Optimal Artificial Neural Network for the Diagnosis of Chagas Disease Using Approximate Entropy and Data Augmentation

• Published in: Computing in cardiology Vol. 50; pp. 1 - 4
• Main Authors: Rodriguez, Maria Fernanda, Cornejo, Diego Rodrigo, Diaz, Luz Alexandra, Ravelo-Garcia, Antonio, Alvarez, Esteban, Cabrera-Caso, Victor, Condori-Merma, Dante, Cornejo, Miguel Vizcardo
• Format: Conference Proceeding
• Language: English
• Published: CinC 01.10.2023
• Subjects: Data augmentation; Distributed databases; Heart; Machine learning; Time series analysis; Training; Urban areas
• ISSN: 2325-887X
• DOI: 10.22489/CinC.2023.143

The use of machine learning for disease diagnosis is gaining popularity due to its ability to process data and provide accurate results, but optimazing it remains a challenge. Chagas disease is endemic in Latin America and has emerged as a health problem in more urban areas. Early and accurate diagnosis is essential to prevent cardiac complications, since an estimated 65 million people are at risk of contracting this disease. This study used a database of 292 subjects distributed into three groups: healthy volunteers (Control group), asymptomatic Chagasic patients (CHI group) and seropositive Chagasic patients with incipient heart disease (CH2 group). A densely connected neural network was used to classify them into their respective groups. The network received as input the Approximate Entropy values of each individual, which were calculated from the 24-hour circadian profiles every 5 minutes (288 RR subsegments). Time series data augmentation algorithms were applied during the training phase to improve the classification results. This approach allowed to achieve 100% accuracy and precision, validated by the ROC curve with AUC values of 1, proving to be a robust approach for early diagnosis and prevention of heart complications in Chagas disease.