An improved model-based vessel tracking algorithm with application to computed tomography angiography
Cezary Bołdak, Yan Rolland, Christine Toumoulin · 2003
This paper reports newest improvements proposed for the segmentation and the characterisation of three-dimensional vessels observed through Computed Tomography Angiography using geometrical moments. Several adaptive controls are introduced, which allow to deal with pathological patterns such as dense and scattered calcifications. The reduced time computation makes the algorithm capable to face clinical constraints in routine use. Examples are given on several data sets that highlight critical situations to handle.