Liver and liver tumor segmentation provides vital biomarkers for surgical planning and hepatic diagnosis. In this paper, we propose and validate a novel level-set method integrating an enhanced edge indicator and an automatically derived initial curve for CT based liver tumor segmentation. At the preprocessing step, the CT image intensity values were truncated to lie in a fixed range to enhance the image contrast surrounding liver and liver tumor. To remove non-liver tissues for subsequent tumor segmentation, liver was firstly segmented using two convolutional neural networks in a coarse-to-fine manner. A 2D slicebased U-net was used to roughly localize the liver and a 3D patch-based fully convolutional network was used to refine the liver segmentation as well as to roughly localize the liver tumor. A novel level-set method was then presented to further refine the tumor segmentation. Specifically, the probabilistic distribution of the liver tumor was estimated using unsupervised fuzzy c-means clustering, which was then utilized to enhance the edge-detector used in level-set. Effectiveness of the proposed pipeline was validated on two publiclyavailable datasets. Experimental results identified the superior segmentation performance of the proposed pipeline over state-of-the-art methods.

No abstract

No abstract

CBCT images have been widely used in digital orthodontics. In CBCT images, there are some tooth boundaries of blurriness and disappearance, and the teeth are of intensity inhomogeneity. In order to identify these boundaries and segment teeth, this paper develops an approach of alternate level set evolutions with controlled switch via slice-by-slice tooth segmentation: the user selects a voxel inside a tooth to initialize a small circle encircling the voxel, and the curve is automatically expanded until the tooth surface is formed. In each slice, the curve is expanded by two level set evolutions: the first level set evolution handles the intensity inhomogeneity and the second level set evolution solves the boundary blurriness and disappearance of the teeth. Experiments showed that the proposed method successfully segmented 12 CBCT datasets of 356 teeth and outperformed the geodesic active contour method and the popular method and the state-of-the-art method.