期刊名称：International Journal of Computer Science & Technology
印刷版ISSN：2229-4333
电子版ISSN：0976-8491
出版年度：2013
卷号：4
期号：2
页码：731-735
语种：English
出版社：Ayushmaan Technologies
摘要：Now these days, breast cancer is the most common disease found in women. Mammography is preferred for the early detection of presence of tumor inside the breast. This paper presents an approach based on feedforward back propagation neural network (FFBNN) for breast tumor classification. Statistical texture features are extracted from mammograms and suitable features are selected and used to train the FFBNN. The fully trained network with different number of neurons in hidden layer is tested with unknown inputs and performance of the FFBNN method is evaluated in the form of accuracy, specificity, sensitivity, and precision for the classification of breast tumor.
关键词：Breast Tumor;Mammograms;Sensitivity;Specificity;Precision and Accuracy
