The EcoSustain project focuses on Computing and Data Science for Sustainability and brings together an interdisciplinary group of researchers with extensive experience in data science and engineering, artificial intelligence, software engineering, simulation, networks, and IoT. Additionally, the team includes experts in environmental sciences, climate change, smart irrigation, urban mobility, health, and economics. Its goal is to develop technological solutions aimed at creating public databases, software systems, machine learning models, dashboards, etc., to monitor and analyze urban and rural environments, contributing to ensuring the effective prevention, prediction, and reduction of environmental degradation processes and their impacts on health, as well as mathematical and econometric models that will serve as a basis for recommendations for evidence-based public policies.

Research Themes

ForestEyes - Citizen Scientists and Machine Learning Assisting Rainforest Conservation

Launched in April 2019, the ForestEyes Project aims to provide supplementary data to assist specialists from governmental and non-profit organizations in their deforestation monitoring efforts. This project uniquely integrates Citizen Science (CS) with Machine Learning (ML) to monitor tropical forests. The renowned Zooniverse platform hosts the CS module, in which volunteers/ordinary citizens, classify specific portions (segments) of satellite images into two predefined categories (forest or non-forest areas). Next, these manual contributions are analyzed and used as training data for machine learning techniques. Finally, different techniques are employed to automatically identify deforestation segments in new images of an extensive forest region. This methodology based on CS and ML techniques introduces a novel computational system compared to the existing monitoring systems for the Amazon Forest. In the future, this proposed system may generate warning signals for the relevant authorities and produce supplementary data for official monitoring programs.

Coordinator: Alvaro Luiz Fazenda
Institution of affiliation: Unifesp
Other researchers: Fabio Augusto Faria, Fabio A.M. Cappabianco - (Unifesp), Fernanda B.J. Dallaqua (VISIONA), Roberto Speicys Cardoso (Scipopulis)
Students: Hugo Rezende, Juan Carlos Guerra Blas, Eduardo Bouhid Neto (Unifesp)
External collaborators: Silvio Jamil Ferzoli Guimarães (PUC-MG)

Smart City Applications for Sustainable Urban Mobility

The increasing use of mobile networks and Internet of Things devices enables communications involving sensors, pedestrians, vehicles, and the wired Internet. Smart Cities and Intelligent Transportation Systems applications leverage the data collection capabilities and the mobility of such devices to improve the citizens' quality of life. Those applications need to address wireless networking challenges, e.g., scalability and latency requirements. The goal of this project is to propose applications that reduce the environmental impacts of urban activities using data collected from users without leaking private information. The first application implements a use case tackling intelligent traffic light control with centralized machine learning. The second application looks at using a more general strategy based on distributed systems. The idea is to implement the traffic light control application using federated learning and, in addition, build a framework that can be reused by other applications from EcoSustain.

Coordinator: Miguel Elias Mitre Campista
Institution of affiliation: UFRJ
Other researchers: Luís Henrique Maciel Kosmalski Costa (UFRJ), Antônio Jorge Gomes Abelém (UFPA)
Students: Lucas de Carvalho Gomes (UFRJ), Guilherme Araujo Thomaz (UFRJ)
External collaborators: Marcelo Dias de Amorim (Sorbonne University), França

Towards a Digital Twin Approach for Health Care Based on Environmental and Digital Phenotype Monitoring

Digital phenotyping can be used to monitor the health of individuals through the collection of physiological data, such as heart rate, respiratory rate, oxygen saturation, pressure, body inclination, combined with location and physical activity data, from which behavioral characteristics can be extracted that allow health problems of interest to be inferred in different areas of health, such as occupational and mental health. Additionally, environmental aspects such as temperature, humidity, light, noise level, air pollution, among others, can be combined with digital phenotyping data to investigate how the environment may be influencing the health of individuals. This ECOSustain sub- project proposal is related to social sustainability and aims to investigate a digital twin approach for combining environmental monitoring with digital phenotyping to enable research in Personalized Healthcare Services (PHS).

Coordinator: Francisco José da Silva e Silva
Institution of affiliation: UFMA
Other researchers: Davi Viana (UFMA) , Luciano Reis Coutinho (UFMA) , Markus Endler(PUC-Rio) , Anderson de Oliveira (PUC-Rio) , Vitor Pinheiro de Almeida (PUC-Rio) , Fabio Kon (IME-USP) , Daniel Cordeiro (EACH-USP), Fábio Costa (UFG)
Students: Luis Eduardo Costa Laurindo (doutorado-UFMA) André Filipe Sousa Barreto (mestrado-UFMA) Amanda Almeida Cardoso (graduação-UFMA) Augusto Gonçalves Santos (graduação-UFMA) Marcos Vinicius dos Santos Oliveira (graduação-UFMA) Francisco Wallison Carlos Rocha (doutorado-EACH-USP) Jedean Simões (Graduação, PUC-Rio) Victor Cortez (mestrando, PUC-Rio)
External collaborators: Cícero Costa Quarto (UEMA, Brazil)

Members