A real time contour tracking system to investigate the cross-sectional area changes of the aorta

The study of the vascular wall motion on sequences of echographic images requires the analysis of a large number of images. Therefore the tedious and time consuming process of having to manually trace the vascular contours is unacceptable. Here, the authors present an automatic and real time contour tracking procedure which they used to investigate the cross sectional changes of the aorta. The procedure is based on the computation of the mass center of the first order absolute moment (FOAM). A pure software solution based on a Personal Computer did not allow the authors to obtain real time performances. Consequently, the procedure was implemented on an integrated software/hardware environment based on the interaction of a PC and a DSP board. The quantitative assessment of the system was performed by using linear regression of the cross-sectional area of expert-traced contours and computer detected contours.

[1]  James S. Duncan,et al.  Shape-based tracking of left ventricular wall motion , 1997, IEEE Transactions on Medical Imaging.

[2]  Andrew Blake,et al.  Robust contour tracking in echocardiographic sequences , 1998, Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271).

[3]  M. Demi,et al.  The first order absolute moment in contour tracking , 1997, Computers in Cardiology 1997.

[4]  M. Demi,et al.  How the map of mass centers of the first order absolute moment can be used to track contours in image sequences [cardiovascular application] , 1998, Computers in Cardiology 1998. Vol. 25 (Cat. No.98CH36292).

[5]  Tomaso A. Poggio,et al.  On Edge Detection , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.