Abstract
Gastric Intestinal Metaplasia (GIM) is a precancerous gastric lesion and its early detection facilitates patient followup, thus lowering significantly the risk of death by gastric cancer. However, effective screening of this condition is a very challenging task, resulting low intra and inter-observer concordance. Computer assisted diagnosis systems leveraging deep neural networks (DNNs) have emerged as a way to mitigate these ailments. Notwithstanding, these approaches typically require large datasets in order to learn invariance to the extreme variations typically present in Esophagogastroduodenoscopy (EGD) still frames, such as perspective, illumination, and scale. Hence, we propose to combine a priori information regarding texture characteristics of GIM with data-driven DNN solutions. In particular, we define two different models that treat pre-trained DNNs as general features extractors, whose pairwise interactions with a collection of highly invariant local texture descriptors grounded on fractal geometry are computed by means of an outer product in the embedding space. Our experiments show that these models outperform a baseline DNN by a significant margin over several metrics (e.g., area under the curve (AUC) 0.792 vs. 0.705) in a dataset comprised of EGD narrow-band images. Our best model measures double the positive likelihood ratio when compared to a baseline GIM detector.

Abstract
Deep learning approaches require a large amount of data to be transferred to centralized entities. However, this is often not a feasible option in healthcare, as it raises privacy concerns over sharing sensitive information. Federated Learning (FL) aims to address this issue by allowing machine learning without transferring the data to a centralized entity. FL has shown great potential to ensure privacy in digital healthcare while maintaining performance. Despite this, there is a lack of research on the impact of different types of data heterogeneity on the results. In this study, we research the robustness of various FL strategies on different data distributions and data quality for glaucoma diagnosis using retinal fundus images. We use RetinaQualEvaluator to generate quality labels for the datasets and then a data distributor to achieve our desired distributions. Finally, we evaluate the performance of the different strategies on local data and an independent test dataset. We observe that federated learning shows the potential to enable high-performance models without compromising sensitive data. Furthermore, we infer that FedProx is more suitable to scenarios where the distributions and quality of the data of the participating clients is diverse with less communication cost.

Abstract
This paper is poised to inform educators, policy makers and software developers about the untapped potential of PWAs in creating engaging, effective, and personalized learning experiences in the field of programming education. We aim to address a significant gap in the current understanding of the potential advantages and underutilisation of Progressive Web Applications (PWAs) within the education sector, specifically for programming education. Despite the evident lack of recognition of PWAs in this arena, we present an innovative approach through the Framework for Gamification in Programming Education (FGPE). This framework takes advantage of the ubiquity and ease of use of PWAs, integrating it with a Pareto optimized gamified programming exercise selection model ensuring personalized adaptive learning experiences by dynamically adjusting the complexity, content, and feedback of gamified exercises in response to the learners' ongoing progress and performance. This study examines the mobile user experience of the FGPE PLE in different countries, namely Poland and Lithuania, providing novel insights into its applicability and efficiency. Our results demonstrate that combining advanced adaptive algorithms with the convenience of mobile technology has the potential to revolutionize programming education. The FGPE+ course group outperformed the Moodle group in terms of the average perceived knowledge (M = 4.11, SD = 0.51).

Abstract
Chest radiography is increasingly used worldwide to diagnose a series of illnesses targeting the lungs and heart. The high amount of examinations leads to a severe burden on radiologists, which benefit from the introduction of artificial intelligence tools in clinical practice, such as deep learning classification models. Nevertheless, these models are undergoing limited implementation due to the lack of trustworthy explanations that provide insights about their reasoning. In an attempt to increase the level of explainability, the deep learning approaches developed in this work incorporate in their decision process eye-tracking data collected from experts. More specifically, eye-tracking data is used in the form of heatmaps to change the input to the selected classifier, an EfficientNet-b0, and to guide its focus towards relevant parts of the images. Prior to the classification task, UNet-based models are used to perform heatmap reconstruction, making this framework independent of eye-tracking data during inference. The two proposed approaches are applied to all existing public eye-tracking datasets, to our knowledge, regarding chest X-ray screening, namely EGD, REFLACX and CXR-P. For these datasets, the reconstructed heatmaps highlight important anatomical/pathological regions and the area under the curve results are comparable to the state-of-the-art and to the considered baseline. Furthermore, the quality of the explanations derived from the classifier is superior for one of the approaches, which can be attributed to the use of eye-tracking data.

Abstract
Road transportation emissions have increased in the last few decades and have been the primary source of pollutants in urban areas with ever-growing populations. In this context, it is important to have effective measures to monitor road emissions in regions. Creating an emission inventory over a region that can map the road emission based on the vehicle trips can be helpful for this. In this work, we show that it is possible to use raw GPS data to measure levels of pollution in a region. By transforming the data using feature engineering and calculating the vehicle-specific power (VSP), we show the areas with higher emissions levels made by a fleet of taxis in Porto, Portugal. The Uber H3 grid system is used to decompose the city into hexagonal grids to sample nearby data points into a region. We validate our experiments on real-world sensor datasets deployed in several city regions, showing the correlation with VSP and true values for several pollutants attesting to the method's usefulness.

Abstract
An online data-driven predictive maintenance approach for railway switches using data logs obtained from the interlocking system of the railway infrastructure is proposed in this paper. The proposed approach is detailed described and consists of a two-phase process: anomaly detection and remaining useful life prediction. The approach is applied to and validated in a real case study, the Metro do Porto, from which seven months of data is available. The approach has been revealed to be satisfactory in detecting anomalies. The results open the possibilities for further studies and validation with a more extensive dataset on the remaining useful life prediction.