The CT Scan examination technique that in the last decade is popular in diagnosing urinary stones is NCCT Urography examination. This examination is Fast and informative, but this examination has a deficiency that the resulting image has a fairly high noise. Iterative Reconstruction is a method of algorithm reconstruction on CT Scan with the basic principle of estimating data to produce reconstruction image by reducing noise. The purpose of this study was to find out the difference in noise value in the use of Iterative Reconstruction in reducing noise and improving the quality of NCCT Urography image. This research was conducted retrospectively, namely on existing image given iterative reconstruction treatment of 20%, 40%, 60%, 80% and 100%. Samples in each iterative reconstruction treatment group as many as 10 images. Image is done noise measurement by doing ROI in kidneys, ureter and VU. Data analysis is done by conducting different tests of One Way Anova using SPSS Software. The results showed that there was a significant difference in noise value after it was done Iterative Reconstruction. Noise decreased as the percentage of used Iterative Reconstruction increased in NCCT Urography image. In the use of Iterative Reconstruction percentage of 100% obtained the lowest average noise value of 9.11. The use of Iterative Reconstruction can reduce noise in NCCT Urography image by 9.386% compared to Filtered Back Projection (FBP). Iterative reconstruction is able to reduce noise and improve the image quality of NCCT Urography.