SHAPE-BASED INTERPOLATION OF MULTIDIMENSIONAL OBJECTS

Three-dimensional image data produced by medical imaging scanners usually have unequal scanning resolution in different dimensions, the slice separation usually being much greater than the pixel size within an individual slice. The general practice in the three-dimensional display of organs based on such data is to first interpolate between slices to obtain isotropic resolution and then perform object identification and display operations. This implies that, if user interaction is required on a slice-by-slice basis to identify the object of interest (which often is the case), the effort needed is much (two-ten times) more than that required if the uninterpolated slice data were used for identifying the object. The situation becomes much (ten-50 times) worse if a fourth dimension of time is also involved, as in the case of a dynamic organ such as a beating heart. We present a new shape-based interpolation scheme for multidimensional images, which consists of first segmenting the given image data into a binary image, converting the binary image back into a gray image wherein the gray value of a point represents its shortest distance (positive value for points of the object and negative for those outside) from the cross-sectional boundary, and then interpolating the gray image. The set of all points with nonnegative values associated with them in the interpolated image constitutes the interpolated object. The method not only greatly minimizes the user involvement in interactive segmentation, but also leads to more accurate representation and depiction of dynamic as well as static objects. We present the general methodology and the implementation details of the new method and compare it on a qualitative and quantitative basis to the existing methods. The generality of the proposed scheme is illustrated with a number of medical imaging examples. © 1990 IEEE