Diagnosis of Breast Cancer Using Neural Network Approach

Abstract

Background: Breast cancer is one of the fatal disorders causing death in people and second most common reason of mortality (death rate) in women. Early detection of cancer, which usually results in reducing the extent of damage, less extensive treatment and better outcomes. In this paper, we present an artificial neural network inspired from biological neural network which is used for pattern classification of benign and malignant cells Methods: A feed forward neural network model with back propagation learning algorithm for training the neural network using breast cancer database is simulated with all the variable network constraints to make it efficient, robust and fault tolerated pattern classifier. Results: The viability of this approach is demonstrated for classification with predictive success of 96.34% with 99.41% sensitivity to malignancy. Hence, it will probably enhance the decision on classifying the malignant cells. Conclusions: Despite the fact, not all general hospitals have the mammogram facilities the neural network prediction model can increase the rate of diagnosis for breast cancer. This scheme can be used as an auxiliary tool to differentiate between benign and malignant breast cancers.