Texture Feature Based Analysis of Brain Ct Images for Discriminating Benign, Malignant Tumors

A computer software system is designed for the automatic discrimination of benign tumor from malignant tumor in brain CT examinations. Image analysis methods were applied to the images of 40 benign images and 40 malignant images. Textural features extracted from the gray level co-occurrence matrix of the tumor images and back propagation neural network classifier were employed for the design of the system. Best classification accuracy was achieved by four textural features and two hidden layers and 6 hidden nodes of the classifier. The proposed system provides new textural information and differentiating benign tumor from malignant tumor, especially in small tumor regions of CT images efficiently and accurately with lesser computational time.