43569 research outputs found
Sort by
A Feminist Analysis of Gender Bias in Artificially Created and Human News Articles
No full textTreball de fi de màster Estudis Internacionals sobre Mitjans, Poder i DiversitatSupervisora: Pilar Medina-BravoThis analysis compares the different kinds of gender bias apparent in human-written and artificially created News articles. Focusing on the 2022 FIFA World Cup in Qatar, articles published by EuroNews on that topic get used to generate similar articles produced by ChatGPT, Perplexity, and DeepAI. By applying a feminist qualitative content analysis, the study explores how gendered language, stereotypes or forms of discrimination can manifest in both types of texts. While results show that both human and AI-generated articles reflect some forms of existing gender bias, the AI-generated texts offered the readers more context on how those inequalities could be dangerous for marginalized groups and included more diverse voices. The outputs by the different AI-models were generally less biased, most likely through content moderation and guidelines inside the algorithms. The findings highlight the mutual shaping of technology and society and urges for more transparency and diversity in AI development
Multi-center Fetal Brain Tissue Annotation (FeTA) challenge 2022 results
No full textSegmentation is a critical step in analyzing the developing human fetal brain. There have been vast improvements in automatic segmentation methods in the past several years, and the Fetal Brain Tissue Annotation (FeTA) Challenge 2021 helped to establish an excellent standard of fetal brain segmentation. However, FeTA 2021 was a single center study, limiting real-world clinical applicability and acceptance. The multi-center FeTA Challenge 2022 focused on advancing the generalizability of fetal brain segmentation algorithms for magnetic resonance imaging (MRI). In FeTA 2022, the training dataset contained images and corresponding manually annotated multi-class labels from two imaging centers, and the testing data contained images from these two centers as well as two additional unseen centers. The multi-center data included different MR scanners, imaging parameters, and fetal brain super-resolution algorithms applied. 16 teams participated and 17 algorithms were evaluated. Here, the challenge results are presented, focusing on the generalizability of the submissions. Both in- and out-of-domain, the white matter and ventricles were segmented with the highest accuracy (Top Dice scores: 0.89, 0.87 respectively), while the most challenging structure remains the grey matter (Top Dice score: 0.75) due to anatomical complexity. The top 5 average Dices scores ranged from 0.81-0.82, the top 5 average 95th percentile Hausdorff distance values ranged from 2.3-2.5mm, and the top 5 volumetric similarity scores ranged from 0.90-0.92. The FeTA Challenge 2022 was able to successfully evaluate and advance generalizability of multi-class fetal brain tissue segmentation algorithms for MRI and it continues to benchmark new algorithms
Catalan long-term care system reaches the majority of age: what have we learned?
No full textCatalonia is one of the regions with the highest life expectancy among OECD countries, and it is expected to remain so in the coming decades. Currently, only half of this total life expectancy at age 65 is spent in good health. Far from healthy aging, in 2021, around 66% of the population over 65 did not rate their health status as good or very good, and 49% reported experiencing limitations in carrying out daily living activities (ADL) (OECD, 2024).
In this context, the Spanish Long-Term Care System (LTCS), which represents the fourth pillar of the welfare state, is coming of age. The main objective of this text is to present the key lessons learned during this period based on a wide range of analyses of the system. To contextualize this evidence, the evolution of the Spanish LTCS is first reviewed, presenting its main characteristics and comparing them with those of its European counterparts. Secondly, the main insights gained from research conducted on the Spanish LTCS are presented.
This research has focused on different aspects. One area of study has examined the –precarious- financial sustainability of the Spanish LTCS, identifying chronic underfunding and a flawed design that have limited the positive effects of benefits and increased the burden on families. On the other hand, the design and implementation of the Spanish LTCS ensure equitable access to and navigation within the system. However, inequities
have been documented in the type of benefits received, as there is a higher demand for service-based benefits among beneficiaries with lower economic status.
The economic impact of the Spanish LTCS has also been explored, revealing significant returns, particularly through job creation. Another aspect analysed is the effects of long-term care benefits on beneficiaries and their families. In this regard, research has shown that the introduction of the system has reduced hospitalizations, especially those resulting from conditions that could be avoided with adequate long-term care. Additionally, Spanish LTCS benefits have led to lower savings levels and an increase in the supply of informal caregivers, improving their well-being and increasing the likelihood of early retirement.
Finally, the COVID-19 pandemic exposed critical weaknesses in residential care, including insufficient funding and staffing shortages, which contributed to high mortality rates during this period
RF-EMF exposure assessment with add-on uplink exposure sensor in different microenvironments in seven European countries
No full textIntroduction: Several devices have been developed to assess exposure to radiofrequency electromagnetic field (RF-EMF). Since the existing solutions to measure the personal exposure induced by emerging 5G New Radio (NR) are expensive, complex, and bulky, a new cost efficient and low-complexity sensor is developed, that aims to measure RF-EMF exposure in different scenarios of data transmission within different areas. Methods: With this novel sensor, activity-based microenvironmental surveys were conducted across seven European countries: Belgium, Hungary, Italy, Poland, Switzerland, the Netherlands, and the United Kingdom. The device is attached to a smartphone to quantify the auto-induced uplink (a-UL) transmission component of the total exposure for a broadband frequency range from 100 MHz to 6000 MHz and is thus denoted as add-on sensor. In-situ measurements were performed for three usage scenarios, namely non-user (i.e., environmental exposure), maximum downlink (max DL), and maximum uplink (max UL) scenarios, in a large city, a secondary city, and three rural villages a priori selected within each country. Results: Power levels were lowest in non-user scenarios (median: -2.64 dBm or 0.54mW), increasing by a factor of 5.00 dB in maximum downlink scenarios and by a factor of 14.15 dB in maximum uplink scenarios. In the maximum uplink scenarios, the highest median a-UL power of 18.68dBm (= 73.79 mW) was recorded in The Netherlands, while the lowest median a-UL power of 4.77dBm (= 3 mW) was observed in the UK. The analysis of the measured data showed a prominent trend of a 2.72 dB lower power in the cities compared to the villages. Further comparisons were made based on microenvironment groups, where the lowest a-UL power levels (median: 12.35dBm) were measured in outdoor areas, with an increase of 1.78 dB and 1.91 dB in power was measured compared to public transport and public places, respectively. Conclusion: This study compares RF-EMF power levels between different countries, urbanization settings, and usage scenarios, which is important for future epidemiological studies
Ressenya de Elizalde, María Dolores, Huetz de Lemps, Xavier, Rodrigo y Alharilla, Martín y De Llobet, Ruth, Los Roxas. Filipinas en el siglo XIX a través de una familia hispano-filipina, Madrid, Marcial Pons, 2020, 470 págs., ISBN: 978-84-17945-11-4.
No full textPhylogenetic signal in primate tooth enamel proteins and its relevance for paleoproteomics
No full textAncient tooth enamel, and to some extent dentin and bone, contain characteristic peptides that persist for long periods of time. In particular, peptides from the enamel proteome (enamelome) have been used to reconstruct the phylogenetic relationships of fossil taxa. However, the enamelome is based on only about 10 genes, whose protein products undergo fragmentation in vivo and post mortem. This raises the question as to whether the enamelome alone provides enough information for reliable phylogenetic inference. We address these considerations on a selection of enamel-associated proteins that has been computationally predicted from genomic data from 232 primate species. We created multiple sequence alignments for each protein and estimated the evolutionary rate for each site. We examined which sites overlap with the parts of the protein sequences that are typically isolated from fossils. Based on this, we simulated ancient data with different degrees of sequence fragmentation, followed by phylogenetic analysis. We compared these trees to a reference species tree. Up to a degree of fragmentation that is similar to that of fossil samples from 1 to 2 million years ago, the phylogenetic placements of most nodes at family level are consistent with the reference species tree. We tested phylogenetic analysis on combinations of different enamel proteins and found that the composition of the proteome can influence deep splits in the phylogeny. With our methods, we provide guidance for researchers on how to evaluate the potential of paleoproteomics for phylogenetic studies before sampling valuable ancient specimens
Detecting markov random fields hidden in white noise
No full textMotivated by change point problems in time series and the detection of textured objects in images, we consider the problem of detecting a piece of a Gaussian Markov random field hidden in white Gaussian noise. We derive minimax lower bounds and propose near-optimal tests.Ery Arias-Castro supported in part by the US National Science Foundation (DMS-1223137, DMS-1120888). Gábor Lugosi supported by the Spanish Ministry of Economy and Competitiveness, Grant MTM2015-67304-P and FEDER, EU. Nicolas Verzelen supported in part by the French Agence Nationale de la Recherche (ANR 2011 BS01 010 01 projet Calibration)
Early synaptic changes and reduced brain connectivity in PD-like mice with depressive phenotype
No full textAnxiety and depression are common in Parkinson’s disease (PD), affecting quality of life. Aggregates of α-synuclein (α-Syn) are found in serotonergic (5-HT) raphe nuclei early in the disease, but their relationship to brain changes is unclear. We investigated synaptic plasticity, neuronal activity, and functional magnetic resonance imaging (fMRI)-based brain connectivity in a PD-like mouse model with depressive phenotype. AAV-induced human α-Syn accumulation in raphe 5-HT neurons causes progressive synaptic pathology in interconnected brain regions. This is marked by lower MAP-2, PSD95 and higher SV2A, VAMP2, which are key to synaptic structure and function, as confirmed in human brain tissue samples. Abnormalities in Egr-1-dependent neuronal activity and region-specific differences in resting-state functional brain activity were also detected eight weeks post-AAV infusion, before neurodegeneration. This provides evidence for synaptic and fMRI markers associated with α-Syn pathology in emotional brain circuits, and has translational importance for identifying PD patients at risk for depression.This work was supported by MCIU/AEI/FEDER, UE grant (PID2019-105136RB-100, PID2022-141700OB-I00, MCIN/AEI/10.13039/501100011033 to A.B.), and AGAUR 2021-SGR-01358, Catalan Government. CB/07/09/0034 Center for Networked Biomedical Research on Mental Health (CIBERSAM). We also thank the Spanish Stress Research Network, MCIN/AEI/10.13039/501100011033. U.S.S. is a recipient of a fellowship from the Non- Doctor Researcher Formation Pre-doctoral Program of Basque Government, Spain. U.A. is a recipient of a fellowship from the State Research Agency for the training of predoctoral research personnel (PREP2022-000676) associated to a generation of knowledge project (PID2022-141700OB-I00). The authors would like to thank the staff members of the Biobank of the Clinic Hospital Barcelona - IDIBAPS for their collaboration in the study
Genomic signatures of island colonisation in highly diverse primates
No full textData de publicació electrònica: 09-07-2025Understanding how small populations cope with loss of genetic diversity and deleterious variation is crucial to address the current biodiversity crisis. Insular populations are particularly interesting as they have often persisted at lower population sizes and higher inbreeding than their mainland counterparts. While the genome-wide consequences of inbreeding in threatened insular species have received some attention, comparative genomics between insular and mainland populations of widespread and genetically diverse species have rarely been performed. Yet, they are particularly well suited to inform about the consequences of drastic population declines from initially large populations-a phenomenon that is becoming increasingly common. The spot-nosed monkey (Cercopithecus petaurista), the Campbell's monkey (Cercopithecus campbelli) and the green monkey (Chlorocebus sabaeus) are common and genetically diverse West African primates. Insular populations can be found at the Bijagós Archipelago, Guinea-Bissau. Here, we assessed the genome-wide diversity, inbreeding, genetic load and adaptive variation using whole genome sequencing data from insular and mainland populations. In the three species, island populations showed lower genome-wide diversity and higher inbreeding. Genetic drift has likely promoted the conversion of masked genetic load into realised load without increased purging of deleterious variation. Additionally, we found no evidence for accumulation of deleterious variation, suggesting that these populations are not yet at risk of extinction by genetic factors and may act as reservoirs of extant mainland genetic diversity. We highlight, however, that other anthropogenic factors are threatening these insular primates, and therefore conservation management should target their immediate threats and safeguard against additional loss of diversity
Abogacía y blanqueo de capitales: asesoramiento letrado en contextos de criminalidad económica
No full textTreball de Fi de Grau en Dret. Curs 2024-2025Tutora: Beatriz Goena VivesEl presente trabajo analiza el papel del abogado en el marco del delito de blanqueo de capitales, prestando especial atención a los límites entre su deber de secreto profesional y las obligaciones de colaboración. El objetivo principal es poder determinar hasta qué punto colisionan los derechos del letrado con los deberes de diligencia e información impuestos por la normativa de prevención del blanqueo. Para ello se examina la normativa española, jurisprudencia nacional y europea, así como las posibles implicaciones del cobro de honorarios procedentes de actividades ilícitas. Se incluye una comparativa entre modelos europeos, concretamente Alemania y Francia, concluyendo con anexo final correspondiente a una propuesta de reglamento que está orientada a clarificar y reforzar el cumplimiento normativo sin comprometer las garantías profesionales del abogado en esta materia