An Algorithm Embedded Prior Shape Information for Segmentation of Medical Image Series

An algorithm based on the active shape model is proposed for the segmentation of medical image series. The edge of soft tissues in medical image usually are not very clearly, so it´s difficult to segment them only by the computer. But a tissue´s shape deform sequentially in medical image series. We selected some images from the image series and segmented them with the help of doctors, the results of the segmentation was used as the training maps to create an active shape model by PCA Algorithm. We used this model to segment remainder images in the image sequence by mutual information registration and active contour model. The experiments show the efficiency of this method.