Estimation of 2-D blood flow velocity map from cine-angiograms: algorithm using overlapping block set and illustration of vortex flow in abdominal aneurysm.

Abstract

To derive blood flow dynamics from cine-angiograms (CAGs), we have developed an image-processing algorithm to determine a two-dimensional blood flow velocity map projected on CAGs. Each image data of CAG is divided into a set of overlapping blocks, and it is assumed that the contrast medium in each block moves only to its 'adjacent blocks' between two serial frames. Based on this assumption, a 'fundamental equation' and the 'maximum flow constraints' are derived. These equations state the relationship between the volume of contrast medium in each block and the volume of contrast medium flowing from/to its 'adjacent blocks'. The volume of the flowing contrast medium is calculated using these relationships, boundary conditions and an additional 'smoothness constraint'. The blood flow velocities are estimated from the volume of the flowing contrast medium and are illustrated with a needle diagram. We applied our algorithm to an abdominal CAG (clinical data). The result showed a vortex flow in the abdominal aneurysm, which was consistent with visual inspection of the CAG movie and with the existence of thrombus in the aneurysm. Our algorithm may be a useful diagnostic tool to assess vascular disease.