Neural-network-based boundary detection of liver structure in CT images for 3-D visualization

This paper describes a segmentation method of liver structure from abdominal CT images using an artificial neural network (NN), together with a priori information about liver location and area in the abdomen cross section and digital imaging processing techniques. This approach based on the NN is to classify each pixel on an image into one of three categories: boundary, liver, and nonliver. Supervised training technique is used. The training data set is obtained from one of the given set of images by creating gray level histograms for the three categories. The histograms are considered as the respective feature values. Prior to NN classification, preprocessing is employed to locally enhance the contrast of the region of interest. Postprocessing are also applied after the NN classification to smooth the detected boundary. Our preliminary results show that the proposed method has potential utility in automatic segmentation of liver structure and other organs in the human body