Abstract
Artificial Intelligence-based tools have shown promising results to help clinicians in diagnosis tasks. Radio-genomics would aid in the genotype characterization using information from radiologic images. The prediction of the mutations status of main oncogenes associated with lung cancer will help the clinicians to have a more accurate diagnosis and a personalized treatment plan, decreasing the need to use the biopsy. In this work, novel and objective features were extracted from the lung that contained the nodule, and several machine learning methods were combined with feature selection techniques to select the best approach to predict the EGFR mutation status in lung cancer CT images. An AUC of 0.756 ± 0.055 was obtained using a logistic regression and independent component analysis as feature selector, supporting the hypothesis that CT images can capture pathophysiological information with great value for clinical assessment and personalized medicine of lung cancer. Clinical Relevance-Radiogenomic approaches could be an interesting help for lung cancer characterization. This work represents a preliminary study for the development of computer-aided decision systems to provide a more accurate and fast characterization of lung cancer which is fundamental for an adequate treatment plan for lung cancer patients.

Abstract
Wastewater treatment plants constitute an essential part of the sewage system. They have the role of removing pollutants from wastewater to enable the safe disposal of the treated effluent in the natural environment. This research seeks to evaluate plants' efficiency and effectiveness, which involves minimizing energy consumption while obtaining a quality level of the treated water aligned with legislation requirements. We explore two policy scenarios regarding the measurement of effluent quality. The first assumes that pollutants' emission quotas (EQs) are fixed at each plant. The second assumes that quotas are set for the receiving waters (e.g., river or watercourse in the natural environment) so that trade-offs in EQs among plants sharing the same discharge site are possible. This latter scenario requires a system-wide analysis to identify optimal targets for pollutants removal at each plant that allow fulfilling the expected average quality levels of the effluent discharged. This paper develops a methodology to fully realize the potential for energy savings based on an innovative mixed-integer linear programming model. This model follows the data envelopment analysis axioms to estimate the frontier of the production possibility set. The approach proposed is tested in a real-world context using the plants of a Portuguese water company. The results show that the two scenarios combining efficiency and effectiveness perspectives have advantages in terms of energy savings compared to the conventional situation focused only on efficiency gains. The saving potential is slightly higher in the scenario allowing reallocation of EQs among plants.

Abstract
[No abstract available]

Abstract

Abstract
We study an evolutionary dynamics for the contributions by agents to a common/public good in a generalized version of Baliga and Maskin's environmental protection model. The dynamical equilibria consist of three scenarios: a single agent contributing to preserve the good with its optimal contribution level, and all the other agents being free-riders: a group of agents with the same optimal contribution level contributing to preserve the good, and all the other agents being free-riders; one where no agents contribute. The dynamics of the contributions can be complex but we prove that each trajectory converges to the equilibrium associated to the single agent (or group of agents) with the highest preference for the good that are contributing since the beginning. We note that while the aggregate contribution is below the optimal contribution level of the agent with the smallest preference for the good, then the aggregate contribution is increasing and there is no free-riding. Hence, if the optimal contribution level of the agent with the smallest preference is enough to not exhaust the good too quickly and the optimal contribution level of the agent with the greatest preference is enough to preserve the good, then, in spite of the appearance of free-riding in the contributions, the good might not be exhausted.

Abstract
Resumo   Ao considerar diversos desafios que surgiram e que contribuíram para (re) significar o ensino e a aprendizagem no campo dos estudos com surdos, no âmbito das tecnologias digitais em rede, muitas vezes pensadas para a produção e o consumo de ouvintes tendo, portanto, características baseadas nas experiências de quem percebe o mundo pela audição, surgiu a necessidade de experimentar e refletir sobre as possibilidades de engajar os surdos em interações que promovam o desenvolvimento das aprendizagens em redes digitais. Diante de tal desafio e a aproximação com Seminário Práticas de Pesquisa, desenvolvida no âmbito da Pós-graduação em Educação da UNISINOS (PPGEDU) e Pós-Graduação em Linguística Aplicada (PPGLA), tivemos a oportunidade de ter contato com ações de co-criação de uma rede internacional que objetiva conectar crianças e adolescentes em diferentes lugares do mundo para desenvolver ações reticulares de Educação OnLIFE. Como resultado desta aproximação, vimos a necessidade de problematizar práticas pedagógicas para surdos, por meio da Libras para oportunizar aos ouvintes reflexões sobre o papel da Libras nas interações e potencializar o desenvolvimento de ações conjuntas e de transformações sociais numa perspectiva que considere, tanto a cultura surda quanto a cultura ouvinte em rede.