AN AUTOMATIC DIAGNOSTIC SYSTEM FOR CT LIVER IMAGE CLASSIFICATION

Computed tomography (CT) images have been widely used for liver diseas e diagnosis. Designing and developing computer-assisted image processi ng techniques to help doctors improve their diagnosis has received con siderable interests over the past years. In this paper, a CT liver ima ge diagnostic classification system is presented which will automatica lly find, extract the CT liver boundary and further classify liver dis eases, The system comprises a detect-before-extract (DBE) system which automatically finds the liver boundary and a neural network liver cla ssifier which uses specially designed feature descriptors to distingui sh normal liver, two types of liver tumors, hepatoma and hemageoma. Th e DBE system applies the concept of the normalized fractional Brownian motion model to find an initial liver boundary and then uses a deform able contour model to precisely delineate the liver boundary. The neur al network is included to classify liver tumors into hepatoma and hema geoma, It is implemented by a modified probabilistic neural network (P NN) [MPNN] in conjunction with feature descriptors which are generated by fractal feature information and the gray-level co-occurrence matri x. The proposed system was evaluated by 30 liver cases and shown to be efficient and very effective.