Measuring the function of the cerebral small vessels can greatly benefit our understanding of cerebral small vessel disease. Recent research enabled the assessment of blood flow velocity and pulsatility in the perforating arteries of the cerebral white matter in healthy subjects. However, in patients this method requires manual elimination of artifacts. This paper explores various methods for excluding false positives. The reproducibility of the velocity, pulsatility and number of selected vessels was investigated in a test-retest study. Results show that the reproducibility of these outcome measures highly depends on the chosen method for vessel selection.