Early and precise detection of breast cancer with artificial neural network

Breast cancer is the most commonly diagnosed cancer with high mortality and morbidity in Iranian women. Early detection of breast cancer is essential in improving survival, therefore in this study artificial neural network was used for early diagnosis of breast cancer. Artificial Neural Networks (ANNs) are non-linear mapping structures based on the function of the human brain. ANNs can identify and learn correlated patterns between input data sets and corresponding target values. 25 clinical and medicinal parameters were taken for 222 persons from Besat hospital of Hamadan. Chi square test was carried out with SPSS software for 25 parameters. With regard to the results of this analysis we selected 7 parameters that had the lowest sig for ANN analysis (among parameters, whose sig were less than 0.01). Selected parameters of 222 persons were applied for training network with Levenberg-Marquardt Learning Algorithm; Learning rate was 0.1. The training process finished at around 22 epochs; Best validation performance was . corelation between target value and output of trained neural network for training, validation & test dataset statistically was plotted; R value for all of them was 1. Recently ANNs have become popular in medical diagnosis. ANNs are particularly attractive for diagnostic problems with nonlinear solution . In summery early detection of breast cancer based on ANN will be reliable, accurate and noninvasive. Because of high performance and accuracy, thisnetwork was used for quick and reliable detection of breast cancer.

Keywords:

breast , cancer , diagnosis , Artificial Neural Network