Various
studies have
demonstrated the potential benefits of image guidance in robot-assisted
surgery
and therapy. One of the most challenging applications of image guidance
is
beating heart tracking in order to automatically synchronize the
motions of a surgical
robot to the motions of the heart during off-pump cardiac surgery,
which has
strong advantages over on-pump arrested heart surgery. For the above
application, I am developing algorithms that use three-dimensional (2D
+ time)
and four-dimensional (3D + time) cardiac ultrasound images for
semi-automatic
or fully-automatic real-time segmentation and tracking of heart
structures
(e.g., mitral valve annulus, atrial septal defect, etc.) that require
surgical
intervention.