Abstract
Accessibility and inclusiveness of people with disabilities is a recurring theme that is already perceived as an issue in the field of human rights. Ramps, elevators, among other devices aim at the inclusion of these individuals with limited mobility. Various types of motor limitations, specially partial limitations, are linked to corresponding physical-motor rehabilitation process, with the purpose of reducing or eliminating the patient's dependence on a caregiver or devices for adaptation. Patients with motor disabilities must practice physiotherapeutical exercises along a physician in order to perform body and muscle analysis to ensure the patient's well-being. To reach a more accurate analysis, physiotherapists use a range of devices to acquire patient data, such as the spirometer, to acquire the patient's breath intensity and lung capacity. Similarly, there are other technologies capable of acquiring motion data and quantifying them. This work aims to develop a system that, paired together with an exercise game project (exergame), can acquire and transmit the motion data acquired in-game for an easier and faster analysis of the patient's growth, relying on graphs, tables, and other visual indicators to improve the evaluation of physiotherapeutic treatments. The usage together with an exergame also has benefits such as increased patient compliance with the treatment and improvements in well-being.

Abstract

Abstract
Research and development of autonomous mobile robotic solutions that can perform several active agricultural tasks (pruning, harvesting, mowing) have been growing. Robots are now used for a variety of tasks such as planting, harvesting, environmental monitoring, supply of water and nutrients, and others. To do so, robots need to be able to perform online localization and, if desired, mapping. The most used approach for localization in agricultural applications is based in standalone Global Navigation Satellite System-based systems. However, in many agricultural and forest environments, satellite signals are unavailable or inaccurate, which leads to the need of advanced solutions independent from these signals. Approaches like simultaneous localization and mapping and visual odometry are the most promising solutions to increase localization reliability and availability. This work leads to the main conclusion that, few methods can achieve simultaneously the desired goals of scalability, availability, and accuracy, due to the challenges imposed by these harsh environments. In the near future, novel contributions to this field are expected that will help one to achieve the desired goals, with the development of more advanced techniques, based on 3D localization, and semantic and topological mapping. In this context, this work proposes an analysis of the current state-of-the-art of localization and mapping approaches in agriculture and forest environments. Additionally, an overview about the available datasets to develop and test these approaches is performed. Finally, a critical analysis of this research field is done, with the characterization of the literature using a variety of metrics.

Abstract
The missing data issue is often found in real-world datasets and it is usually handled with imputation strategies that replace the missing values with new data. Recently, generative models such as Variational Autoencoders have been applied for this imputation task. However, they were always used to perform the entire imputation, which has presented limited results when comparing to other state-of-the-art methods. In this work, a new approach called Variational Autoencoder Filter for Bayesian Ridge Imputation is introduced. It uses a Variational Autoencoder at the beginning of the imputation pipeline to filter the instances that are later fitted to a Bayesian ridge regression used to predict the new values. The approach was compared to four state-of-the-art imputation methods using 10 datasets from the healthcare context covering clinical trials, all injected with missing values under different rates. The proposed approach significantly outperformed the remaining methods in all settings, achieving an overall improvement between 26% and 67%.

Abstract
The dropout of university students has been a factor of concern for educational institutions, affecting various aspects such as the institution’s reputation and funding and rankings. For this reason, it is essential to identify which students are at risk. In this study, algorithms based on decision trees and random forests are proposed to solve these problems using real data from 331 students from the University of Trásos-Montes and Alto Douro. In this work with these learning algorithms together with the training strategies, we managed to obtain an 89% forecast of students who may abandon their studies based on the evaluations of both semesters related to the first year and personal data. © 2021, Springer Nature Switzerland AG.

Abstract
Saliency maps have been used as one possibility to interpret deep neural networks. This method estimates the relevance of each pixel in the image classification, with higher values representing pixels which contribute positively to classification. The goal of this study is to understand how the complexity of the network affects the interpretabilty of the saliency maps in classification tasks. To achieve that, we investigate how changes in the regularization affects the saliency maps produced, and their fidelity to the overall classification process of the network. The experimental setup consists in the calculation of the fidelity of five saliency map methods that were compare, applying them to models trained on the CIFAR-10 dataset, using different levels of weight decay on some or all the layers. Achieved results show that models with lower regularization are statistically (significance of 5%) more interpretable than the other models. Also, regularization applied only to the higher convolutional layers or fully-connected layers produce saliency maps with more fidelity.