Research on Classification Prediction for Prostate Neoplasm Patients Based on Machine Learning
In view of the clinical difficulties of not being able to screen prostate patients efficiently and carry out classification in real time,four Machine Learning models based on BP Neural Network,Random Forest(RF)Algorithm,Radial Basis Function(RBF),and Convolutional Neural Network(CNN)are constructed to identify different types of prostate patients quickly.The models are continuously optimized using parameters and Cross-Validation,and the performance of the models is evaluated using four indicators of accuracy,precision,recall,and the harmonic mean of the two.The accuracy of the BP Neural Network,RF Algorithm,RBF and CNN is 0.930,0.965,0.877 and 0.982,respectively,indicating that the four methods can all perform classification prediction of prostate patients well.Among them,CNN has the best classification prediction effect and can provide a reference for the early clinical screening of prostate cancer.
hyperplasia of prostateprostate adenocarcinomaMachine Learningclassification predictionConfusion Matrix
国家科技期刊平台
NSTL
万方数据
