SurfDist: Interpretable Three-Dimensional Instance Segmentation Using Curved Surface Patches

We present SurfDist, a convolutional neural network architecture for three-dimensional volumetric instance segmentation. SurfDist is a modification of the popular model architecture StarDist-3D which enables learning instance boundaries as closed piecewise compositions of smooth parametric surfaces. This parameterization breaks StarDist-3D’s coupling of instance dimension and instance voxel resolution, and it produces predictions which may be upsampled to arbitrarily high resolutions without introduction of voxelization artifacts. For datasets with blobshaped instances, common in biomedical imaging, SurfDist can achieve higher segmentation accuracy than StarDist3D with more compact instance parameterizations.