parviz abdolmaleki - department of bio medical physics tarbiat modares university

to develop and artificial neural network ANN to differentiate malignant from benign in a group of patients with histopathologically proved breast lesions on the base of data derived independently from time intensity curve the performance of ANN was compared with that of experienced observers using a database with 120 patients records each of which consisted of 14 quantitative parameters mostly derived from time intensity curve.

breast cancer, quantitative features dynamic magnetic resonance imaging , artificial neural network ANN , receiver operating characteristic curve ROC