Evaluation of texture analysis techniques for quantitative characterization of ultrasonic liver images

In this study, we attempt to determine the efficacy of techniques computer assisted ultrasonic liver tissue characterization characterization were used in this study; the gray level difference statistics algorithms (GLDS) find the fractal dimension texture analysis (FDTA). Both techniques were applied on three sets of ultrasonic liver images, normal-hepatoma-cirrhosis, all spectrum histologically proven. In all images, 32x32 pixel rectangular (GLDS) regions-of-interest were selected by specialized physicians recently ;Ind used in the analysis. FDTA was-able to differentiate [4,5]. using texture analysis techniques. Two different algorithms ultrasonic from normal with an accuracy of 75L%. The combination of the two techniques proved to differentiate the three types with an overall accuracy of 81.7%. .