21587 research outputs found
Sort by
From cues to engagement: a comprehensive survey and holistic architecture for computer vision-based audience analysis in live events
No full textThe accurate measurement of audience engagement in real-world live events remains a significant challenge, with the majority of existing research confined to controlled environments like classrooms. This paper presents a comprehensive survey of Computer Vision AI-driven methods for real-time audience engagement monitoring and proposes a novel, holistic architecture to address this gap, with this architecture being the main contribution of the paper. The paper identifies and defines five core constructs essential for a robust analysis: Attention, Emotion and Sentiment, Body Language, Scene Dynamics, and Behaviours. Through a selective review of state-of-the-art techniques for each construct, the necessity of a multimodal approach that surpasses the limitations of isolated indicators is highlighted. The work synthesises a fragmented field into a unified taxonomy and introduces a modular architecture that integrates these constructs with practical, businessoriented metrics such as Commitment, Conversion, and Retention. Finally, by integrating cognitive, affective, and behavioural signals, this work provides a roadmap for developing operational systems that can transform live event experience and management through data-driven, real-time analytics.ALGARVE-FEDER-0118050
Cold agglutinin syndrome in a patient with metastatic breast cancer: a Case report
No full textBackground: Cold agglutinin syndrome (CAS) is a form of autoimmune hemolytic anemia (AIHA), most often associated with lymphoproliferative disorders or infections. Its occurrence in breast cancer is rare and may be triggered by systemic treatment. Case presentation: We report the case of a woman in their fifties diagnosed with breast cancer in 2019. She underwent surgery followed by adjuvant chemotherapy and radiotherapy and subsequently received 3 years of endocrine therapy before developing bone and hepatic metastases. First-line treatment with ribociclib plus letrozole achieved partial response, and fulvestrant was administered at progression. Following further progression, paclitaxel was introduced as third-line metastatic therapy. After four weekly administrations, the patient was admitted to our hospital with severe anemia and diagnosed with CAS. Prompt management and a multidisciplinary approach resulted in partial hematological recovery. Nevertheless, paclitaxel was permanently discontinued, and subsequent therapies provided only transient benefit. The disease continued to progress, her performance status declined, and she ultimately transitioned to exclusive palliative care until death. Conclusion: This case illustrates a rare and severe immune complication of paclitaxel in metastatic breast cancer. The emergence of CAS not only limited systemic options but also reshaped the therapeutic trajectory, highlighting the need for close monitoring during cancer treatments. Early recognition, multidisciplinary approach, and prompt management can provide some improvement, although overall prognosis remains determined by the underlying malignancy.CEECIND/03721/2018/CP1540/CT000
SAMMBA is a high-throughput pipeline for isolating and phenotyping macroalgal strains
No full textDespite successful preservation efforts, macroalgal diversity remains under-represented in global biobanks. A major limitation is the extreme morphological diversity of seaweed thalli, which hinders standardized isolation and phenotyping and often requires taxon-specific protocols. Here we present SAMMBA (Seaweed Automatable Microplate Microscopy for Breeding Approaches), an end-to-end pipeline for the high-throughput isolation, phenotyping and storage of macroalgal propagules in 384-well plates. By optimizing live-cell manipulation for chlorophyll autofluorescence (CAF) imaging and segmentation, multiple unialgal propagules can be isolated by dilution-based workflows. In a single plate, we obtained 68 singlet gametophyte fragments of Laminaria ochroleuca (17.7%) and 60 meiospores of Phyllariopsis purpurascens (31.25%). We demonstrated taxonomic and morphological versatility by isolating 60 unialgal cultures from three distinct Rhodophyta morphotypes (filamentous, crustose and foliose) and 10 strains of Ulva sp., also in a single plate. Furthermore, CAF-based area increase over 30 days enabled high-precision estimates of specific growth rates, yielding 0.130 ± 0.006 and 0.117 ± 0.01 day− 1 for male and female L. ochroleuca gametophytes, respectively (n = 768; p = 1.27e− 53). This precision substantially increases experimental reproducibility and statistical power compared to conventional methods, supporting high-throughput recovery of unialgal strains without motorized platforms, while remaining fully compatible with automation. SAMMBA expands operational capacity for strain discovery and phenotyping, providing a scalable foundation for phenomics, domestication workflows, and standardized macroalgal biobanking. We outline how the platform can benefit multiple areas of phycological research and facilitate the development of improved strains that can support aquaculture and restoration efforts.UID/04326/2025; UID/PRR/04326/202
Deception volcano (Antarctica): an example of advances on ground displacement surveillance in extreme and isolated environments using GNSS satellites
No full textDeception Island is an active, caldera-forming volcano whose surveillance is critically constrained by its extreme Antarctic isolation, scarce resources and limited seasonal human access. This study addresses these challenges by presenting an innovative Global Navigation Satellite Systems (GNSS) geodetic surveillance framework specifically adapted for such a remote environment. Our approach establishes a key operational distinction between non-real-time monitoring and near-real-time surveillance via a semi-continuous wireless network. We effectively resolve the inherent trade-off between promptness and precision by applying signal enhancement methods (e.g. Kalman filtering) to maintain millimetric accuracy in deformation detection, even when utilizing the high sampling rates (1 Hz) essential for rapid assessment. The resulting model allows for the rigorous isolation of the local volcanic signal from the complex regional tectonic kinematics. Crucially, data analysis reveals recurrent 3 year cycles of inflation and deflation in the magmatic system, strongly correlated with seismicity, which validates ground deformation as a reliable volcanic precursor. The primary practical advance is the validation of a dual-term hazard forecasting system: 1) mid-term (months) forecasts based on long-term time-series analysis to facilitate safe inter-campaign operations and 2) short-term (days) forecasting during periods of unrest using ground displacement acceleration, complemented by a magma injection model to predict the spatial location of potential vent openings. This validated and technologically adjusted framework provides an optimized and transferable template for continuous geodetic surveillance in other isolated, active polar volcanoes
Basal progenitors as drivers of neocortical expansion
No full textThe diversification and expansion of distinct progenitor cell subtypes during embryogenesis are essential to form the sophisticated brain structures present in vertebrates. In particular, the emergence of highly proliferative basal progenitors contributed to the evolutionary enlargement of the mammalian neocortex. Basal progenitors are at the center of indirect neurogenesis and can be divided into two main subtypes: the classical TBR2-positive intermediate progenitor cells and the outer radial glial cells, which are especially abundant in gyrencephalic species. While the function of some transcriptomic regulators is conserved across the mammalian clade, recent studies have identified human-specific genes and enhancers that uniquely affect progenitor biology, possibly driving the increased neocortical complexity and disease-susceptibility of the human brain. Here, we review the evolution of basal progenitors, highlighting species-specific traits, molecular drivers of proliferation, and how imbalances in neurogenesis contribute to human brain disorders
From cues to engagement: a comprehensive survey and holistic architecture for computer vision-based audience analysis in live events
No full textThe accurate measurement of audience engagement in real-world live events remains a significant challenge, with the majority of existing research confined to controlled environments like classrooms. This paper presents a comprehensive survey of Computer Vision AI-driven methods for real-time audience engagement monitoring and proposes a novel, holistic architecture to address this gap, with this architecture being the main contribution of the paper. The paper identifies and defines five core constructs essential for a robust analysis: Attention, Emotion and Sentiment, Body Language, Scene Dynamics, and Behaviours. Through a selective review of state-of-the-art techniques for each construct, the necessity of a multimodal approach that surpasses the limitations of isolated indicators is highlighted. The work synthesises a fragmented field into a unified taxonomy and introduces a modular architecture that integrates these constructs with practical, businessoriented metrics such as Commitment, Conversion, and Retention. Finally, by integrating cognitive, affective, and behavioural signals, this work provides a roadmap for developing operational systems that can transform live event experience and management through data-driven, real-time analytics.ALGARVE-FEDER-0118050
Optimizing the input: can large language models standardize radiology requisitions?
No full textRadiology stands as a central pillar of modern healthcare, non-invasively visualizing anatomy and physiology to guide critical diagnostic and treatment decisions. Over the last decade, the radiology community has made significant strides in standardizing its “outputs,” the radiology report [1]. Through initiatives like the various Reporting and Data Systems (RADS) for breast, liver, prostate, and thyroid imaging, we have improved communication, reduced ambiguity, and enhanced the clinical utility of our findings. However, a high-quality output depends fundamentally on a high-quality input
Seasonal, spatial, and high-frequency monitoring of dissolved oxygen and net ecosystem metabolism in a shallow coastal lagoon, Ria Formosa – Portugal
No full textThis study provides a detailed assessment of dissolved oxygen (DO) dynamics and net ecosystem metabolism (NEM) in the Ria Formosa coastal lagoon, highlighting how spatial and temporal variability shape mesotidal ecosystems. Complementary approaches were applied: short-term (24 h) diel oxygen open-water method at two stations (eastern and western boundaries) in comparison with the bottle incubation method, and a 2.5-year high-frequency dataset at an inner station lagoon. DO showed clear seasonal patterns, largely driven by photosynthesis and respiration, but strongly modulated by water circulation and hydrodynamics. Sporadic hypoxic events (DO ≤ 2 mg L−1) were observed but do not represent a risk. The boundary stations are shallower and colonized by submerged vegetation, displayed stronger diel fluctuations than at the inner station, underscoring the role of biological processes in lagoon metabolism. NEM estimates revealed strong spatial contrasts: the western station was autotrophic, while the eastern station predominantly attributed to restricted water exchange. Long-term records from the inner station indicated a slightly heterotrophic status, reinforcing the value of sustained high-frequency monitoring for capturing ecosystem trends overlooked by short-term approaches. The results also highlight lagoon–coastal ocean connectivity, suggesting dual exchanges: local production may be exported offshore to sustain adjacent coastal areas, while coastal processes, such as upwelling, can also influence lagoon metabolism. Methodological comparisons confirmed that bottle incubation method underestimated NEM relative to diel oxygen open-water method, emphasizing the role of hydrodynamics in mesotidal system metabolism. Although sampling focused on western and eastern boundaries and inner lagoon, the strong tidal renewal in the main and secondary channels suggests Ria Formosa overall is close to metabolic balance, with spatial heterogeneity shaped by hydrodynamics, vegetation, and coastal forcing. By integrating short-term, high-frequency, and long-term observations, this study advances understanding of metabolism in coastal lagoons, offering key insights for predicting ecosystem responses to climate change and for guiding management of vulnerable coastal environments.UID/00350/202
A reforma da lei laboral é uma prioridade?
No full textPor uma atualização pontual da legislação do trabalho e foco na Reforma do Estado e da Administração Pública para uma redução significativa nos custos de contexto.FEUAlg - Faculdade de Economia da Universidade do Algarv
Microplastics hot spots at the South Iberian Margin
No full textThe persistent accumulation of microplastics (MPs) in sediments poses ecological risks to benthic organisms and contributes to the broader issue of marine pollution. This study quantitatively analysed MPs in sediments, water and mussels Mytilus galloprovincialis from eleven sites of the South Portuguese coast in two contrasting climatic seasons (summer and winter). MPs were detected in sediments, water, and M. galloprovincialis at all study sites, although their abundance, colour, size, and type varied across compartments, locations, and seasons. Three hot spots of MPs contamination were identified at the South Portuguese coast. In these areas, the concentration of MPs in sediments was three orders of magnitude higher than water and mussels. The MPs identified had distinct colour patterns: transparent particles dominated in sediments, while blue was the most common in water and mussels. A size-dependent accumulation pattern was observed in the sediments, suggesting selective retention of MPs according to natural particle size, and a relationship was observed between MP levels in sediments and in mussels. Polypropylene (PP) and polyethylene (PE) were dominant polymers in sediments while PP, PE and polyethylene terephthalate (PET) were consistently present in water and mussels. Polybuthyl methachrylate (PBMA) was also detected in surface water and ingested by mussels. These findings suggest that variability in MPs abundance and polymer composition is linked to differences in local human activities. They also provide strong evidence for the importance of controlling land-based sources of MPs, particularly those transported to the coastal area by transported by rivers.CEECINSTLA/00018/202