Fuzzy Rule-based Boundary Extraction of Plaque in Intravascular Ultrasound Image

In this paper, we propose a fuzzy rule- based boundary extraction method for a plaque in an intravascular ultrasound (IVUS) image. In the present method, a plaque boundary to be extracted is piecewise approximated by the polynomials inferred by fuzzy rules. The coecients of the polynomials are determined by the weighted least square method us- ing the separability measure obtained from the IVUS image. The proposed method achieves a precise and noise-robust extraction of the plaque with use of a small number of seed points and without any time- consuming iterative processes. The validity and the eectiveness of the proposed method have been con- firmed by applying it to the boundary extraction problem of the plaque in the IVUS image.