by Keyword: Recursive identification
Calvo, M., Le Rolle, V., Romero, D., Béhar, N., Gomis, P., Mabo, P., Hernández, A. I., (2019). Recursive model identification for the analysis of the autonomic response to exercise testing in Brugada syndrome Artificial Intelligence in Medicine 97, 98-104
This paper proposes the integration and analysis of a closed-loop model of the baroreflex and cardiovascular systems, focused on a time-varying estimation of the autonomic modulation of heart rate in Brugada syndrome (BS), during exercise and subsequent recovery. Patient-specific models of 44 BS patients at different levels of risk (symptomatic and asymptomatic) were identified through a recursive evolutionary algorithm. After parameter identification, a close match between experimental and simulated signals (mean error = 0.81%) was observed. The model-based estimation of vagal and sympathetic contributions were consistent with physiological knowledge, enabling to observe the expected autonomic changes induced by exercise testing. In particular, symptomatic patients presented a significantly higher parasympathetic activity during exercise, and an autonomic imbalance was observed in these patients at peak effort and during post-exercise recovery. A higher vagal modulation during exercise, as well as an increasing parasympathetic activity at peak effort and a decreasing vagal contribution during post-exercise recovery could be related with symptoms and, thus, with a worse prognosis in BS. This work proposes the first evaluation of the sympathetic and parasympathetic responses to exercise testing in patients suffering from BS, through the recursive identification of computational models; highlighting important trends of clinical relevance that provide new insights into the underlying autonomic mechanisms regulating the cardiovascular system in BS. The joint analysis of the extracted autonomic parameters and classic electrophysiological markers could improve BS risk stratification.
JTD Keywords: Autonomic nervous system, Brugada syndrome, Computational model, Recursive identification