Artificial Neural Network and Permutation Entropy in the Stratification of Patients with Chagas Disease

Luz Diaz1, Maria Rodriguez1, Diego Cornejo1, Antonio Ravelo-Garcia2, Esteban Alvarez3, Miguel Vizcardo1
1Universidad Nacional de San Agustin de Arequipa, 2Universidad de Las Palmas de Gran Canaria, 3Universidad Central de Venezuela, Venezuela


Chagas disease is a potentially life threatening illness that in the last decades was becoming a public health problem because it became mainly an urban disease, thus putting 70 million people at risk of infection. It may be silent and asymptomatic in the chronic phase (where 40% of the population has cardiac compromise). That is the reason of why it is necessary the development of early markers. To achieve this, we propose a MLP architecture in order to classify 292 patients into three different groups: The Control group with 83 volunteers, the CH1 group with 102 patients with positive serology and no cardiac involvement and the CH2 group with 107 patients with positive serology and mild to moderate incipient heart failure. The used data comes from 24-hour ECG, the RR intervals from each subject were divided in 288 frames of 5 minutes each.Then the RR data was preprocessed using permutation en- tropy obtaining a circadian profile for each patient which was used in the training of the proposed architecture. The classification performed with 89 % accuracy and 89% precision, consisting in a great work of classification vali- dated by the AUC in each ROC curve. All these results were obtained with limited data, so this study can be im- proved with more available data, making this model a tool for analyzing ECG in order to try to do an early evalua- tion and diagnosis of a cardiac compromise related to the generally silent chronic phase.