Abstract
According to the World Health Organization forecast, over 55 million people worldwide have dementia, and about 10 million new cases are detected yearly. Early diagnosis is essential for patients to plan for the future and deal with the disease. Machine Learning algorithms allow us to solve the problems associated with early disease detection. This work attempts to identify the current relevance of the application of machine learning in dementia prediction in the scientific world and suggests open fields for future research. The literature review was conducted by combining bibliometric and content analysis of articles originating in a period of 20 years in the Scopus database. Twenty-seven thousand five hundred twenty papers were identified firstly, of which a limited number focused on machine learning in dementia diagnosis. After the exclusion process, 202 were selected, and 25 were chosen for analysis. The recent increasing interest in the past five years in the theme of machine learning in dementia shows that it is a relevant field for research with still open questions. The methods used to identify dementia or what features are used to identify or predict this disease are explored in this study. The literature review revealed that most studies used magnetic resonance imaging (MRI) and its types as the main feature, accompanied by demographic data such as age, gender, and the mini-mental state examination score (MMSE). Data are usually acquired from the Alzheimer's Disease Neuroimaging Initiative (ADNI). Classification of Alzheimer's disease is more prevalent than prediction of Mild Cognitive Impairment (MCI) or their combination. The authors preferred machine learning algorithms such as SVM, Ensemble methods, and CNN because of their excellent performance and results in previous studies. However, most use not one machine-learning technique but a combination of techniques. Despite achieving good results in the studies considered, there are new concepts for future investigation declared by the authors and suggestions for improvements by employing promising methods with potentially significant results.

Abstract

Abstract
The use of several distributed generators as well as the energy storage system in a local microgrid require an energy management system to maximize system efficiency, by managing generation and loads. The main purpose of this work is to find the optimal set-points of distributed generators and storage devices of a microgrid, minimizing simultaneously the energy costs and the greenhouse gas emissions. A multi-objective approach called Pareto-search Algorithm based on direct multi-search is proposed to ensure optimal management of the microgrid. According to the non-dominated resulting points, several scenarios are proposed and compared. The effectiveness of the algorithm is validated, giving a compromised choice between two criteria: energy cost and GHG emissions.

Abstract
Proper waste management has been recognized as a tool for the green transition towards a more sustainable economy. For instance, most studies dealing with municipal solid wastes in the literature focus on environmental aspects, proposing new routes for recycling, composting and landfilling. However, there are other aspects to be improved in the systems that deal with municipal solid waste, especially in the transportation sector. Scholars have been exploring alternatives to improve the performance in waste collection tasks since the late 50s, for example, considering the waste collection problem as static. The transition from a static approach to a dynamic is necessary to increase the feasibility of the solution, requiring faster algorithms. Here we explore the improvement in the performance of the guided local search metaheuristic available in OR-Tools upon different execution times lower than 10 seconds to solve the capacitated waste collection problem. We show that increasing the execution time from 1 to 10 seconds can overcome savings of up to 1.5 km in the proposed system. Considering application in dynamic scenarios, the 9 s increase in execution time (from 1 to 10 s) would not hinder the algorithm's feasibility. Additionally, the assessment of the relation between performance in different execution times with the dataset's tightness revealed a correlation to be explored in more detail in future studies. The work done here is the first step towards a shift of paradigm from static scenarios in waste collection to dynamic route planning, with the execution time established according to the conclusions achieved in this study. © 2023 ITMA.

Abstract
Low Power Wide Area Networks (LP-WAN) are receiving a lot of attention because of their ability to communicate using radio frequency in long distances, with low-power consumption and low-cost devices. In this paper, we provide a comparison between the two LP-WAN platforms that are leading the market, the Sigfox and the LoRaWAN, based on the literature. Both platforms are analyzed considering the context of the forest fire detection and verification systems. Many aspects are being considered to identify which LP-WAN is more adequate to be used in this kind of systems, such as battery lifetime, coverage range, business model and costs. The comparison shows that both platforms are very similar in most of the aspects, although LoRaWAN is more flexible than Sigfox on the deployment and management of the network infrastructure. LoRaWAN allows customers to implement and manage their own infrastructure network, which is essential in systems which monitor vast forest areas.

Abstract
Unmanned Aerial Vehicles (UAVs) have emerged as valuable tools in precision agriculture due to their ability to provide timely and detailed information over large agricultural areas. In this sense, this work aims to evaluate the semi-autonomous navigation capacity of a multirotor UAV when applied in the field of precision agriculture. For this, a small aircraft is used to identify and track a set of fiducial markers (Ar Track Alvar) in an environment that simulates inspections of insect traps in olive groves. The purpose of this marker is to provide a visual reference point for the drone's navigation system. Once the Ar Track Alvar marker is detected, the robot will receive navigation information based on the marker's position to approach the specific trap. The experimental setup evaluated the computer vision algorithm applied to the UAV to make it recognize the Ar Track Alvar marker and then reach the trap efficiently. Experimental tests were conducted in a indoor and outdoor environment using DJI Tello. The results demonstrated the feasibility of applying these fiducial markers as a solution for the UAV's navigation in this proposed scenario.