The objective of this thesis is to examine the possibility to determine the mechanical heart rate using intracardiac impedance in the time domain. Deducing the mechanical heartrate from the impedance could help improve the performance of implanted devices that today depend on the measurement of the heart’s electrical activity. Cardiogenic – also known as intracardiac – impedance is based on the difference in conductivity between heart muscle tissue and blood, making the impedance vary as the heart is filled and emptied. The data used in this thesis was acquired from three previous studies performed by St Jude Medical, two clinical and one preclinical. Two impedance measurement configurations were chosen from these studies, one bipolar and one quadropolar. To deduce the heart rate from the intracardiac impedance six algorithms were evaluated. Three using continuous peak detection and three evaluating small frames of the impedance signal.The peak detection algorithms were peak detection on the impedance signal itself, on its derivative  and on its integral. The three others were an Auto Correlation Function (ACF), an Average Magnutide Difference Function (AMDF) and an Average Wave Comparison Function (AWCF). In order to assess the heart rates deduced from the intracardiac impedance by the algorithms, these rates were compared to both the IEGM or the ECG (depending on which study was at hand) and the blood pressure. Several issues affected the performance of the algorithms. Impedance morphology can vary between patients. Some display so called “double peaks”, making it hard to decide whether a patient has for example a pulse of 80 bpm or of 160 bpm. The impedance morphology was also affected by amplitude modulation with the respiration frequency which in some patients cause difficulties to analyze the impedance signal. The results show that the two impedance measurement configurations perform equally well and that the ACF method was the overall best performing algorithm. They also show that individual patient impedance morphology has a large influence on the results and for future studies it should therefore be interesting to calibrate the algorithms for each patient, as this should improve performance.