COMPARISON OF A GENETIC ALGORITHM NEURAL-NETWORK WITH LOGISTIC-REGRESSION FOR PREDICTING OUTCOME AFTER SURGERY FOR PATIENTS WITH NONSMALL CELL LUNG-CARCINOMA

BACKGROUND. Neural networks have been used to predict outcome in cance r patients. Their accuracy compared with standard statistical methods has nor been fully assessed. METHODS. In this study, the authors exami ned the ability to predict the outcome of surgery in 620 patients with nonsmall cell lung carcinoma (NSCLC) by a genetic algorithm neural ne twork (GANN) using Bayes' theorem compared with logistic regression, a nd the predictive value of tumor volume measures in addition to standa rd indices such as histologic type and stage, Predictive methods were compared by examining accuracy of classifying target outcome of patien ts living or dead at 6, 12, 18, and 21 months after surgery. RESULTS. GANN was a significantly better predictor of outcome than logistic reg ression at all time points (McNemar, P < 0.01). Measures of tumor volu me produced significant improvement in rile prediction of 12-, 18-, an d 24-month time points with GANN, and at 18- and 24-month time points with logistic regression (Wilcoxon matched pairs signed rank test, P < 0.02). CONCLUSIONS. In this study of surgically treated NSCLC patient s, outcome predictions were significantly Improved by including measur es of tumor volume, For predicting individual patient outcome, GANN wa s found to be highly accurate and significantly better than logistic r egression. (C) 1997 American Cancer Society.