Search

For healthcare datasets, it is often impossible to combine data samples from multiple sites due to ethical, privacy, or logistical concerns. Federated learning allows for the utilization of powerful machine learning algorithms without requiring the pooling of data. Healthcare data have many simultaneous challenges, such as highly siloed data, class imbalance, missing data, distribution shifts, and non-standardized variables, that require new methodologies to address. Federated learning adds significant methodological complexity to conventional centralized machine learning, requiring distributed optimization, communication between nodes, aggregation of models, and redistribution of models. In this systematic review, we consider all papers on Scopus published between January 2015 and February 2023 that describe new federated learning methodologies for addressing challenges with healthcare data. We reviewed 89 papers meeting these criteria. Significant systemic issues were identified throughout the literature, compromising many methodologies reviewed. We give detailed recommendations to help improve methodology development for federated learning in healthcare.

ABSTRACTIn recent years, there has been a rising awareness about the relevance of spatial ability and integrating spatial information into educational curricula. It is now considered essential to problem-solving and understanding a variety of natural and cultural phenomena. To analyze people’s global-scale cognitive map and which factors influence it, a short playful test was developed that allowed participants to estimate the real land area of certain countries, regions, and continents. These estimates are analyzed to define how accurately the size of countries and continents is perceived. The results show that country of residence does not significantly affect people’s spatial knowledge. However, gender, migration background, and familiarity with maps have a significant impact.

AbstractThis paper addresses the sustainability of academic conferences in terms of their format and their choice of location(s). Environmental sustainability requires the minimisation of travel-related emissions, ideally through the consideration of geography in the location decision process and/or by adopting a multi-venue strategy. Social considerations command a fair distribution of advantages among academic association members in view of the monetary and non-monetary benefits that conferences convey to their host cities, universities and local researchers. This can, inter alia, be achieved through a rotation principle among members. From an economic standpoint, conferences are ideally large gatherings which allow for temporary agglomeration effects, and which should be easily accessible at low costs.Against this backdrop, this paper develops a conceptual framework in which the sustainability outcomes in terms of these format and location mechanisms are hypothesized. Following this, the framework is applied to the case of A ESOP, the Association of European Schools of Planning, to explore to what extent this conference aligns with these considerations. We illustrate how A ESOP qualifies as a rotating and predominantly dispersed single-venue conference, and consider to what extent its sustainability could be improved when other format and location strategies are adopted. We conclude the empirical section by making a case for more experimentation with, and research on, the multi-venue conference format, as it may represent the best option in balancing the three dimensions of sustainability. We argue that spatial planning and research organisations with sustainability aspirations may want to translate these goals more explicitly through their decisions on conference format and location.

ABSTRACTIn geomorphology, PhotoScan is a software that is used to produce Digital Surface Models (DSMs). It constructs 3D environments from 2D imagery (often taken by Unmanned Aerial Vehicles (UAV)) based on Structure-from-Motion (SfM) and Multi-View Stereo (MVS) principles. However, unpublished computer-vision algorithms used, contain random elements which can affect the accuracy of the outputs. For this letter, ten model runs with identical inputs were performed on UAV imagery of a rock glacier to analyse the magnitude of the variation between the different model outputs. This variation was quantified calculating the standard deviation of each cell value in the respective DSMs and derivatives (curvature). Places with steep slope gradients have considerably more DSM variation (up to 10 cm) but stay within the range of the model’s accuracy (10 vertical cm) for 88 – 96% of the area. The edges of the model also show a larger variability (0.10 – 3 m), related to a lower number of overlapping images. These results should be accounted for when performing a geomorphological research at centimetre scale using PhotoScan, especially in areas with a complex relief. Using medium-quality runs, additional oblique viewpoints and respecting a minimum of five overlapping images can minimize the software’s variations.

The goal of this research is to design, use and evaluate a set of web lectures, specifically tailored to the needs of students in higher education who follow geographic information system -related courses. Since education in geographic information system includes theoretical concepts and practical experience, both of these teaching strategies will be implemented in the web lectures. The user-centered design approach is used in the design process to increase the acceptance of the web lectures and the motivation to use them: perceived usefulness and ease of use. The results show that the students appreciate the initial set of web lectures, but that they need to be motivated more to use them (especially when theoretical topics are covered). Students still value the “traditional” face-to-face lectures and see the web lectures as an ideal complement.