Dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) renography, in common with other medical imaging techniques, is influenced by respiratory motion. As a result, data quantification may be inaccurate. This work presents a novel on-line approach for motion correction by implementing a spatio-temporal independent component analysis method (STICA). This methodology firstly results in removal of motion artefacts and secondly provides independent components that have physiological characteristics. The STICA was applied to 10 healthy volunteers' renal DCE-MRI data. The results were evaluated using independent component curve gradients (ICGs) from different regions of interest and by comparing them with the Rutland-Patlak (RP) analysis. The r values for the ICGs were significantly higher compared to the RP curves. The standard deviations of the IC curve gradients also showed less dispersion with comparison to the RP curve gradients across all the ten volunteers' renal data. © 2012 IEEE