Real-time free breathing cardiac cine imaging is a reproducible method with shorter acquisition time and without breath-hold for cardiac magnetic resonance imaging. However, the detection of end-diastole and end-systole frames of real-time free breathing cardiac cine imaging for left ventricle function analysis is commonly completed by visual identification, which is time-consuming and laborious. In order to save processing time, we propose a method for semi-automatic identification of end-diastole and end-systole frames. The method fits respiratory motion signal and acquires the expiration phase, end-diastole and end-systole frames by cross correlation coefficient. The procedure successfully worked on ten healthy volunteers and validated by the analysis of left ventricle function compared to the standard breath-hold steady-state free precession cardiac cine imaging without any significant statistical differences. The results demonstrated that the present method could correctly detect end-diastole and end-systole frames. In the future, this technique may be used for rapid left ventricle function analysis in clinic.
