IEEE Transactions on Biomedical Engineering · 1989 · 35 citations · 8 references
MedicineCardiologyPhysiologyBiomedical Signal ProcessingElectrophysiologyPhysiological Systems ModelingCardiac Action PotentialsCardiac Conduction SystemCardiac PhysiologyElectrophysiological ModelingVentricular ElectrophysiologyCardiac Chronotropic FunctionCardiac Electrophysiological ModelingCardiac Electrophysiology ModelingElectrophysiologic EffectsVentricular Excitability
The objective of this study was to determine the ability of the empirical Lapicque and theoretically derived Blair expressions for excitation to fit experimentally obtained threshold current values to evoke a ventricular extrasystole using rectangular-wave stimuli applied to the dog heart. The data points were fitted to both expressions and the ability of each to predict the measured values was determined. The Levenberg-Marquardt (L-M) algorithm was used to fit the Lapicque and Blair expressions. The Lapicque data were also fitted to the linear charge-duration expression of Weiss (W). It was found that the ratio of the predicted to measured current was slightly different from one 0.95 (L-M) and 1.06 (W) for the Lapicque and 0.92 (L-M) for the Blair expression. Thus, there appears to be little difference between the ability of the expressions to fit the same experimentally obtained data. The L-M Lapicque fit is best for the short durations range; the Weiss-Lapicque fit overestimates in the short duration range and underestimates near chronaxie. The L-M Blair fit is best for the short duration range and poor for the durations near the membrane time constant.
8
Nonlinear Regression Modeling.
Jeffrey A. Witmer, David A. Ratkowsky · Journal of the American Statistical Association · 1984 · 502 citations