Bones segmentation techniques in computed tomography, a survey

Image segmentation is one of the most critical functions in image analysis and processing. In orthopedics, bones segmentation of extremities, joints and fractures is a sensitive step, crucial for many applications as surgery planning, fractures fixation and additive manufacturing. A huge number of segmentation algorithms exist that simplify this tedious process with automatic or semiautomatic approach. The aim of this study is to present techniques used for Computed Tomography (CT) images segmentation, for each group of algorithms examples in orthopedic field are provided, focused on extremity, joints, and fractures segmentation algorithm. Segmentation techniques are discussed divide in five groups, pixel based, region based, edge based, deformable models and artificial neural networks. Each approach advantages and disadvantages are given in tabular form. This study underlines that pixel and region-based methods are used in orthopedics bones segmentation, however all more recent literature uses deformable models or artificial neural network for more complex bone district.