Our web app is a decision support platform for smart health resource management, identifying risk areas and disease pattern according to environmental characteristics such as rainfall, climate, deforestation, fluviometry, air pollution and others.

As such, we focus on obtaining information before disasters occur. For this, we performed:

1. Creation of a database with health-related variables, such as year, month, diagnosis, gender, race, education, type of closure (discharge, death), place of occurrence and many others.

2. Creating a database with environmental variables such as burning, rainfall, deforestation, flooding and others using NASA data.

3. We gathered the databases according to year, month and place of occurrence, using the Unified Health System database (SUS, Brazil). (Importantly, many countries have health information systems, and this makes our system widely scalable !!).

4. We use machine learning techniques to predict locations that are vulnerable to certain diseases.

Finally, this data may be used for consultation by anyone, but our target audience is mainly local health managers and researchers.

BIBLIOGRAPHIC REFERENCES

1. American tegumentary leishmaniasis: a multivariateanalysis of the spatial circuits for production ofcases in Minas Gerais State, Brazil, 2007 to 2011;

2. Deforestation and malaria incidence in thelegal Amazon Region between 1996 and 2012;

3. Impacts of deforestations, agriculture, economic growth and municipal expenciture on the incidence of malaria in amazon: an analisis space;

4.Impact of deforestation on incidence of diseases in Amazon.

Pilot project:

https://github.com/arthurgsf/GAIDA