AMNIOTIC FLUID volume (AFV) is one of the important parameters in the assessment of fetal well-being. The ability of ultrasound measurements to represent the actual AFV is unproven. This study was undertaken to compare correlation of conventional AMNIOTIC FLUID index (AFI) and radial AMNIOTIC FLUID index (RAFI) as a new method with actual FLUID volume on phantom. As an experimental study, 10 to 100 ml of water with 5 ml intervals was injected to a rubber bladder as a uterus phantom containing a 15 week gestational age fetus. The vertical diameter was measured in largest FLUID pouch at each quadrant. Four diameters were summed as conventional AFI. The largest radial diameter perpendicular to uterus and fetus was measured at four quadrants and were summed as RAFI. Databases were analyzed based on correlation and regression methods. RAFI and conventional AFI predicted 91.6% and 65% of variations of FLUID volume, respectively (P < 0.001). In conclusion, RAFI is more accurate and reliable than conventional AFI in the prediction of injected FLUID volume.