Computer-Aided Diagnosis of Breast Cancer Based on Probabilistic Neural Networks
Jianghong Chen · Jisuanji fangzhen · 2012
In order to overcome the limitation of Fine Needle Aspiration Cytology,and to further enhance the accuracy of breast cancer auxiliary diagnosis,this paper proposed a computer-aided diagnosis method based on Probabilistic Neural Networks.First,a classifying model based on probabilistic neural networks was established.Second,the samples were distributed to training set and test set.Third,after finding out the optimal radial basis density function of probabilistic neural networks,the accuracy of 5-fold cross validation wass calculated.Finally,the method was checked and analysze with the results of simulation,and compared with other mothods.The results show that the probabilistic neural network has higher speed and better accuracy.