@inbook{32565, keywords = {clustering, Data mining, Epidemiological analysis, Epidemiology, Kohonen networks, leprosy}, author = {Silva YED and Salgado CG and Gomes Conde VM and Conde GBA}, title = {Application of Clustering Technique with Kohonen Self-organizing Maps for the Epidemiological Analysis of Leprosy}, abstract = {
Leprosy is still a worldwide public health problem, and Brazil ranks second in the largest number of cases of leprosy. Social and economic conditions drastically influence people’s lives, making them more vulnerable to disease and increasing dissemination, contributing to endemicity in the country. Based on this, this study aimed to analyze the epidemiology of leprosy by analyzing data from patients and their Household contacts using Artificial Intelligence techniques in the data mining process. The best results were obtained with Kohonen’s Self-Organizing Maps algorithm in 2 × 3 matrix. A data set with SINAN patients and new leprosy cases (schoolchildren and HHCs) was found in an active search conducted in the municipality of Santarém in the year 2014. The results analyzed call attention to a high number of late diagnoses and the values found for the Anti PGL-1 in clusters 1, 2 and 5 indicating a high burden of the leprosy bacillus and, therefore, a high risk of contagion. The study demonstrated that the identification of the relationship profile of the leprosy patient with their home and their family appears as promising tools, such as leprosy control strategy.
}, year = {2018}, number = {869}, pages = {295-309}, publisher = {Springer International Publishing}, address = {Cham}, issn = {2194-5357}, isbn = {978-3-030-01056-0}, doi = {10.1007/978-3-030-01057-7_24}, language = {eng}, }