Detection of systole and diastole start in cardiac output and arterial pressure recordings

In coronary artery bypass graft patency research, the quality of the graft is often quantified by the mean graft flow, pulsatility index (PI) and diastolic filling (DF). Calculation of PI and DF requires segmentation of the heart cycle into systolic and diastolic intervals. In this paper we present two methods to identify systolic and diastolic intervals from arterial pressure and cardiac output (pulmonary flow) signals. The methods were evaluated by comparing the start of the systole and diastole in measurements from 6 pigs undergoing bypass surgery. Furthermore, we compared DF and PI calculated using the timings from the two different signals. We obtained a mean difference between systole start of -2.76ms (± 18.26ms) and 59.59ms (± 81.17ms) for diastole start. Finally, we obtained a correlation of 95.96% for PI and 60.30% for DF. The use of more than one recording type can be used to ensure more stable results when performing coronary graft patency quality assessments.