Data science applied to epidemiological and demographic information as a strategy for simulation and surveillance of malaria cases in the Brazilian Amazon

A major challenge in the fight against malaria is the sustainability of actions aimed at its control in a context of reduced health spending. Malaria occurs almost entirely in the North, where health systems are less adaptable to change. A situation to be faced is also the reduction in the reduction of transmission containment efforts when there is a reduction in the number of cases, which in a short time leads to the resurgence of the disease. It is clear that the elimination of malaria in Brazil is possible, but that to achieve it, a medium and long term plan with strategies that lead to sustainability is necessary. Among these actions, it highlights the adaptation of vector control measures, with integration and strengthening of the local health system. This requires both technical innovations and political commitment to guarantee human and financial resources. Currently, there are few resources for carrying out surveillance of malaria cases in the Amazon and, generally, they are based on historical data and carried out by highly specialized human resources, making the work costly and often unviable because it is not possible to infer knowledge in a timely manner. to carry out preventive or even corrective actions. In addition, the actions that could be taken by public managers are limited by the lack of tools for exploring this data, especially those that are based on information visualization methods, which allow analysis in a dynamic, interactive and even in real time (if the data is integrated). The vast majority of actions for the problem in question are reactive, carried out after verified by observation of real cases. It is in this scenario that the present research project is inserted. It is important to note that several previous studies carried out in the country have explored the issue of malaria transmission in the Brazilian Amazon, both from an epidemiological, as well as biological, vector and geographical perspective. It is necessary to develop new approaches and strategies that propose more efficient identification and control. It is extremely important to identify and analyze the factors that have the greatest impact on malaria cases in the Amazon Region. It is believed that with the application of machine learning methods using demographic, epidemiological, climatic and clinical data, a prediction of the behavior of malaria cases in the region can be carried out and, consequently, favor the implementation of more effective preventive actions and better management. 

Epidemiology
Health Systems

Feb 2020 — Feb 2023

Brazil