A Benign and Malignant Mass Classification Algorithm Based on an Improved Level Set Segmentation and Texture Feature Analysis

In this paper, we investigate the classification of masses with texture features. We propose an improved level set method to find the boundary of a mass, based on the initial contour provided by radiologists. After the boundary of a mass is found, texture features from Gray Level Co-occurrence Matrix (GLCM) are extracted from the surrounding area of the boundary of the mass. The extracted texture features are used as the input of Linear discriminant analysis and a support vector machine to classify the mass as benign or malignant. Mammography images from DDSM were used in the experiments and the classification accuracy was evaluated using the area under the receiver operating characteristic (ROC) curve. In the proposed method the area under the ROC curve is A_z =0.7. The experimental result shows the effectiveness of the proposed method.