The singular value decomposition deconvolution of cerebral tissue concentration-time (C-T) curves with the arterial input function (AIF) is commonly used in dynamic susceptibility contrast (DSC) cerebral perfusion MR imaging. However, it is sensitive to the time discrepancy between the arrival of the bolus in the tissue C-T curve and the AIF signal. This normally causes inaccuracy in the quantitative perfusion maps due to delay and dispersion effects. A comprehensive correction algorithm has been achieved through slice-dependent time-shifting of the AIF, and a delay-dependent dispersion correction model. The correction algorithm was tested in 11 healthy subjects and 3 ischemic stroke patients scanned with a quantitative perfusion pulse sequence at 1.5T. A validation study was performed on 5 patients with confirmed cerebrovascular occlusive disease scanned with MRI and positron emission tomography (PET) at 3.0T. A significant effect (p