Abstract: An image segmentation method. An image is defined by voxels representing at least two different structure types A and B, by having corresponding different intensities that may be subject to intensity inhomogeneities. A basic method comprises (a) selecting a first voxel to be classified; (b) defining criteria by which a comparison of the intensity associated with the first voxel with an intensity associated with a second voxel is to be considered to indicate that the first voxel represents structure type B assuming the second voxel represents structure type A; (c) determining a path of voxels passing through the first voxel; (d) defining a maximum distance along the path; and (e) determining whether there is a second voxel on the path, within the maximum distance of the first voxel, such that the intensities associated with the first and second voxels together satisfy the criteria, for potentially classifying the first voxel as representing structure type B.

Abstract: An image segmentation method. An image is defined by voxels representing at least two different structure types A and B, by having corresponding different intensities that may be subject to intensity inhomogeneities. A basic method comprises (a) selecting a first voxel to be classified; (b) defining criteria by which a comparison of the intensity associated with the first voxel with an intensity associated with a second voxel is to be considered to indicate that the first voxel represents structure type B assuming the second voxel represents structure type A; (c) determining a path of voxels passing through the first voxel; (d) defining a maximum distance along the path; and (e) determining whether there is a second voxel on the path, within the maximum distance of the first voxel, such that the intensities associated with the first and second voxels together satisfy the criteria, for potentially classifying the first voxel as representing structure type B.