Systems and methods which provide active optimized spatio-temporal sampling (AOSTS) for image generation are shown. Embodiments actively select one or more regions of interest (ROIs) in a multi-beam ultrasound sampling mode, to minimize temporal image artifacts in the ROIs and thereby provide AOSTS. Such selection of ROIs according to embodiments results in various multi-beam sampling parameters, such as the number of rays that are used, the spacing between the rays that are used, the width of the rays that are used, the sequence of the rays that are used, the angle of the rays that are used, etc., being selected to provide optimized spatio-temporal sampling with respect to the selected ROIs. Selection of ROIs according to embodiments may include selecting parameters such as position, size, shape, orientation, direction, rate of change, etc.
