Image processing method and apparatus

摘要

A method of selecting image data representative of a subject from an image data set comprises determining regions of image data, wherein each region of image data consists of a respective plurality of connected voxels, and selecting at least one region as being representative of the subject based upon at least one of the size and shape of the region.

主权项

1. A method of selecting image data representative of a subject from an image data set, the image data set corresponding to the subject and a structure except the subject, comprising
dividing image data into a plurality of regions; identifying from amongst the plurality of regions at least one region that is representative of a predetermined region based on a value of a geometrical classifier, the geometrical classifier distinguishing between parts of the subject and parts of the structure and being a function of a shape of said at least one region; excluding data representative of the identified at least one region that is representative of the predetermined region; determining first regions and second regions from the plurality of the regions, the first regions being representative of parts of the subject and the second regions corresponding to candidates of parts of the structure; selecting at least one region, from the second regions corresponding to said candidates of said parts of the structure except the subject; and connecting the selected at least one region, selected from the second regions, to the first regions to define said image data representative of the subject.