The goal of this study was to investigate,in vivo, human liver tissue differentiation by ultrasound. This investigation was carried out through implementation of pattern recognition techniques. Values of the power spectra of the attenuation coefficient for normal liver tissues were estimated, and 25 of these values were selected to form the reference feature vector. A decision rule based on the normalized Euclidean distance between the feature vector of the liver under consideration and the reference vector was applied to decide whether tissue was normal or abnormal. This study was a continuation of a previous one (Chu, et al., 1985). In the previous study, curve fitting techniques were used to differentiate between the spectra of normal and abnormal liver cases. It was anticipated that results based on this study would be found more accurate and more informative than the results of the previous study.