A program is described that performs Patlak graphical analysis of the brain uptake curve from the bolus phase of an ECD or HMPAO study. The aortic arch curve is used as the input function (Matsuda, EJNM 1992). This yields the unidirectional uptake coefficient that is related to absolute cortical bloodflow (aCBF) in ml/100g/min and transport of the agent into the neurons. The absolute nature of this measure has the potential to detect global cerebral microvascular or neuronal pathology that is not possible with the SPECT scan. Measuring aCBF with multiple camera heads permits its evaluation beyond the obligatory anterior view that primarily reflects frontal aCBF. Additional program features include correction for contamination of the aortic-arch (input) curve by activity from underlying vessels, better evaluation of the aortic-arch to brain delay, and subtraction of residual activity from a previous injection. We have analysed 17 patients at random (aged 40 to 90) who presented for HMPAO brain scans on a 3-head camera. Non aortic arch activity in the input curve underestimated aCBF by as much as 15%. aCBF was relatively unaffected by small changes in the aortic arch ROI or the bolus delay. Measured aCBF ranged from 10% above to 32% below the age-corrected normal value. The aCBF from the LPO and RPO views were at most 11% below the ANT value. When a reliable age-dependence for aCBF in the normal aged population has been established, this technique can be evaluated for the detection of changes in aCBF in early dementia.