Beat-to-Beat Variability in Computational hv-CM and hiPSC-CM Models

Beat-to-beat variability (B2BV) in the electrical activation of human-induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) has been previously demonstrated in in vitro studies.Here, we study how B2BV is recapitulated in computational CM models.We studied the dynamic properties of two human ventricular (hv) CM models, Bartolucci et al. (BPS2020) and Margara et al. (MBR2021), and one hiPSC-CM model from Paci et al. (PHS2020).The interbeat intervals (IBIs) were used as inputs and the action potential duration at 90 % repolarization (APD90) values as outputs in the models.The cross-correlation between the IBI and APD90 time series was calculated with different lags.In the PHS2020 model, the simulation was extended with randomly modulatable J rel and I f parameters to create B2BV in the model.The most realistic variability was created by modulating I f the parameter, with pink noise applied to mimic the biological variability and memory effects.The results show that it is possible to add B2BV to the models.As heart rate variability (HRV) serves as an important clinical marker, B2BV could offer a new biomarker for computational CM models.