Quantitative measurement of cerebral blood flow by dynamic CT

We propose a new “BOX-MTT method” to measure the regional cerebral blood flow (rCBF) quantitatively using a fourth-generation CT scanner. In this method, the time-density curve (TDC) of the region of interest (ROI) on the brain tissue is assumed to be proportional to convolution of the TDC of the artery feeding the ROI and the modulation transfer function (MTF). Although the transit pattern of the blood through the brain tissue is not well known, we hypothesise that the MTF is box-shaped and have named this model the BOX-MTT model. The rCBF and regional cerebral blood volume rCBV were measured by our method in eight patients with various neurological diseases. Our model fitted the original TDC well, and rCBV and rCBF values in various parts of the normal hemispheres were similar to those measured by PET or SPECT. By applying deconvolution analysis, we found that the MTF in the brain tissue was approximately box-shaped. Our method offers quantitative rCBF values by fitting TDCs directly to the BOX-MTT model.