The team did an analysis of all the challenges and discussed which ones most fit the team's profile. Challenge 9 involved data collection, and as researchers, we feel more familiar and motivated to find some innovative way from the information that would be collected. As an approach, a brainstorming methodology was used to decide our initial problem, followed by Crazy-8 to find several solutions and group them in topics and the analysis of literature in Nvivo software, discussing with the group which ones would be the most relevant for a later one data survey. After defining the topics and completing the search, it was decided in a team that the best way for a more assertive forecast would be using artificial intelligence in our favor, in order to identify the patterns between the cities most affected by the covid-19 and its correlation with data collected in the literature. Before defining the most important topic among the solutions, we would use the satellite images of the land-based infrastructure, but as the project progressed, the team decided to use the Brazilian model as an analysis and from there the UN data, Datasus and Secretariat of Health. The project included the development of artificial intelligence using python and visual studio code. The first problem found was that the team was not sure of the direct relationship between the indexes that were being raised in the literature with confirmed cases of covid-19. As much as at the end of the analysis we have come to the conclusion that these indices interfere in the forecasts and that they can be a tool for identifying the next cities that would be affected, the direct correlation has not yet been confirmed.

Another problem encountered was the very long deliveries requested, which compromised the prototyping time.

Human development indicators were used to predict the most contagious points of pathology, a problem that affects the environment as a whole. Then, artificial intelligence was related to these indicators, allowing the creation of a model that foresees the critical areas of dissemination, through an AI that seeks patterns in the data that are difficult to perceive and performs the forecast of new epicenters, enabling actions anticipated, eradicating the pandemic.

Data from the United Nations Development Program (UNDP), an organ of the United Nations (UN) were used, where human development indexes were obtained, Datasus was found, the Gini indexes of municipalities were found, and health secretariats linked to the Ministry of Health from Brazil, collected the crude population and confirmed cases, collected the crude population and confirmed cases.