213476 research outputs found
Sort by
MuFlex: A Scalable, Physics-based Platform for Multi-Building Flexibility Analysis and Coordination
No full textWith the increasing penetration of renewable generation on the power grid, maintaining system balance requires coordinated demand flexibility from aggregations of buildings. Reinforcement learning (RL) has been widely explored for building controls because of its model-free nature. Open-source simulation testbeds are essential not only for training RL agents but also for fairly benchmarking control strategies. However, most building-sector testbeds target single buildings; multi-building platforms are relatively limited and typically rely on simplified models (e.g., Resistance-Capacitance) or data-driven approaches, which lack the ability to fully capture the physical intricacies and intermediate variables necessary for interpreting control performance. Moreover, these platforms often impose fixed inputs, outputs, and model formats, restricting their applicability as benchmarking tools across diverse control scenarios. To address these gaps, MuFlex, a scalable, open-source platform for multi-building flexibility coordination, was developed. MuFlex enables synchronous information exchange and co-simulation across multiple detailed building models programmed in EnergyPlus and Modelica, and adheres to the latest OpenAI Gym interface, providing a modular, standardized RL implementation. The platform’s physics-based capabilities and workflow were demonstrated in a case study coordinating demand flexibility across four office buildings using the Soft Actor–Critic (SAC) algorithm. The results show that under four buildings’ coordination, SAC effectively reduced the aggregated peak demand by nearly 12% with maintained indoor comfort to ensure the power demand below the threshold. Additionally, the platform’s scalability was investigated through computational benchmarking on building clusters with varying sizes, model types, and simulation programs. The platform is released open-source on GitHub: https://github.com/BuildNexusX/MuFlex
Pregnancy and perinatal outcomes stratified by sFlt-1 and PlGF gestation-specific thresholds in women with suspected preeclampsia
No full textOBJECTIVES: To evaluate gestational age appropriate soluble fms-like tyrosine kinase-1(sFlt-1) and placental growth factor (PlGF) concentrations in determining abnormal outcomes in women with suspected preeclampsia. STUDY DESIGN: Retrospective single-center study of 457 singleton pregnancies from a tertiary referral center in the United Kingdom. Four subgroups were defined using gestational age–adjusted thresholds for sFlt-1 (>95th centile) and PlGF (95th, PlGF 95th, PlGF >5th; n=58), Group 3 (sFlt-1 5th; n=268). Distinct angiogenic and ultrasonographic profiles were observed between groups (one-way ANOVA, F(3,453), p<0.001). On post-hoc Dunnett’s, when compared to controls, Groups 1 and 3 with low PlGF (<5th centile) had a significantly lower mean first trimester PAPP-A (0.6, [95% CI 0.6-0.7] and 0.5, [0.4-7.0] vs 0.9 [0.8-1.0]; df=1, p<0.001) and higher second trimester combined uterine artery Doppler pulsatility index (1.47 [1.35-1.59] and 1.31 [1.13-1.51] vs 0.98 [0.93-1.02]; df=1, p<0.001).
Markers of maternal end-organ involvement also demonstrated strong group-level differences (one-way ANOVA, F(3,453), p95th centile) had a significantly lower mean platelet count (182.8 [168-198.8] and 176.7 [160.4-194.6] vs 212.5 [204.9-220.4]; df=1, p<0.01), higher mean creatinine levels (70.1 [65.2-75.4] and 70.1 [63.3-78.3] vs 59.4 [57.7-61.2]; df=1, p<0.01) and higher mean urine protein:creatinine ratios (93 [68.7-126.1] and 58.9 [38.9-89.0] vs 25.8 [22.2-29.9]; df=1, p<0.01).
The incidence of preeclampsia occurring before 37 weeks was higher in Groups 1-3 when compared to the controls, occurring in 78.0%, 39.7%, 31.3% and 6.7% of pregnancies respectively (p<0.001). Adverse neonatal outcomes were clustered particularly in Group 1 which had the lowest median gestational age at birth (median 33.4 [29.2-35.7] and lowest birthweight centile (0.26 [0.00-2.71]. The incidence of small for gestational age infants and need for neonatal intensive care unit admission was significantly higher across Groups 1-3 when each of the groups was individually compared to the controls (p<0.001 for all). Stillbirths occurred only in Groups 1 (5.3%, n=5) and 3 (6.9%, n=2) which had low PlGF levels <5th centile. CONCLUSION: Stratification of women with suspected preeclampsia using gestational age-specific thresholds for sFlt-1 and PlGF can identify distinct maternal and fetal phenotypes with significantly different clinical outcomes. Elevated sFlt-1 concentrations appear to be strongly associated with severe maternal disease, whilst low PlGF appears to correlate with adverse fetal and neonatal outcomes including growth restriction, stillbirth and prolonged neonatal intensive care stay
Rapidly changing attitudes of university students on the use of generative artificial intelligence
No full textGenerative AI tools such as ChatGPT are increasingly impacting the academic landscape, provoking discussions about their potential role in student learning. A clearer understanding of the extent to which students are adopting generative AI is critical to enable educators and universities to properly adapt to advances in AI technology. This longitudinal study examines the attitudes and usage patterns of first-year undergraduate students towards generative AI and whether these have evolved over a 12-month period. The results indicate a substantial increase in student engagement with generative AI, highlighting a growing familiarisation and confidence in using these tools. Moreover, there is a significant increase in students willing to incorporate generative AI into their studies, suggesting a notable shift towards students recognising the benefits of AI. The rapid evolution of student adoption of generative AI underscores the importance of higher education institutions adapting to and supporting student engagement with generative AI, emphasising the need for flexible institutional policies, targeted technology support frameworks and innovative curriculum designs to align with the evolving competencies and acceptance of AI among students
Invertebrate Automated Phenotyping Platform (INVAPP): An Automated High-Throughput System with Applications in Understanding and Combating Human Diseases
No full textThe nematode Caenorhabditis elegans is a eukaryotic genetic model organism introduced for studies of animal development and behavior (Brenner S, Genetics 77:71-94, 1974). It is also proving useful to expedite our understanding of human diseases and to explore potential therapies (Ahringer J, Curr Opin Genet Dev 7:410-415, 1997; Culetto E, Sattelle DB, Hum Mol Genet 9:869-877, 2000). Monitoring phenotypic changes and the impact of drug candidates is particularly convenient in the case of C. elegans models of neuromuscular or neurological disorders, where changes in motility and growth are often easily observed and can be conveniently assayed. We therefore developed an Invertebrate Automated Phenotyping Platform (INVAPP) together with an algorithm (Paragon) to facilitate such work (Buckingham SD, Partridge FA, Sattelle DB, Int J Parasitol Drugs Drug Resist Int J Parasitol Drugs Drug Resist 4:226-232, 2014; Partridge FA, Brown AE, Buckingham SD, Willis NJ, Wynne GM, Forman R et al., Int J Parasitol Drugs Drug Resist 8:8-21, 2018). Similarly, in the search for novel chemicals to combat invertebrate pathogens, such as parasitic worms, and disease vectors, such as the mosquito that serves as the malaria parasite vector, the phenotyping of worms and insects in the presence of new candidate drugs and control chemicals (anthelmintics and insecticides) can be extremely useful. This is especially important in view of the current challenges in controlling the malaria vector Anopheles gambiae and the soil-transmitted helminth, the whipworm Trichuris trichiura. For example, the development of resistance to the hitherto highly successful pyrethroid insecticides threatens the impressive gains made by the deployment of insecticide-treated nets (ITNs) and indoor residual sprays (IRS) in reducing malaria cases in the period 2000-2015. Also, there is a need for new anthelmintic drugs to combat soil-transmitted helminths such as whipworm, now that the widely used benzimidazoles are becoming much less effective. In both cases, automated phenotyping assays have a role to play. Here, we describe the use of a simple invertebrate automated phenotyping system and provide some examples that illustrate its utility
Inclusive Engagement in Action: LGBTQ+ Perspectives on Research
No full textThis resource reflects on a panel discussion marking LGBTQ+ History Month 2026, which closed IOE’s inaugural Public Engagement Masterclass Series. Showcasing funded and doctoral research projects, the panel explored how LGBTQ+ perspectives are shaping inclusive public engagement across the research lifecycle. From co-production with young people to community-embedded mixed-methods research, the discussion highlighted the importance of visibility, shared decision-making and ethical partnership. This piece considers how inclusive engagement strengthens research design, societal relevance and institutional research culture
