735962 research outputs found
Sort by
The Role of Public and Buyer-Driven Governance for Supplier Sustainability Development in China
No full textGovernments and regulatory agencies in emerging countries are increasingly enforcing labor and environmental protection laws. However, how this growing public governance interrelates with existing buyer-driven governance to influence supplier sustainability development remains unclear. To address this gap, we conducted a longitudinal case study involving two rounds of extensive data collection at nine Chinese suppliers of COSMOS, a European multinational company in the consumer electronics industry. We identify three distinctive configurations of how public governance and buyer-driven governance jointly influence supplier sustainability development. Identifying these configurations builds on prior research that distinguishes between competitive, complementary, and coexistent interactions between public and private governance. We extend this literature by uncovering the causal mechanisms through which each configuration gives rise to a unique sustainability development trajectory. Specifically, we unpack the different elements of each configuration and analyze their influence on suppliers' willingness and ability to pursue sustainability improvement. We conclude by delineating the boundary conditions that define the generalizability of our findings across contexts
An initial map of European intertidal seagrass
No full textSeagrass meadows directly and indirectly provide a wide range of ecosystem services, but their close proximity to anthropogenic activities renders them highly vulnerable. Regardless of their vulnerability and importance as ecosystem health indicators, intertidal seagrasses, unlike their subtidal relatives, have yet to be assessed at continental scales, and current global estimates of seagrass extent either do not mention intertidal seagrasses or combine them with subtidal seagrasses. Here, we present a first of its kind map of intertidal seagrasses in Europe using a harmonised methodology. Over a total intertidal area of 15,100 km, we found that seagrasses cover an area comparable to the combined areas of Paris and Lisbon: 212 18.8 km. The proportion of available intertidal area covered by seagrass decreased towards higher latitudes (from ∼5.4 % at 35°N to ∼0 % at 58°N). Regardless of this pattern, the top three hotspots of seagrass covered the full latitudinal range of Europe, notably in the North Frisian Wadden Sea (Germany), Arcachon Bay (France) and Ria Formosa (Portugal). Furthermore, we showed latitudinal gradients in seagrass cover, with higher cover in low latitudes and lower cover in high latitudes. Finally, an almost linear change in intertidal seagrass peak timing (day of the year of maximum seagrass cover) with latitude was found, with peak timing later the further south. This is a ‘Call-to-Action’ for data that are Findable, Accessible, Interoperable and Reusable (FAIR). We present critical data for prioritising and developing policies, management and protection mechanisms across local, regional or international scales to safeguard these important ecosystems and the societies that depend upon them
Perioperative quantification of clinical bradykinesia measurements in patients with Parkinson's disease using accelerometry
No full textOBJECTIVE: The gold standard for assessing bradykinesia in Parkinson's disease (PD) is the Movement Disorder Society Unified PD Rating Scale (MDS-UPDRS), although it is inherently subjective and relies on experienced raters. Therefore, we translated MDS-UPDRS upper limb bradykinesia assessments into an objective scoring method with 3D-accelerometry.METHODS: In this study, 44 PD-patients and 44 healthy controls (sex and age matched) were included. Two raters assessed MDS-UPDRS criteria for bradykinesia, while accelerometric measurements were conducted at the index fingers. Measurements were performed in an off-medication state. The 3D-acceleration data produced quantitative measures, like frequency and amplitude of hand movement. The algorithm for calculating accelerometric scores was based on MDS-UPDRS criteria with thresholds from healthy controls. Agreement between MDS-UPDRS bradykinesia scores and accelerometric scores was estimated with Cohen's kappa-coefficient (κ).RESULTS: The accelerometric scores showed moderate agreement (κ ≥ 0.548) with MDS-UPDRS ratings. The inter-rater agreement between the two MDS-UPDRS raters was moderate for all tests (κ ≥ 0.595). Accelerometric test-retest reliability was good to excellent (ICC ≥ 0.764, p < 0.001).CONCLUSIONS: This study introduces an accelerometric algorithm to classify upper extremity bradykinesia according to MDS-UPDRS criteria, yielding high test-retest reliability.SIGNIFICANCE: Given its consistency, this method could reduce MDS-UPDRS rater-dependency and improve objective monitoring of upper limb bradykinesia.</p
Towards synthesizing physiologically accurate EEG signals with generative adversarial networks:An exploration of hyperparameter impact on sine samples generation
No full textElectroencephalography (EEG) is important in medical diagnostics and cognitive neuroscience for brain activity recording and understanding, but limited data availability restricts the application of machine learning methods for classification and prediction purposes. Therefore, our ultimate aim is to synthesize physiologically accurate EEG data using Generative Adversarial Networks (GANs). In this preliminary study, we first investigate how the choice of GAN hyperparameters influences the quality of simpler EEG-like generated signals. Thereto, we first created sinusoidal samples as target datasets with single (12 Hz), double (5, 12 Hz), or triple frequencies (5, 12, 28 Hz) and added Gaussian noise. We trained 2,592 models to generate fake sinusoidal samples and examined the effects of varying the hyperparameters batch size, learning rate of generator (LrG) and discriminator (LrD), and the number of epochs for vanilla GAN, Wasserstein GAN (WGAN), and WGAN with Gradient Penalty (WGAN-GP). We compared the distributions of relevant target and fake signal features, such as amplitude, phase, and signal-to-noise ratio, using the Kolmogorov-Smirnov test. Our results showed that WGAN-GP models had better training stability for sinusoidal sample generation compared to WGAN, and significantly outperformed vanilla GAN. Smaller batch sizes accelerated convergence at a fixed number of epochs. Matching LrG and LrD balanced networks during training. Increasing the number of epochs (up to 100) enhanced performance. We conclude that the WGAN-GP model provides the best quality signals, when hyperparameters, particularly learning rate, are well tuned. We therefore propose that effective combinations of hyperparameters enhance training stability and performance of GANs for EEG signal generation.</p
Nature shaping nurture:the impact of parental genotypes on offspring outcomes
No full textThis dissertation examines how parental genotypes can influence their offspring through environmental pathways, a process known as genetic nurture. Although similarities between parents and offspring have traditionally been attributed to either direct genetic transmission (“nature”) or the rearing environment (“nurture”), recent studies show that parental alleles that are not transmitted can also affect offspring through the rearing environment. Using data from the Dutch Lifelines cohort, this work investigates the impact of genetic nurture across a broad set of complex traits. To achieve this aim, the HINTA (Haplotype-based Inference of Non-Transmitted Alleles) method was developed to infer non-transmitted alleles in both parent–offspring trios and parent–offspring pairs. This approach was applied to multiple complex traits, including educational outcomes, depression and anxiety, substance use, and body mass index. Genetic nurture effects were identified for educational outcomes, where non-transmitted parental alleles accounted for a substantial proportion of the direct genetic effect. More modest effects were observed for smoking quantity and body mass index, while little to no evidence emerged for depression, anxiety, alcohol use, smoking initiation, or cannabis use. The findings demonstrate that genetic nurture varies across traits and is generally modest in adulthood. By integrating methodological development with large-scale family data, this dissertation advances the understanding of how parental genotypes shape the environments in which their offspring develop
Profiles of fluctuating prolonged grief disorder reactions using experience sampling data
No full textBACKGROUND: The death of a significant other can lead to prolonged grief disorder (PGD), characterized by intense grief reactions interfering with daily life. This study aims to identify latent profiles, and predictors thereof, of severity and fluctuation indices in PGD using experience sampling methodology (ESM) data from a novel Findable, Accessible, Interoperable, and Reusable (FAIR) data-archive.METHODS: Bereaved adults (N = 228) completed PGD items five times per day for two weeks using ESM. Latent profile analyses were run to identify profiles of fluctuating PGD reactions using four person-specific indicators derived from ESM-PGD items: mean, standard deviation, autocorrelation indicating inertia, and the root mean square of successive differences indicating instability. Background, loss-related characteristics, and psychopathology levels assessed before the ESM period were explored as potential predictors of profile membership.RESULTS: Five distinct profiles were identified: moderate PGD/low inertia/moderate fluctuation (46 %), moderate PGD/low inertia/high fluctuation (20 %), low PGD/low inertia/low fluctuation (19 %), high PGD/low inertia/low fluctuation (8 %), and low PGD/high inertia/low fluctuation (7 %). The most consistent differentiating predictors of profiles were gender, time since loss, and baseline psychopathology levels.CONCLUSIONS: We found that at comparable levels of PGD severity, different fluctuation profiles exist in daily life of bereaved people suggesting that ESM may be an important tool for understanding the dynamic nature of PGD. Identifying these profiles enhances our understanding of individual differences in response to loss and could help to personalize bereavement care.</p
Understanding disease outcome in cancer by integrating large-scale transcriptomic data
No full textCancer remains one of the leading causes of death worldwide. While treatments such as chemotherapy, targeted therapy, and immunotherapy have made great progress, many patients still do not benefit from them. A key challenge is understanding why some patients respond well to treatment while others do not, as these differences are driven by a complex interplay of tumor-intrinsic and patient-specific factors.In this thesis, we applied advanced machine learning methods to analyze large collections of gene expression data from tumors and patient blood samples. These methods helped to uncover hidden biological patterns that capture specific processes related to cancer progression and treatment response.Our studies showed that certain genetic alterations in tumors can affect the immune environment, potentially making tumors more resistant to immunotherapy. We have also identified gene expression patterns in the blood related to how well patients respond to immunotherapy. In breast and ovarian cancers, we found specific patterns associated with treatment benefit and survival, offering opportunities to refine risk assessment and guide therapy decisions. In metastatic breast cancer, we highlighted new molecular features that could contribute to early treatment resistance.Overall, our research demonstrates how large-scale gene expression analysis can identify biological processes associated with cancer progression and treatment response. These findings open the door to more precise, personalized treatment strategies and point to new biological targets that could improve patient outcomes in the future