Computer model of cardiac activation and recovery times for evaluating indirect electrocardiographic measures of cardiac properties

The purpose of this thesis was to develop a cardiac model which would allow the investigation of characteristics of the electrocardiogram for assessing the local cardiac properties of activation sequence, recovery sequence and changes of resting potential in the underlying tissue. The cardiac model stimulated a strip of cardiac tissue and was assumed to be situated in a homogenous infinite volume conductor. A distribution of action potentials throughout the cardiac strip was used as the simulation source for electrocardiogram generation. One method investigated to assess activation and recovery properties was through the use of mapping the QRS, STT and QRST areas along the cardiac strip. The results of this thesis revealed that, in the absence of resting potential gradients throughout the cardiac strip, QRS area maps are indicative of the activation sequence, STT area maps are indicative of the recovery sequence and QRST area maps are indicative of the distribution of recovery durations. However, when a resting potential gradient is introduced the results showed that STT areas and QRST areas were highly sensitive to the resting potential gradients while the QRS areas were only mildly affected by the gradients. Thus, on the basis of STT and QRST area maps alone, it is not possible to discern a particular set of recovery properties in the presence of a nonuniform distribution of resting potentials. A second method investigated as a potential index in assessing activation and recovery properties was through differentiating the electrocardiogram with respect to time. In this approach the map of the times of the the masimum QRS downstroke was shown to be a useful index for assessing activation sequence. This result was true regardless of any resting potential gradients. To assess recovery sequences, maps of the times of maximum upstroke and maximum downstroke in the t-wave were examined. The results of this model showed that in both of these approaches the relative order of recovery could be easily recognized. The difference between these two approaches was that the time of maximum down stroke the T wave more accurately indicated the actual time of recovery for given localized areas of the cardiac strip than the time of maximum upstroke. Overall, neither the time of maximum downstroke nor the time of maximum upstroke in the T wave were affected by resting potential gradients. The final method investigated was to independently assess resting potential gradients. The conclusion reached in this portion of the study was that any DC shift of the TQ segment is directly resultant from a gradient of resting potentials and therefore a map of the DC shift of the TQ segment would be index in assessing resting potential disparities.