on buildings’ spatial attributes (e.g., area, perimeter) and functional attributes (e.g., use). The experimental results indicate that urban livability can indeed be predicted by analyzing a building’s spatial and functional characteristics. Specifically, higher-density categories such as stacked residential, industrial, and business areas positively contribute to livability, whereas single-family residences, detached residential areas, row residential zones, and the presence of certain public services negatively impact livability. These findings highlight the significant role building characteristics play in shaping urban livability, offering valuable insights for urban planning and policy-making in alignment with Sustainable Development Goal 11.Traditional methods for evaluating urban livability rely on surveys and statistical data, which are often time-consuming, costly, and updated irregularly. While chapters 4 demonstrated that building characteristics can partially assess urban livability, the information these attributes provide is limited. Additionally, errors in building use classification can reduce the accuracy of livability regression. To enhance the accuracy of urban livability evaluations, chapter 5 explored the use of multiple data sources and deep learning methods. A Transformer-based multi-task multimodal regression (TMTMR) model was proposed to estimate livability scores for five associated domains and their overall score using RS images, NLRS images, DSM, and POI data. 13 Dutch research areas were involved, and experimental results indicate that geospatial data can effectively predict urban livability conditions with this method, outperforming models based solely on building characteristics. Among the four modalities, their contributions to livability assessment are ranked as follows: RS images, NLRS images, DSM, and POI data.In summary, this thesis investigates the effectiveness of deep-learning methods using multiple geospatial data to analyze the complex urban spatial structure, with a focus on building use classification and urban livability evaluation. By employing multimodal deep learning methods, this research demonstrates how information from diverse data modalities can be effectively extracted and integrated.<br/
Patient-reported outcomes after immediate and delayed DIEP-flap breast reconstruction in the setting of post-mastectomy radiation therapy—results of the multicenter UMBRELLA breast cancer cohort
Get PDFPurpose: Timing of Deep Inferior Epigastric artery Perforator (DIEP)-flap breast reconstruction in the context of post-mastectomy radiotherapy for breast cancer patients is topic of debate. We compared the impact of immediate (before radiotherapy) versus delayed (after radiotherapy) DIEP-flap breast reconstruction (IBR versus DBR) on short- and long-term patient-reported outcomes (PROs). Methods: Within the prospective, multicenter breast cancer cohort (UMBRELLA), we identified 88 women who underwent immediate or delayed DIEP-flap breast reconstruction and received PMRT. At 6 and 12 months post-mastectomy, as well as on long-term (≥ 12 months post-reconstruction) body image, breast symptoms, physical functioning, and pain were measured by EORTC-QLQ-30/BR23. Additionally, long-term evaluation included satisfaction with breast(s), physical well-being and self-reported adverse effects of radiation as measured by BREAST-Q, and late treatment toxicity. PROs were compared between groups using independent sample T-test. Results: IBR was performed in 56 patients (64%) and DBR in 32 patients (36%), with 15 months of median time to reconstruction. At 6 and 12 months post-mastectomy, better body image and physical functioning were observed after IBR. No statistically nor clinically relevant differences were observed in long-term EORTC and BREAST-Q outcomes (median follow-up 37–41 months for IBR vs. 42–46 months for DBR). Patients with IBR reported more fibrosis and movement restriction (median follow-up 29 vs. 61 months, resp.). Conclusion: Long-term PROs were comparable for patients with IBR and DBR, despite more patient-reported fibrosis and movement restriction after IBR. Therefore, both treatment pathways can be considered when opting for autologous breast reconstruction in the setting of PMRT.</p
What Next for the Science of Patient Preference? Interoperability, Standardization, and Transferability
Get PDFUsing patient preference information (PPI) to incorporate patient voices into the drug development lifecycle can help align therapies with the needs and values of patients. However, several barriers have limited the use of PPI, including a lack of clarity on its use by decision-makers, a need for greater decision-maker trust in PPI, and a lack of time, budgets, and access to specialist expertise. The value proposition for PPI could be enhanced by making it FAIR: Findable, Accessible, Interoperable, and Reusable. To support the development of a research agenda to deliver FAIR PPI, we reviewed related endeavors in the development of repositories of existing studies, disease models, benefit transfer, and common data standards. We concluded that developing FAIR PPI would require advances in the science of PPI, including the establishment of a consortium, mirroring the Clinical Data Interchange Standards Consortium (CDISC) or Observational Medical Outcomes Partnership (OPOM), to develop PPI data standards, and research into the sources of variation in patient preferences. This will require the science of PPI to graduate from being a body of empirical observations to developing theories that explain variations in patient preferences, simultaneously driving both efficiency in the generation of PPI and trust in PPI.</p