Use of Fuzzy Entropy for Risk Stratification in Patients with Chagas Disease

Santiago Ismael Flores-Chavez1, Antonio Gabriel Ravelo-Garcı́a2, Miguel Vizcardo1
1Universidad Nacional de San Agustin de Arequipa, 2Institute for Technological Development and Innovation in Communications, Universidad de Las Palmas de Gran Canaria, 35017 Las Palmas de Gran Canaria, Spain Interactive Technologies Institute (ITI/LARSyS and ARDITI), 9020-105 Funchal, Portugal


Abstract

Abstract: According to the World Health Organization, the number of people infected with Trypanosoma cruzi is estimated between 6 and 7 million, this being the causative agent of Chagas disease, with 550,000 people exposed to the risk of affectation. This study used fuzzy entropy to quantify the regularity of tachograms in patients with Chagas disease. The study population consisted of three groups of volunteers: 92 controls (C), 102 patients with positive serology without cardiac involvement diagnosed by conventional non-invasive methods (CH1), and 107 patients with positive serology and mild to moderate incipient heart failure (CH2). RR segments of 5 minutes were analyzed, 288 segments, corresponding to 24 hours per patient. Methodology: Tachograms were obtained by processing the electrocardiogram (ECG) and obtaining the RR interval. For QRS complex detection, the Pan-Tomkins algorithm was used, generating 288 tachograms of 5 minutes. Fuzzy entropy was employed to evaluate the irregularity and complexity of the time series, calculating it for the 288 tachograms of 5 minutes and obtaining 24-hour circadian profiles. The Kruskal-Wallis test was used to compare the 288 segments corresponding to the Control-CH1, Control-CH2, and CH1-CH2 groups to evaluate differences between groups. Logistic regression was also applied to evaluate the circadian profiles of the mean values of fuzzy entropy. Results: Fuzzy entropy demonstrated potential utility for assessing irregularity and complexity patterns in the circadian profiles across all three study groups. The analysis suggested that during afternoon periods, Control group measurements exhibited different characteristics compared to those from the CH1 and CH2 groups, possibly indicating alterations in autonomic response. These preliminary findings suggest this approach could potentially serve as an indicator of heart rate variability patterns and might contribute to risk assessment methodologies for patients with Chagas disease, with particular implications for the CH1 group.