Modeling the Electrical Activity of the Heart via Transfer Functions and Genetic Algorithms

• Published in: Biomimetics (Basel, Switzerland) Switzerland), 2024-05, Vol.9 (5), p.300
• Main Authors: Rodríguez-Abreo, Omar, Cruz-Fernandez, Mayra, Fuentes-Silva, Carlos, Quiroz-Juárez, Mario A, Aragón, José L
• Format: Article
• Language: English
• Subjects: Algorithms; Artificial intelligence; Cardiac arrhythmia; Cardiac function; Cardiovascular research; Diagnosis; ECG; EKG; Electrocardiogram; Electrocardiography; Electrodes; Electrophysiology; genetic algorithm; Genetic algorithms; Heart; Heart diseases; Mathematical models; metaheuristic optimization; modeling; Ordinary differential equations; Oscillators; Pacemakers; transfer function; Variables
• ISSN: 2313-7673; 2313-7673
• DOI: 10.3390/biomimetics9050300

Although healthcare and medical technology have advanced significantly over the past few decades, heart disease continues to be a major cause of mortality globally. Electrocardiography (ECG) is one of the most widely used tools for the detection of heart diseases. This study presents a mathematical model based on transfer functions that allows for the exploration and optimization of heart dynamics in Laplace space using a genetic algorithm (GA). The transfer function parameters were fine-tuned using the GA, with clinical ECG records serving as reference signals. The proposed model, which is based on polynomials and delays, approximates a real ECG with a root-mean-square error of 4.7% and an R2 value of 0.72. The model achieves the periodic nature of an ECG signal by using a single periodic impulse input. Its simplicity makes it possible to adjust waveform parameters with a predetermined understanding of their effects, which can be used to generate both arrhythmic patterns and healthy signals. This is a notable advantage over other models that are burdened by a large number of differential equations and many parameters.