Motion artifact in T2-weighted fast spin-echo images of the liver: effect on image contrast and reduction of artifact using respiratory triggering in normal volunteers.

Abstract

The purpose of our study was to evaluate the effect of respiratory motion on the image contrast of T2-weighted fast spin-echo (FSE) images of the liver as well as the reduction of motion artifact using respiratory triggering of the data acquisition. We imaged the livers of 10 healthy volunteers using a fast spin-echo T2-weighted sequence. Images were obtained both without and with patient triggering. Triggered images were acquired in a segmented fashion during multiple sequential breath-holds using an echo train of 8 or 16, both with and without flow compensation (gradient moment nulling). Ratios of signal difference to noise (SD/N) of the liver and gallbladder as well as the liver and spleen were compared for all sequences. All of the triggered images showed statistically significant improvement of SD/N for the liver and gallbladder as well as for the liver and spleen when compared with the nontriggered images. Triggered images obtained with an echo train length of 8 and, with flow compensation, showed the highest SD/N ratios. In one volunteer whose liver contained multiple small cysts, the triggered images showed improved visualization of individual cysts and identified a larger number of cysts. Respiratory motion causes a significant loss of contrast on T2-weighted fast spin-echo images of the liver. This can be reduced by using a segmented data acquisition triggered by the respiratory cycle obtained during sequential breath-holds.