creating an experiment using questionnaires and subliminal images

I am trying to create an experiment using a series of questionnaires, followed by a video containing subliminal images (flashed at approximately 16 milliseconds each), followed by more questionnaires. The video containing subliminal images was created in DirectRT. I have created the experiment file in MediaLab, and everything works. However, the images are still able to be consciously perceived. From other posts, I have gathered that MediaLab might not be able to handle millisecond timings. If this is the case, how would I go about creating an experiment that runs smoothly through the process of questionnaires, video, more questionnaires? Ideally, I would like this to run automatically with no experimenter involvement throughout the process (as it does in an experiment file in MediaLab).

Thanks!