The aim of this paper is to describe a knowledge-based system that int
erprets three-dimensional (3D) coronary artery movement, using data fr
om digital subtraction angiography image sequences. Dynamic informatio
n obtained from artery centerline 3D reconstruction and optical flow e
stimation, is classified according to experimental evidence indicating
that artery displacements are quasi-homogeneous by a segment analysis
. Characteristic motion features like displacement direction, perpendi
cular/radial components, rotation direction, curvature and torsion are
qualitatively described from an image sequence using symbolic labels.
These facts are then related and interpreted using anatomical-functio
nal knowledge provided by a specialist, as well as spatial and tempora
l knowledge, applying spatio-temporal reasoning schemes. Facts, knowle
dge and reasoning rules are stated in a declarative form. Detailed exa
mples of local and global interpretation results, using a real reconst
ructed angiographic biplane image sequence are presented in order to i
llustrate how our system suitably interprets coronary artery dynamic b
ehavior. (C) 1998 Elsevier Science B.V. All rights reserved.