Parallel process for level set segmentation of volume data

A method comprising:

• a) using a processor initializing in parallel operation a level set field within a multi-dimensional dataset comprising a plurality of voxels, the multi-dimensional dataset being indicative of an object;
• b) using the processor generating in parallel operation an initial set of active voxels in dependence upon the initialized level set field;
• c) using the processor determining in parallel operation new active voxels corresponding to an intersection of active voxels changing in time and active voxels changing in space;
• d) using the processor removing in parallel operation duplicate active voxels;
• e) using the processor compacting in parallel operation the non-duplicate active voxels;
• f) using the processor repeating c) to e) until a quantitative stopping criterion is met;
• g) using the processor determining level set segmentation data in dependence upon the level set field, the segmentation data being indicative of a feature of the object;
• and,
• h) using the processor providing the level set segmentation data;
• and
• wherein steps b) to f) the active voxels are stored in a plurality of auxiliary buffers.

A non-transitory storage medium having stored therein executable commands for execution on a processor, the processor when executing the commands performing:

• a) initializing in parallel operation a level set field within a multi-dimensional dataset comprising a plurality of voxels, the multi-dimensional dataset being indicative of an object;
• b) generating in parallel operation an initial set of active voxels in dependence upon the initialized level set field;
• c) determining in parallel operation new active voxels;
• d) removing in parallel operation duplicate active voxels;
• e) compacting in parallel operation the non-duplicate active voxels;
• f) repeating c) to e) until a quantitative stopping criterion is met;
• g) determining level set segmentation data in dependence upon the level set field, the segmentation data being indicative of a feature of the object;
• and,
• h) providing the level set segmentation data.