Abstract
Developing autonomous assistants to help with domestic tasks is a vital topic in robotics research. Among these tasks, garment folding is one of them that is still far from being achieved mainly due to the large number of possible configurations that a crumpled piece of clothing may exhibit. Research has been done on either estimating the pose of the garment as a whole or detecting the landmarks for grasping separately. However, such works constrain the capability of the robots to perceive the states of the garment by limiting the representations for one single task. In this paper, we propose a novel end-to-end deep learning model named GarmNet that is able to simultaneously localize the garment and detect landmarks for grasping. The localization of the garment represents the global information for recognising the category of the garment, whereas the detection of landmarks can facilitate subsequent grasping actions. We train and evaluate our proposed GarmNet model using the CloPeMa Garment dataset that contains 3,330 images of different garment types in different poses. The experiments show that the inclusion of landmark detection (GarmNet-B) can largely improve the garment localization, with an error rate of 24.7% lower. Solutions as ours are important for robotics applications, as these offer scalable to many classes, memory and processing efficient solutions.

Abstract
This study presents and applies combined phase and magnitude metrics for validation of multibody dynamics (MBD) estimated muscle actions with simultaneous registered sEMG of lower limb muscles. Subject-specific tests were performed for acquisition of ground reaction forces and kinematic data from joint reflective markers during NG, SKG and SR. Inverse kinematics and dynamics was performed using AnyBody musculoskeletal personalized modeling and simulation. MBD estimated muscle activity (MA) of soleus medialis (SM) and tibialis anterior (TA) were compared on phase, magnitude and combined metric with simultaneous acquisition of sEMG for the same muscles. Results from quantitative metrics presented better agreement between MDB MA and sEMG on phase (P) than on magnitude (M) with combined (C) metric following the same pattern as the magnitude. Soleus medialis presented for specific subject lower P and M error on NG and SKG than at SR with similar P errors for tibialis anterior and higher error on M for TA at NG and SKG than SR. Separately and combined quantitative metrics of phase and magnitude presents as a suitable tool for comparing measured sEMG and MBD estimated muscle activities, contributing to overcome qualitative and subjective comparisons, need for intensive observer supervision, low reproducibility and time consuming.

Abstract
Delineation of the cardiac structures from 2D echocardiographic images is a common clinical task to establish a diagnosis. Over the past decades, the automation of this task has been the subject of intense research. In this paper, we evaluate how far the state-of-the-art encoder-decoder deep convolutional neural network methods can go at assessing 2D echocardiographic images, i.e., segmenting cardiac structures and estimating clinical indices, on a dataset, especially, designed to answer this objective. We, therefore, introduce the cardiac acquisitions for multi-structure ultrasound segmentation dataset, the largest publicly-available and fully-annotated dataset for the purpose of echocardiographic assessment. The dataset contains two and four-chamber acquisitions from 500 patients with reference measurements from one cardiologist on the full dataset and from three cardiologists on a fold of 50 patients. Results show that encoder-decoder-based architectures outperform state-of-the-art non-deep learning methods and faithfully reproduce the expert analysis for the end-diastolic and end-systolic left ventricular volumes, with a mean correlation of 0.95 and an absolute mean error of 9.5 ml. Concerning the ejection fraction of the left ventricle, results are more contrasted with a mean correlation coefficient of 0.80 and an absolute mean error of 5.6%. Although these results are below the inter-observer scores, they remain slightly worse than the intra-observer's ones. Based on this observation, areas for improvement are defined, which open the door for accurate and fully-automatic analysis of 2D echocardiographic images.

Abstract
The localization algorithms have different errors which can impair the robot's navigation. In this way, we propose an approach that will supervise the localization while the robot navigate. Our approach is based on another work present in the literature, where we detected a problem during its analysis. Therefore, this article will present a new method based on the RLS algorithm, to solve the identified problem. Besides, we propose the supervision of two more localization algorithms, being now four the supervised algorithms, namely: Augmented Monte Carlo Localization, Extended Kalman Filter with Beacons, Perfect Match and Odometry. The results show that the robustness and reliability of the system were increased.

Abstract
We introduce the concept of the envelope dyadic Green's function (EDGF) and present a formalism to study the propagation of electromagnetic fields with slowly varying amplitude (EMFSVA) in dispersive anisotropic media with two dyadic constitutive parameters: the dielectric permittivity and the magnetic permeability. We find the matrix elements of the EDGFs by applying the formalism for uniaxial anisotropic metamaterials. We present the relations for the velocity of the EMFSVA envelopes which agree with the known definition of the group velocity in dispersive media. We consider examples of propagation of the EMFSVA passing through active and passive media with the Lorentz and the Drude type dispersions, demonstrating beam focusing in hyperbolic media and superluminal propagation in media with inverted population. The results of this paper are applicable to the propagation of modulated electromagnetic fields and slowly varying amplitude fluctuations of such fields through frequency dispersive and dissipative (or active) anisotropic metamaterials. The developed approach can be also used for the analysis of metamaterial-based waveguides, filters, and delay lines.

Abstract
We face a technology growth period that promotes STEM (Science, Technology, Engineering and Mathematics) as basic set of competences in diversified professional backgrounds. However, the process of teaching this areas is complex and abstract. We propose using game-based learning (GBL) for increased engagement and efficacy. More specifically, we use pervasive games as the drive of the learning process, promoting challenges in the context of the learner. To develop pervasive games, we require mobile devices (for location-based games) and a set of functionalities regarding location, time, and context of the user. In this article, we present the design of a mobile app to implement pervasive games. The perks of the mobile devices will be used to get unique and engaging challenges, to stimulate learning. Therefore, the learning process will become more enjoyable, and it will accomplish the goal of learning "anytime anywhere". This work sets upon previous work developed in the H2020 BEACONING project and the mobile app presented will allow other developers to create pervasive games for distinct areas of learning in a more engaging and effective way.