High Temporal Resolution Reconstruction Algorithm to Reduce Cardiac Motion Artifact

Abstract

Cardiac pulsation is the most important cause of serious image degradation on high resolution CT(HRCT) images of lung adjacent to heart and cardiac images. The purpose of this study was to propose a special high temporal resolution reconstruction(HTRR) algorithm, and to assess the effects on the reduction in motion artifacts caused by cardiac pulsation on HRCT images of lung and cardiac images by using this algorthm combined with sub-second CT. The HTRR algorithm was developed on the basis of a special weighting function. The results of phantom and clinical studies demonstrated that the HTRR algorithm can provide high temporal resolution, substantially reduce moting artifacts and noise, and markedly improve the quality of cardiac images and HRCT lung images at the cost of a slight degradation in spatial resolution. Further, the HTRR algorithm can obtain phase information by combining subsecond CT with the overlapping cine reconstruction technique and ECG information. Using the HTRR algorithm, the effective temporal resolution(FWHM) is 0.4 sec for a 0.8 sec scan.