作者: Bob Djavan , Mesut Remzi , Alexandre Zlotta , Christian Seitz , Peter Snow
DOI: 10.1200/JCO.2002.20.4.921
关键词:
摘要: PURPOSE: Two artificial neural networks (ANN) for the early detection of prostate cancer in men with total prostate-specific antigen (PSA) levels from 2.5 to 4 ng/mL and 10 were prospectively developed. The predictive accuracy ANN was compared that obtained by use conventional statistical analysis standard PSA parameters. PATIENTS AND METHODS: Consecutive a serum level between (n = 974) 272) analyzed. A separate model developed each group patients. Analyses performed determine presence cancer. RESULTS: area under receiver operator characteristic (ROC) curve (AUC) 87.6% 91.3% models, respectively. For latter model, AUC generated significantly higher than produced single variables PSA, percentage free density transition zone (TZ), TZ volum...