Velocity measurement in microcirculation: an estimation-theoretic method from microcirculatory temporospatial image

The movement of blood flow in microvessels has a great influence on exchange of substance and energy in tissue and cells. In this paper, a new estimation-theoretic method for measuring blood flow from microcirculatory temporospatial(T-S) image was carried out. In this method, a T-S image was formed from an image sequence. An implicit assumption about T-S image was that the movement of blood flow was one-dimensional. The orientation of local image intensity structure in T-S image contained a lot of information of the movement of blood flow. The error distribution over the search area was computed using sum_of_absolute_differences. The velocity information was recovered in the form of an estimate by using estimation-theoretic techniques. This method offered the following principal advantage. It recovered the velocity of pixels of partial or no information from pixels that have sufficient intensity variation on the edge of the stria. To illustrate the usefulness of the method, the velocity estimates and the errors were recovered from some representative T-S images, and the results were discussed.