Objective To discuss the application of dual-source computed tomography （CT） low dose technology in the upper abdomen enhanced inspection. Methods Six hundred consecutive patients from July 2011 to February 2012 in this hospital were orderly divided into ordinal tube current （210 mAs） group and low dose tube current group （200 mAs，190 mAs，180 mAs，170 mAs，and 165 mAs）. The standard deviation （SD） of subcutaneous fat，signal to noise ratio （SNR） of liver and pancreas， contrast to noise ratio （CNR） of liver-erector spinae and pancreas-erector spinae，score of subjective diagnosis and the indexes of radiation dose，including CT dose index （CTDI），dose length product （DLP），effective dose （ED） were measured，calculated，and assessed respectively on CT images of arterial phase and portal phase from each group. Results The SD of subcutaneous fat， SNRs of liver and pancreas，score of subjective diagnosis，and the CTDI，DLP，ED of CT images in arterial phase and portal phase were significantly different from each other （P＜0.05），while CNRs of liver-erector spinae and pancreas-erector spinae were not statistically significant （P＞0.05）. The SNR，radiation dose，and score of subjective diagnosis of the 165 mAs tube current group were the lowest among all the groups，but the images of the 165 mAs tube current group could not fulfill the need of diagnosis. Conclusions It is really feasible that the method of decreasing tube current gradually in the upper abdomen enhanced CT inspection could ensure that the radiologists could adapt the low dose image bit by bit，and this methods could be popularized to all kinds of CT facilities we own currently. The images with 170 mAs as tube current in the upper abdomen enhanced inspection of dual-source CT could fulfill the need of diagnosis，and the radiation dose of patients is apparently lower than that the conventional scan．