Heart rate fragmentation and variability from real-time cardiovascular magnetic resonance imaging using TREND-CMR software

Real-time cardiovascular magnetic resonance (rtCMR) is advantageous for imaging irregular cardiac cycles. However, software interpretation of the varied cardiac cycles imaged has been lacking. We developed TREND-CMR software to assess temporal variability in rtCMR. The temporal parameters were extracted from rtCMR scans using principal component analysis and corroborated by corresponding results from ECG and segmentation of the left ventricular blood pool. TREND-CMR provides multiple plots of temporal characteristics and video excerpts of the aberrant heartbeats of a subject. We adapted heart rate fragmentation (HRF), an innovation in electrophysiology, to rtCMR. We also describe new HRF metrics called jerkiness and fragmentation of left ventricular filling time. Applied to the rtCMR scans of 20 volunteers (mainly short axis, but also two-chamber views), jerkiness and HRF inflections suggested differences of two subjects with heart conditions. Regular alteration of beats at inspiration were noted in a subject with respiratory sinus arrhythmia apparent. TREND-CMR also supports motion correction of rtCMR imaging by sorting by respiratory phase. The simple interface, multi-panel plots, video excerpts of atypical beats, will make it easy for the community to quantify HRF and inspect aberrant cycles among a series of imaged beats.