Real-Time 3D Interactive Segmentation of Echocardiographic Data Through User-based Deformation of B-spline Explicite Active Surfaces

Despite the efforts towards automatic segmentation methods, this task remains to date a supervised process in daily clinical practice. Therefore, there is a need to ensure that the operating physician is able to interact with the segmented objects in order to correct any eventual problem in a straightforward manner. We have previously developed a framework for real-time 3D segmentation of challenging data. In the present manuscript we expand this method in order to deal with contour information introduced by the user, allowing a synergistic collaboration between the operator and the underlying segmentation algorithm. In order to reduce the number of manual corrections, we propose a geometric regularization scheme which makes the interactive segmentation process less cumbersome to the user. The resulting algorithm is able to significantly increase the segmentation accuracy while keeping a low inter-observer variability. Moreover, the proposed method Lab. on Cardiovascular Imaging and Dynamics, Dept. of Cardiovascular Sciences, KU Leuven, Belgium Université de Lyon, CREATIS, CNRS UMR5220, Inserm U1044, INSA-Lyon, Université Lyon 1, France Preprint submitted to Computerized Medical Imaging and Graphics December 4, 2012 provides real-time feedback to the user, reducing the total analysis time and globally contributing to a more efficient use in daily clinical practice. Furthermore, it compares favourably to a state-of-the-art interactive segmentation software.