Parenchymal echo patterns of cirrhotic liver analysed with a neural network for risk of hepatocellular carcinoma

Background: To objectively evaluate the parenchymal echo patterns of the li ver in cirrhosis, an image analysing system in which a neural network is us ed has been found capable of numerically calculating coarse score (CS). Usi ng this system, we analysed whether or not CS can serve as a predictive fac tor for the development of hepatocellular carcinoma (HCC). Methods: The risk factors for HCC were evaluated in 95 patients with liver cirrhosis with an average follow-up period of 2041 +/- 823 days. We used a three-layer feed-forward neural network and a backpropagation algorithm to calculate CS. Results: There were strong correlations between CS, alanine aminotransferas e (ALT) and alpha-fetoprotein (AFP) and the average cumulative incidence ra te of HCC evaluated by the Cox's proportional hazards model. The adjusted r ate ratios were estimated to be 3.00, 2.80 and 2.01, respectively. The cumu lative risks of HCC were significantly higher with an initial CS greater th an or equal to 1.5 than with an initial CS < 1.5, with ALT greater than or equal to 80 IU/L than with initial ALT < 80 IU/L and with AFP greater than or equal to 20 ng/mL than with initial AFP < 20 ng/mL, all analysed by the log-rank test. Conclusions: Coarse score is a useful predictor for development of HCC. (C) 1999 Blackwell Science Asia Pty Ltd.