Chamber reconstruction from a partial volume

A method, comprising:

• mapping, to respective first voxels, respective coordinates of a plurality of locations within an anatomical cavity;
• assigning, to each of the first voxels, a first value f(0),
• f(n) being a monotonic function defined for each integer n between 0 and M inclusive, M being a predefined positive integer;
• valuing a plurality of second voxels {vi}, each of which is at a distance 1≤d(vi)≤M voxels from a nearest one of the first voxels, by assigning, to the second voxels, respective second values {f(d(vi))};
• subsequently to valuing the second voxels, iteratively revaluing the second voxels by iteratively assigning, to each voxel of at least some of the second voxels, a weighted average of respective values of immediate neighbors of the voxel,
• in which weighted average any one of the values differing from f(M) by not more than a first predefined threshold Δ is given a higher weight than any other one of the values;
• subsequently to iteratively revaluing the second voxels, identifying a subset of the second voxels, each of which has a value differing from f(M) by more than a second predefined threshold;
• and
• by applying a mesh-generating algorithm to a volume that includes the first voxels and the subset of the second voxels, generating a mesh representing a surface of the volume.

A computer software product comprising a tangible non-transitory computer-readable medium in which program instructions are stored, which instructions, when read by a processor, cause the processor to:

• map, to respective first voxels, respective coordinates of a plurality of locations within an anatomical cavity,
• assign, to each of the first voxels, a first value f(0),
• f(n) being a monotonic function defined for each integer n between 0 and M inclusive, M being a predefined positive integer,
• value a plurality of second voxels {vi}, each of which is at a distance 1≤d(vi)≤M voxels from a nearest one of the first voxels, by assigning, to the second voxels, respective second values {f(d(vi))},
• subsequently to valuing the second voxels, iteratively revalue the second voxels by iteratively assigning, to each voxel of at least some of the second voxels, a weighted average of respective values of immediate neighbors of the voxel,
• in which weighted average any one of the values differing from f(M) by not more than a first predefined threshold Δ is given a higher weight than any other one of the values,
• subsequently to iteratively revaluing the second voxels, identify a subset of the second voxels, each of which has a value differing from f(M) by more than a second predefined threshold, and
• by applying a mesh-generating algorithm to a volume that includes the first voxels and the subset of the second voxels, generate a mesh representing a surface of the volume.