A new level-set-based protocol for accurate bone segmentation from CT imaging

Abstract

A new medical image segmentation pipeline for accurate bone segmentation from computed tomography (CT) imaging is proposed in this paper. It is a two-step methodology, with a pre-segmentation step and a segmentation refinement step, as follows. First, the user performs a rough segmenting of the desired region of interest. Second, a fully automatic refinement step is applied to the pre-segmented data. The automatic segmentation refinement is composed of several sub-steps, namely, image deconvolution, image cropping, and interpolation. The user-defined pre-segmentation is then refined over the deconvolved, cropped, and up-sampled version of the image. The performance of the proposed algorithm is exemplified with the segmentation of CT images of a composite femur bone, reconstructed with different reconstruction protocols. Segmentation outcomes are validated against a gold standard model, obtained using the coordinate measuring machine Nikon Metris LK V20 with a digital line scanner LC60-D and a resolution of 28 µm. High sub-pixel accuracy models are obtained for all tested data sets, with a maximum average deviation of 0.178 mm from the gold standard. The algorithm is able to produce high quality segmentation of the composite femur regardless of the surface meshing strategy used.