Cerebral autoregulation (CA) is a complex mechanism which aims at sta-bilizing cerebral blood flow (CBF) against variations of arterial pressure (AP). CBF is commonly surrogated with the CBF velocity (CBFV) recorded through transcranial Doppler device from the middle cerebral artery. Most of the studies evaluating CA need to compute mean CBFV (MCBFV) on a beat-to-beat basis in combination with mean AP (MAP), but there is not a standard approach to derive MCBFV from CBFV signal. In this study, we compare three different strategies to calculate MCBFV from CBFV signal: i) between two consecutive diastolic points detected on the CBFV signal (MCBFV_CBFV); ii) between two consecutive diastolic points detected on the AP signal (MCBFV_AP); iii) between two consecutive R-wave peaks detected on the electrocardiogram (MCBFV_ECG). We analyzed electrocardiogram, noninvasive photoplethysmographic AP and CBFV signals simultaneously recorded from 23 subjects (age: 28 ± 9 yrs, 13 female) at rest in supine posi-tion (REST) and during head-up tilt at 60° (TILT). While mean values were similar regardless of the considered strategy, MCBFV variances significantly varied with MCBFV_CBFV and MCBFV_ECG strategy producing the largest and the smallest variance respectively, both in REST and in TILT condition. This result stresses the need to standardize the approach for MCBFV computa-tion to reduce the variability of the results solely due to the method adopted for its computation and favor clinical applications of CA assessment.