0 avis
Modeling Variability in Cardiac Electrophysiology: A Moment Matching Approach
Archive ouverte
Edité par CCSD ; the Royal Society -
International audience. The variability observed in action potential (AP) cardiomyocyte measurements is the consequence of many different sources of randomness. Often ignored, this variability may be studied to gain insight into the cell ionic properties. In this paper, we focus on the study of ionic channel conductances and we describe a methodology to estimate their probability density function (PDF) from action potential recordings. The method relies on the matching of observable statistical moments and on the maximum entropy principle. We present four case studies using synthetic and experimental AP measurements sets from human and canine cardiomyocytes. In each case, the proposed methodology is applied to infer the PDF of key conductances from the exhibited variability. The estimated PDFs are discussed and, when possible, compared to the true distributions. We conclude that it is possible to extract relevant information from the variability in AP measurements and discuss the limitations and possible implications of the proposed approach.