The purpose of the study was to overcome inherent motion problems in MRI based fetal scans by applying an interactive real-time MRI technique and to evaluate the diagnostic usefulness of the applied real-time MRI sequence in relation to standard protocols. Ten healthy women with uncomplicated pregnancies and a mean age of 29.6 years (24-33 years of age) in their 21.3 gestational week (week 20+4 days to week 22+0 days) were included in the study. The project was reported to the local ethical committee. The experiments were carried out on a pulse programmed Siemens Avanto 1.5T MRI scanner (Siemens, Erlangen, Germany, software release VB17a) with a 32 channel cardiac receive coil. All interactive real-time MRI images were reconstructed using the 'Gadgetron' software package (3). The interactive real-time protocol was a dynamic balanced SSFP 2D single-shot sequence with a field of view and matrix size adapted to the size of each volunteer. Images from the real-time group were compared to standard fetal images from 14 women (mean age 32.1 years; 24-40 years) who had previously been scanned at our department using standard fetal protocols. The two groups were matched by gestational week only (week 21.0; week 20+1 days to week 22+2 days). The image data from the standard protocol group was acquired using a range of sequences including 2D fast spin-echo T2-weighted single shots protocol. Images from both groups were analyzed retrospectively by an experienced radiologist with regard to fetal cerebral anthropometrics and presence of cerebral structures. Fetal motion, image quality, ability to obtain exact slice planes and the apparent diagnostic quality of the images were all scored on a 1-5 scale. MRI images were successfully acquired, reconstructed and viewed in real-time in all 10 pregnant volunteers at a frame-rate of approximately 1.7 images per second. Examples of the real-time and standard protocol image quality are presented in Figure 1. Table 1 lists the anthropometrics data, sums up the assessment of fetal brain structures, lists the degree of movement during scanning, the image quality and the diagnostic value and the ability to obtain correct slice planes. The presented interactive real-time MRI slice positioning system was used to obtain correct slice planes in all three orthogonal directions. Levine et al. (4) also implemented an interactive real-time system for fetal imaging. However, the frame-rate was low (0.5-0.8 images per second) and therefore more susceptible to motion artifacts. Jiang et al. (5) achieved high-resolution 3D images of the fetal brain by a dynamic single shot scan technique using multiple overlapping slices, but required considerable post-processing time which increases the risk of re-scanning if the quality of the final images are not of diagnostic quality. Anthropometrics of the fetal brain could be measured using both protocols. The size of the fourth ventricle was significantly smaller in the real-time group while the cerebral biparietal diameter was significantly larger in the real-time group (Figure 2). Both structures need exact oriented slice planes to be correctly measured. In situations of profound fetal head movement, the real-time approach provides the operator with the ability to interactively alter the slice orientation while the standard protocol would have to repeat the scan until a satisfactory slice orientation is obtained. The real-time sequence yielded images with a mean diagnostic value score of 3.2±1.2 indicating that the images had an acceptable diagnostic value and may therefore be considered as a useful supplementary tool when imaging the fetal brain. Interactive real-time MRI is a promising alternative to traditional fetal MRI for anthropometrics or as a supplement for representation of fetal brain structures in cases in which fetal motion causes challenges in relation to obtaining optimal slice planes using conventional MRI techniques.