1. CLASSIFICATION OF BREAST CANCER MAMMOGRAPHIC DATA.
- Author
-
Lairenjam, Benaki and Singh, Yengkhom Satyendra
- Subjects
COMPUTER-aided diagnosis ,BREAST cancer ,BREAST ,ARTIFICIAL neural networks ,TUMOR classification ,DIAGNOSIS ,BAYES' theorem - Abstract
Breast cancer is a serious disease that causes death in women if not diagnose early. Early diagnosis of the disease is considered as the best way to save life. Automatic detection of the disease from mammogram image using CAD (computer aided diagnosis) assist radiologist in predicting the disease accurately. In this paper a new model NNC ( Neural Network Classifier) based on artificial neural network is developed for automatically detecting Breast cancer from mammography data. The model NNC is a hybrid of CMAR (Classification based on multiple association rule), Bayes theorem with NN (Neural network) for classifying breast cancer mammographic data as benign or malignant. The model consists of three layers: an input layer, a hidden layer and an output layer. CMAR is used in the initial step for creating the structure of the network and Bayes' Theorem is used for calculating initial weights for hidden layer. It is tested on Wisconsin breast cancer data from UCI repository. Experimental results show that NNC model gives better convergence rate as well as classification accuracy in breast data to both benign and malignant. [ABSTRACT FROM AUTHOR]
- Published
- 2021