PCA-Based Region Growing Method and Its Application on Spleen Segmentation

The spleen segmentation in medical images is one of the most important steps for computer aided diagnosis (CAD). In this paper, we propose a novel automatic segmentation algorithm for spleen segmentation in the CT images based on Principal Component Analysis (PCA) and region growing method. Firstly, a set of CT image sequence is trained by PCA and the result is used for reconstruction. This step is able to remove the abdominal muscle area adjacent to spleen. Secondly, we obtain the image without abdominal muscle through the use of the template generated from reconstruction. Finally, we adopt region growing method to extract the interested spleen area. According to the experimental results, the PCA based region growing method can overcome the over-segmentation problem of traditional region growing method.