Use of artificial neural networks in modeling associations of discriminant factors: towards an intelligent selective breast cancer screening

摘要: In order to improve the costs/benefits ratio of breast cancer (BC) screenings, author evaluated performance a back-propagation artificial neural network (ANN) predict an outcome (cancer/not cancer) be used as classificator. Networks were trained on data from familial history cancer, and sociodemographic, gynecoobstetric dietary variables. The ANN achieved up 94.04% positive predictive value 97.60% negative value. Results could operate guidelines for preselecting women who would considered BC high-risk subpopulation. procedure--not only based age factor, but multifactorial basis--appears promising method achieving more efficient detection preclinical, asymptomatic cases.