APPLICATION OF FUZZY-MLP MODEL TO ULTRASONIC LIVER IMAGE CLASSIFICATION
Aborisade David. O, Ojo John. A, Amole Abraham. O · 2014
In this paper, we propose the application of fuzzy-MLP in the classification of ultrasonic liver images. The four sets of ultrasonic liver images used in the experiment are: normal, liver cysts, alcoholic cirrhosis and carcinoma. To deal with the sample images efficiently, we extract textural features from the Pathology Bearing Regions (PBRs) of the ultrasound liver images. The selected features for the classification are entropy, energy and maximum probability-based texture features extracted using gray level co-occurrence matrix second-order statistics. The fuzzy-MLP model is constructed for the selected features classify various categories of ultrasonic liver images. The efficacy of Fuzzy-MLP model and conventional artificial neural network (ANN) has been compared on the basis of the same feature vector. A test with 82 training data and 110 test data for all the four classes shows 92.73% classification accuracy for the proposed fuzzy-MLP model. It is compared with the 81.82 % counterpart provided by conventional ANN method.