Monitoring and prediction of the AIDS epidemic in Chile using neural networks

When a person is interested in looking for information about AIDS, they usually go to Internet search engines, such as Google, using appropriate search terms, such as: “HIV test”, “Tritherapy”, “AIDS diagnosis and treatment”, etc. . Considering that the search interest may be related to some suspicion or certainty about suffering from the disease, we believe that it is possible to monitor and predict AIDS incidences and deaths in Chile using data patterns from Google search trends. Based on this hypothesis, the objective of our project is to develop artificial neural networks that, based on Google search trend data, can monitor and predict AIDS incidences and deaths in the country.

More broadly, through health policies based on adequate epidemiological information, attempts are made to reduce the lethality and spread of the HIV epidemic. The information, coming from health centers throughout the country, must be updated and readily available for use in predictive statistical models. However, the delay or lack of all the necessary information can affect the performance of these techniques. Considering that on the Internet there is a large amount of data in real time, in the condition of big data, associated with the search behavior of users on diagnostic topics, symptoms and treatment of diseases, this information can be used in conjunction with methods of artificial neural networks to establish artificial intelligence systems that monitor epidemics of different types, such as AIDS. This project proposes the implementation of those techniques, adapting them to the reality and needs of our country in the improvement of epidemiological control systems of incidence and deaths from HIV/AIDS.