EPIDEMIOLOGICAL ANALYSIS OF HANSENIASIS USING DATA MINING TECHNIQUES WITH ARTIFICIAL NEURAL NETWORKS IN A HYPERENDEMIC MUNICIPALITY IN THE STATE OF PARÁ
Data Mining, Clustering, Kohonen Networks, Epidemiology
Leprosy is a chronic granulomatous infectious disease caused by the obligate intracellular organism that primarily affects the skin and peripheral nerves that can lead to severe physical impairments and deformities if not diagnosed and treated in its early stages. The transmission and dynamic causes of leprosy are complex. The relationship between poverty and leprosy is a very close relationship, as most places with high socioeconomic vulnerability indicators have very favorable characteristics for the transmission of the bacillus. In more vulnerable areas, problems such as lack of knowledge about the disease mean that it is not treated with its due priority and undiagnosed people are passing the disease on to their contacts. In addition, the houses have many similarities when it comes to characteristics such as: population density of the house, water quality, family income, education and place of disposal of faeces and urine. These indicators can lead people to have a low immunological resistance, together with the favorable location for the transmission of the bacillus can make that region hyperendemic. This study seeks to use computational intelligence techniques, based on clustering, to analyze the epidemiology of leprosy through the relationship of patients and their contacts in the city of Santarém, Pará.