Abstract
The application of machine learning (ML) algorithms to electronic health records (EHR) data allows the achievement of data-driven insights on various clinical problems and the development of clinical decision support (CDS) systems to improve patient care. However, data governance and privacy barriers hinder the use of data from multiple sources, especially in the medical field due to the sensitivity of data. Federated learning (FL) is an attractive data privacy-preserving solution in this context by enabling the training of ML models with data from multiple sources without any data sharing, using distributed remotely hosted datasets. The Secur-e-Health project aims at developing a solution in terms of CDS tools encompassing FL predictive models and recommendation systems. This tool may be especially useful in Pediatrics due to the increasing demands on Pediatric services, and the current scarcity of ML applications in this field compared to adult care. Herein we provide a description of the technical solution proposed in this project for three specific pediatric clinical problems: childhood obesity management, pilonidal cyst post-surgical care and retinography imaging analysis.

Abstract
In this paper we explain how to compute gradients of functions of the form G = 1/2 Sigma(m)(i=1) (Ey(i) - C-i )(2), which often appear in the calibration of stochastic models, using automatic adjoint differentiation and parallelization. We expand on the work of Goloubentsev and Lakshtanov and give approaches that are faster and easier to implement. We also provide an implementation of our methods and apply the technique to calibrate European options.

Abstract
The Atacama desert is a region with exceptional conditions for solar power production. However, despite its relevance, the impact of climate change on this resource in this region has barely been studied. Here, we use regional climate models to explore how climate change will affect the photovoltaic solar power resource per square meter (PVres) in Atacama. Models project average reductions in PVres of 1.5% and 1.7% under an RCP8.5 scenario, respectively, for 2021-2040 and 2041-2060. Under RCP2.6 and the same periods, reductions range between 1.2% and 0.5%. Also, we study the contribution to future changes in PVres of the downwelling shortwave radiation, air temperature and wind velocity. We find that the contribution from changes in wind velocity is negligible. Future changes of downwelling shortwave radiation, under the RCP8.5 scenario, cause up to 87% of the decrease of PVres for 2021-2040 and 84% for 2041-2060. Rising temperatures due to climate change are responsible for drops in PVres ranging between 13%–19% under RCP2.6 and 14%–16% under RCP8.5. We conclude that climate change has the potential to impact the PVres in the Atacama region while retaining exceptional conditions for solar power production.

Abstract
Ensuring a sustainable transition process, whether at a global or local level, involves designing an energy mix appropriate to the user's needs. The aim is to identify the optimal future strategy that maximizes both socio-economic benefits and sustainability. To address these challenges, multiple modeling tools are now available to assess different scenarios before implementation. However, modeling tools are generally designed for economic optimization. This is the case with the Open-Source Energy Modeling System (OSeMOSYS) a CLEWs tool. This paper proposes an optimization methodology in terms of sustainability, introducing Life Cycle Assessment (LCA) as a global estimator. In particular, the effects of energy payback times (EPBT) on the selection of transition scenarios will be evaluated. To ensure the reproducibility of the study, we present an exercise that uses data for a fictitious country that shares features of both a developing and a developed country (Atlantis).

Abstract
In this paper the design of a Sustainability Plan for the Salve brewery is modelled. To support transition for decarbonization of the company ensuring a triple zero production (energy, waste, and transport), the critical points during the production process have been identified from a SWOT analysis. Using the CLEWs tools (through the OSeMOSYS software), five scenarios are designed and evaluated according to the company energetic needs. Both historical set of real consumption data of the company, as well as the growth expectations raised by the owners have served as the basis for the work. To simplify the decision making to the company, both investment and total associated costs as well as the corresponding CO2 emissions has been estimated in the period 2023–2030. As main results, an energetic scenario including the use of bagasse waste, the implementation of photovoltaic solar energy and avoiding the fossil fuels, allows to achieve the planned objective with an economic contribution assumable by the company. In addition, several potential business niches associated with the circular economy have been identified.

Abstract
The objective of this work is to evaluate the effects of different energy policies designed to favor decarbonization by increasing renewable sources. In particular, the implementation of energy efficiency policies and the application of hourly differential electricity tariffs. The Open-Source Energy Modelling System (OSeMOSYS) has been adopted to visualize the effects of each of the actions in the short, medium, and long term, from 2024 till 2046. From our results, the application of hourly differentiation tariffs does not favor either the increase in the implementation of renewable sources or decarbonization processes. The implementation of energy efficiency policies (1-1.25% annual demand decrease), in the long term, allows to reach 80% of energy production from renewable sources. In all the scenarios, the energy sources with a greater level of intermittency, such as wind or solar, strongly increased their contribution in the medium-term, thereby stabilizing their long-term contribution. Finally, the implementation of photovoltaic solar energy becomes necessary only in the long-term. It seems clear that this contribution, up to 20% of the renewable, is associated with the nuclear blackout.