Abstract.
Automatic delineation of anatomical structures in 3-D vol- umetric data is a challenging task due to the complexity of the ob ject appearance as well as the quantity of information to be processed. This makes it increasingly di?cult to encode prior knowledge about the ob- ject segmentation in a traditional formulation as a perceptual grouping task. We introduce a fast shape segmentation method for 3-D volumet- ric data by extending the 2-D database-guided segmentation paradigm which directly exploits expert annotations of the interest ob ject in large medical databases. Rather than dealing with 3-D data directly, we take advantage of the observation that the information about position and appearance of a 3-D shape can be characterized by a set of 2-D slices. Cutting these multiple slices simultaneously from the 3-D shape allows us to represent and process 3-D data as e?ciently as 2-D images while keeping most of the information about the 3-D shape. To cut slices consis- tently for all shapes, an iterative 3-D non-rigid shape alignment method is also proposed for building local coordinates for each shape. Features from all the slices are jointly used to learn to discriminate between the ob ject appearance and background and to learn the association between appearance and shape. The resulting procedure is able to perform shape segmentation in only a few seconds. Extensive experiments on cardiac ultrasound images demonstrate the algorithm’s accuracy and robustness in the presence of large amounts of noise.