This study presents a comparative evaluation of state-of-the-art Machine Learning (ML) and Explainable Artificial Intelligence (XAI) methods, specifically SHAP and LIME, for predicting electromagnetic field (EMF) strength in urban environments. Using more than 19,000 in situ EMF measurements across Catalonia, Spain, combined with high-resolution geospatial features such as building height, built-up volume, and population density, six ML algorithms were trained and assessed over 50 randomized train–test splits. The k-Nearest Neighbors (kNN) model achieved the highest predictive accuracy (RMSE = 0.623), followed by XGBoost (RMSE = 0.711) and LightGBM (RMSE = 0.717). Explainability analysis showed that SHAP consistently identified built-up volume, building height, degree of urbanization, and population density as the dominant global predictors of EMF strength, whereas LIME revealed that degree of urbanization, population density, and building height were the most influential at the local (micro-scale) level. The results demonstrate that integrating interpretable ML frameworks with enriched geospatial datasets improves both predictive performance and transparency in EMF exposure modeling, supporting data-driven urban planning and public health assessment.
Purpose: Updated evidence for the treatment of obesity in cancer survivors includes behavioural lifestyle interventions underpinning at least one theoretical framework. The aim of this systematic review was to assess the effectiveness of theory-based lifestyle interventions for the treatment of overweight/obesity in breast cancer survivors and to report effective behavioural change techniques (BCTs) and components used in these interventions. Methods: Four databases were searched for RCTs published between database inception and July 2022. The search strategy included MeSH terms and text words, using the PICO-framework to guide the eligibility criteria. The PRISMA guidelines were followed. Risk-of-bias, TIDier Checklist for interventions’ content, and the extent of behaviour change theories and techniques application were assessed. To evaluate the effectiveness of interventions, trials were categorised as “very,” “quite,” or “non” promising according to their potential to reduce body weight, and BCTs promise ratios were calculated to assess the potential of BCTs within interventions to decrease body weight. Results: Eleven RCTs met the inclusion criteria. Seven trials were classified as “very”, three as “quite” and one study was “non” promising. Studies’ size, design, and intervention strategies varied greatly, but the weight-loss goal in all studies was ≥ 5% of the initial body weight through a 500–1000 kcal/day energy deficit and a gradually increased exercise goal of ≥ 30 min/day. Social Cognitive Theory was the most commonly used theory (n = 10). BCTs ranged from 10 to 23 in the interventions, but all trials included behaviour goal setting, self-monitoring, instructions on the behaviour, and credible source. The risk-of-bias was “moderate” in eight studies and “high” in three. Conclusion: The present systematic review identified the components of theory-based nutrition and physical activity behaviour change interventions that may be beneficial for the treatment of overweight/obesity in breast cancer survivors. The strategies mentioned, in addition to reported behavioural models and BCTs, should be considered when developing weight-loss interventions for breast cancer survivors.
The purpose of this study was to evaluate the selection of foraging habitats by the Lesser Kestrel Falco naumanni during the different phases of breeding and the post breeding season on the Plain of Thessaly, central Greece. Foraging habitat selection by the species was recorded by direct visual observation in 2014 and 2015 during their presence in the area. Habitat selection was analyzed using the Savage selectivity index. The results indicate that Lesser Kestrels were likely to positively select fallow and set-aside land from pair formation to the pre-migratory phase, but with some annual variation. Legumes (Alfalfa) and cereals were also positively-selected crop categories, whereas cotton and other kinds of cultivation were significantly avoided. During the incubation and chick rearing phase, cereals (the most widespread form of cultivation in the area), were positively selected, which highlights the importance of these crops during this critical breeding phase. Prey abundance and availability, which depend on farming practices, seem to be the main reasons for the periodic variation in the species' habitat selection in the area.
This study examines the adverse environmental impacts of solar photovoltaic parks located in established protected areas, aiming to determine the level of landscape fragmentation through the calculation of relevant landscape metrics. For this purpose, a case study was carried out in a Mediterranean Natura 2000 Special Protection Area (SPA), and landscape metrics were calculated using Geographic Information System spatial analysis tools. The analysis of metrics showed that the installation of renewable energy parks within the designated protected area negatively affect landscape fragmentation and the absence of carefully defined and evidence-based mitigation measures. The land cover categories that are significantly affected are those considered critical habitats of bird species that have been designated as SPAs. The results of this study highlight the need to integrate, in the National Renewable Energy Spatial Plans, specific biodiversity objectives, such as conservation objectives and the suspension of the installation of photovoltaic parks in certain areas that are important for conservation of biodiversity, in order to ensure the overall sustainability of renewable energy production.
The Lesser Kestrel Falco naumanni is a small migratory falcon, foraging in areas covered by relatively low vegetation. In the Thessalian Plain, Central Greece, it feeds mainly on large Orthoptera and Coleoptera, and is characterized by an opportunistic feeding strategy. The purpose of this study was to investigate the composition of the Lesser Kestrel diet in the Thessalian plain. Systematic visits to two large representative colonies of the Lesser Kestrel in the study area were performed, in order to collect pellets during 2014 and 2015 breeding and post breeding periods of the species. Pellet analysis indicated that Orthoptera and Coleoptera were the main prey categories, which seem to have been the most specialized and dominant feeding choices of the species in the study area, while all other prey categories were rare and not specialized. Lesser Kestrel relied its diet on Orthoptera (Tettigoniidae and Acrididae), mainly during the breeding and post breeding phases, while the feeding strategy of the species during these two phases can be characterized as opportunistic. On the contrary, prior to the breeding phase, main prey category of the species in the study area was Coleoptera (Carabidae and Scarabaeidae). Moreover, a narrower niche breadth of the species based on Levins’ index, was recorded during the breeding phase for both years of the study, indicating that the species restricted the variety of the diet during this phase. Conclusively, the species during its whole breeding season made different prey choices, depending on the breeding phase and its specialized needs during each phase.
(1) Background: The rapid expansion of 4G and 5G networks has raised public concerns about potential health risks associated with electromagnetic radiation. This study addresses these fears by investigating electromagnetic measurements in crowded squares, where human exposure is heightened. Despite rigorous scientific methodologies showing generally low levels of radiation from these technologies, public apprehension persists; (2) Methods: We conducted 496 measurements in Larissa's major squares, using a portable field strength analyzer and GIS mapping. Squares were categorized by traffic and population density; (3) Results: Low-traffic squares exhibited exposure ratios consistently below ICNIRP limits, with median values of 0.0012 and 0.00173 in non-active and active mobile phone cases, respectively. In high-traffic squares, the median exposure ratio was 0.00335 in non-active mobile phone cases, remaining below recommended limits; (4) Conclusions: These findings indicate no significant risk of electromagnetic radiation in crowded squares in Larissa, emphasizing the importance of factual communication and dispelling public misconceptions.