Abstract
Deep Neural Networks using histopathological images as an input currently embody one of the gold standards in automated lung cancer diagnostic solutions, with Deep Convolutional Neural Networks achieving the state of the art values for tissue type classification. One of the main reasons for such results is the increasing availability of voluminous amounts of data, acquired through the efforts employed by extensive projects like The Cancer Genome Atlas. Nonetheless, whole slide images remain weakly annotated, as most common pathologist annotations refer to the entirety of the image and not to individual regions of interest in the patient's tissue sample. Recent works have demonstrated Multiple Instance Learning as a successful approach in classification tasks entangled with this lack of annotation, by representing images as a bag of instances where a single label is available for the whole bag. Thus, we propose a bag/embedding-level lung tissue type classifier using Multiple Instance Learning, where the automated inspection of lung biopsy whole slide images determines the presence of cancer in a given patient. Furthermore, we use a post-model interpretability algorithm to validate our model's predictions and highlight the regions of interest for such predictions.

Abstract
This paper explores the problem of model-based detecting and reconstructing occurring actuator and sensor faults in direct current (DC) microgrids (MGs) connected to resistive and constant power loads (CPLs) and energy storage units. Both the actuator and sensor faults are modeled as an additive time-varying term in the state-space representation, which highly degrade the system response performance if they are not compensated. In this paper, a novel advanced extended Kalman filter (EKF), called dualEKF (D-EKF) is proposed to estimate the system states as well as the accruing actuator and sensor faults. The main property of the developed approach is that it offers a systematic estimation procedure by dividing the estimating parameters into three parts and these parts are estimated in parallel. A first-order filter is utilized to turn the sensor faulty system into an auxiliary sensor faults-free representation. Thereby, the artificial output contains the filter states. The proposed D-EKF estimator does not require restrictive assumptions on the power system matrices and is highly robust against stochastic Gaussian noises. At the end, the proposed approach is applied on a practical faulty DC MG benchmark connected to a CPL, a resistive load, and an energy storage system and the obtained simulation results are analyzed form the accuracy and convergence speed viewpoints.

Abstract
Tourism is an information-intensive sector and today's tourist is hungry for information about everything that surrounds him and is increasingly demanding about the mechanisms that are made available for access and interaction with information. This new reality requires rethinking many of the existing solutions. In this context, the Internet of Things (IoT) is revolutionizing the way we think, design and implement Information and Communication Technologies (ICT) solutions for the tourism sector, opening up unprecedented opportunities in terms of how we can provide information and services. This new reality is enabling reengineering the interaction-process between tourists and its surrounding space. For heritage spaces, typically visited by countless tourists, there is an opportunity to rethink the entire process of supporting the interpretation and fruition of heritage, carried out by tourists. In order to understand how this reengineering can be carried out, a review of the state of the art is carried out with regard to how the IoT has been applied in the context of tourism. Then, the methodology that governed the creation of a conceptual model based on IoT is clearly defined, capable of transforming the way physical spaces of tourist interest can be interpreted and how their fruition can be improved. Particular importance is given to the contextualization of the experience, since the information provided must be adjusted to the visitor, according to their profile, which may necessarily reflect different types of interest or prior knowledge about the space. Finally, this article presents a conceptual model where its components are described and where it is discussed how the model can transform the experience of visiting touristic spaces and how tourists can access information and services that entities promoters of these spaces wish to make available. In the dissertation carried out, important aspects of the model and the gains it may generate for the revitalization and promotion of heritage are discussed.

Abstract
Serious Games have been used in professional training to increase employee engagement and improve the results of training initiatives in this context. This work intends to investigate the influence of game elements, in adaptable Serious Games, according to the users' interactions, in the increase of engagement in the game itself and, as the main objective, in the learning outcomes and the transfer of the acquired knowledge and practised skills to activities in the daily work. Using the Design Science Research methodology, this study is intended to develop a framework for the development and evaluation of Serious Games to improve the user experience, the learning outcomes, the transfer of knowledge to work situations, and the application of skills practised in the game in real professional scenarios. © 2021 Associacao Portuguesa de Sistemas de Informacao. All rights reserved.

Abstract
Traditional post-hoc high-fidelity scientific visualization (HSV) of numerical simulations requires multiple I/O check-pointing to inspect the simulation progress. The costs of these I/O operations are high and can grow exponentially with increasing problem sizes. In situ HSV dispenses with costly check-pointing I/O operations, but requires additional computing resources to generate the visualization, increasing power and energy consumption. In this paper we present LOOM, a new interweaving approach supported by a task scheduling framework to allow tightly coupled in situ visualization without significantly adding to the overall simulation runtime. The approach exploits the idle times of the numerical simulation threads, due to workload imbalances, to perform the visualization steps. Overall execution time (simulation plus visualization) is minimized. Power requirements are also minimized by sharing the same computational resources among numerical simulation and visualization tasks. We demonstrate that LOOM reduces time to visualization by 3 × compared to a traditional non-interwoven pipeline. Our results here demonstrate good potential for additional gains for large distributed-memory use cases with larger interleaving opportunities. © 2021 ACM.

Abstract