Previously, we have demonstrated that we can extract systemic low frequency oscillation (sLFO) from resting state (RS) fMRI data. This sLFO is present throughout the brain; temporally-shifted versions of the signal appear in different voxels. A dynamic flow of sLFO can therefore be derived from the RS data ^1,2. We hypothesized that the flow pattern of sLFO reflects the passage of cerebral blood through the brain. We tested this hypothesis by conducting a Dynamic Susceptibility Contrast (DSC) MRI scan (i.e. bolus tracking)³ following the resting state scans on 8 healthy subjects. The studies were conducted in 8 healthy participants. The RS scan was conducted before the DSC scan. In the RS scan, multiband EPI data were obtained (University of Minnesota sequence cmrr_mbep2d_bold R010 ^4,5) with the following parameters: TR/TE = 720/32 ms, 500 time points, matrix 86 x 86 on a 212 x 212 mm FOV, multiband factor=8, 64 2.5-mm slices with no gap. This was followed by a Gd-DSC scan (TR/TE = 1510/21ms, 120 time points, flip angle = 60°, 1.8x1.8 mm inplane resolution, 29 3.5mm slices with a 0.5 mm gap). Twenty seconds after the beginning of this scan, 20 ml of Prohance was injected into the antecubital vein at 4 ml/s, followed by a flush of 20ml of saline using an Medrad Spectris Solaris power injector. The flow pattern of sLFO was derived from the RS data of each subject with the following steps: 1) a seed time series was extracted from superior sagittal sinus, and 2) used to cross-correlate with all the other voxels. The time lag of maximum correlation coefficient of each voxel represents the relative arrival time of the sLFO at that voxel. A time lag map was generated for each subject. Similarly, the arrival time of the Gd contrast bolus at each voxel was estimated by the time to peak (TTP) value of the DSC time course. These two time lag maps are compared. Finally, dynamic movies can be generated from the time lag maps. Each non-overlapping consecutive frame of the movie consists of voxels whose time lags are within a 0.72s time window. A 4-second movies were made by concatenating all these frames in sequence.The time lag maps derived from RS data and DSC are robust among all the subjects. The averaged time lag maps are shown in Figure 1(a) and (b) respectively. Since the signal to noise ratio of sLFO in RS data is much smaller than the bolus contrast in DSC, time lag map of RS has more noise. In Figure 1(a), early time lags can be found in symmetric central regions, such as motor cortex. Late time lags are mainly concentrated in white matter and draining veins. Compared to RS time lag maps, DSC map showed much more uniform region in most of the gray matter. The voxel with late time lags are concentrated in white matter and draining veins, like that of RS time lag map. The dynamic movie frames of two flows are shown in Figure 2. Red represents the flow of sLFO (RS) in the brain, while blue represents the flow of the Gd contrast bolus (DSC). Both similarities and differences can be observed. To a large extent, the flow of sLFO matches the contrast flow (DSC). The discrepancy is most likely due to the fact that fMRI is only sensitive to blood in capillaries and veins (due to high deoxyhemoglobin concentration⁶), while DSC is sensitive to blood in all the vessels (due to exogenous contrast (i.e. Gd) in all vessels). In other word, DSC observes much faster blood flows than that of RS even from the same voxels, resulting in shorter time lags (e.g. uniform blue in the Figure 1(b)).In this study, we tested the hypothesis that sLFO extracted from RS data travel with the blood flow, arriving in different parts of the brain with different delays. By comparing the flow patterns of the injected bolus with that of sLFO, we found that the flow of sLFO reflects blood flow in the brain, to large extent. However, since the contrast mechanism of fMRI is BOLD, which is primarily sensitive to deoxyhemoglobin concentration, the sLFO method reflects mostly the blood flow in the capillaries and veins. The results from this study can be used to develop new and simple perfusion methods without exogenous contrast; because of the higher sensitivity to venous blood, this perfusion method offers complementary information to DSCMRI.