Deformable Spatio-Temporal Shape Models

Principal Investigator: Ghassan Hamarneh

We are developing tools for modelling and analysis of spatio-temporal shape variation from dynamic medical image, as in echocardiography.

Tracking left-ventricular shape while enforcing prior spatio-temporal shape knowledge.

Comparing convergence speed of active contour models towards the contrast-agent leading front in the right ventricle (snakes vs optical-flow snakes)

Tracking the leading edge in echocardiography of the contrast agent reaching the right ventrilce

Frequency-based constrained snake deformations

Statistically constrained snakes flowchart

G. Hamarneh, T. Gustavsson. “Deformable Spatio-Temporal Shape Models: Extending ASM to 2D+Time”. Journal of Image and Vision Computing, vol 22(6), 2004, pp. 461-470.

G. Hamarneh, K. Althoff, T. Gustavsson. “Snake Deformations Based on Optical Flow Forces for Contrast Agent Tracking in Echocardiography”. Swedish Symposium on Image Analysis, 2000, pp. 45-48.

G. Hamarneh, T. Gustavsson. “Statistically Constrained Snake Deformations”. IEEE International Conference on Systems, Man, and Cybernetics, 2000, vol. 3, pp. 1610-1615.

Site by Faculty of Applied Science web team, Simon Fraser University