An Improved Random Walker with Bayes Model for Volumetric Medical Image Segmentation

Figure 3

Steps of the RWBayes method. (a) The segmented liver (red) of the previous slice; (b) the current slice; (c) candidate pixel by thresholding the GM; (d) the rough object (red) and background (green) seed points; (e) the fine seed points using a NBT method; (f) the initial segmentation result by RWBayes; (g) smoothing the boundary by Fourier transform; and (h) visualisation of the segmented liver volume.