Figure 5: Interobserver reliability for CBF quantification between two trained, independent physicians, from different institutions, who constructed CTP datasets using standard post-processing guidelines (15). The x-axis represents the 10 different patient’s datasets, and the y-axis represents mean CBF values (mL/100gm/min) for each patient, measured at IDENTICAL ROI locations within the infarcted territories. The mean CBF, CBV and MTT quantitative values were not statistically different between the two observers (p<0.001) (16). The variability in perfusion parameters among the different patients were reflected in each of the observers’ values (blue and magenta plots). There was a strong correlation between quantitative CTP values for maps created at the different institutions, representing good interobserver reliability.