Has conducted research in Radiology Instalation of K.R.M.T.Wongsonegoro Hospital Semarang by using water phantom and MSCT PhilipsBrilliance CT 16 Slice, in the MSCT abdominal protocols with variations slicecollimation 4 x 0.75 mm, 4 x 1.5 mm, 16 x 0,75 mm and 16 x 1.5 mm to see thevalue of dose length product (DLP) and image noise. The purpose of thisresearch is to know differences in the value of dose length product (DLP) andimage noise in the variation of slice collimation, knowing the value of dose lengthproduct (DLP) and image noise minimum value by displaying the image of theabdominal MSCT optimal.Quantitative research with an experimental approach. The data takenfrom the results of scanning the water phantom using MSCT abdominal protocolin 4 variations of slice collimation with each variation done 3 times scanning toget the value of DLP and image noise. Then the value of DLP value is recordedand measured results of MSCT image noise by utilizing the image region ofinterest (ROI). Data analysis was performed using a different test. Statistical testthe value of DLP with descriptive test and image noise with ANOVA Repeatedmethods as well as to see the lowest score of the DLP and image noise with themean rank.The results showed that there was a differences in the value of DLP andimage noise in the variation of slice collimation, to the value of DLP lowest averageas 143,2 mGy.cm on a slice collimation 16 x 1.5 mm, for low noise image also onthe slice collimation of 16 x 1 , 5 mm. Image noise shows the p-value of 0.562,which means p-value > 0.05 so that Ho refused and Ha is received, there aredifferences in image noise on variations slice collimation, Results showed that themean rank of the lowest average for a value that is equal to 143.2 mGy.cm DLPand to image noise is 3.637 HU on slice collimation of 16 x 1.5 mm slicecollimation so that variation is the right choice.