Preoperative prediction of malignancy of ovarian tumors using least squares support vector machines

摘要: In this work, we develop and evaluate several least squares support vector machine (LS-SVM) classifiers within the Bayesian evidence framework, in order to preoperatively predict malignancy of ovarian tumors. The analysis includes exploratory data analysis, optimal input variable selection, parameter estimation, performance evaluation via receiver operating characteristic (ROC) curve analysis. LS-SVM models with linear radial basis function (RBF) kernels, logistic regression have been built on 265 training data, tested 160 newly collected patient data. model nonlinear RBF kernel achieves best performance, test set area under ROC (AUC), sensitivity specificity equal 0.92, 81.5% 84.0%, respectively. averaged over 30 runs randomized cross-validation is also obtained by an model, AUC, 0.94, 90.0% 80.6%, These results show that potential obtain a reliable preoperative distinction between benign malignant tumors, assist clinicians for making correct diagnosis.