一种改进的SVM算法在乳腺癌诊断方面的应用

发布时间：2018-09-19 12:26

【摘要】：针对计算机辅助诊断(CAD)技术在乳腺癌疾病诊断准确率的优化问题,提出了一种基于随机森林模型下Gini指标特征加权的支持向量机方法(RFG-SVM)。该方法利用了随机森林模型下的Gini指数衡量各个特征对分类结果的重要性,构造具有加权特征向量核函数的支持向量机,并在乳腺癌疾病诊断方面加以应用。经理论分析和实验数据验证,相比于传统的支持向量机(SVM),该方法提升了分类预测的性能,其结果与最新的方法相比也具有一定的竞争力,而且在医疗诊断应用方面更具优势。
[Abstract]:In order to optimize the diagnostic accuracy of computer-aided diagnosis (CAD) in breast cancer, a new support vector machine (RFG-SVM) method based on Gini index weighted under stochastic forest model is proposed. This method uses the Gini index under the stochastic forest model to measure the importance of each feature to the classification result, constructs the support vector machine with weighted eigenvector kernel function, and applies it to the diagnosis of breast cancer disease. Through theoretical analysis and experimental data verification, compared with the traditional support vector machine (SVM), this method improves the performance of classification and prediction, and the results are competitive compared with the latest methods, and it has more advantages in medical diagnosis application.
【作者单位】：兰州交通大学电子与信息工程学院;
【基金】：国家自然科学基金(61163010) 甘肃省自然科学基金(1308RJZA111) 兰州市科技计划(2015-2-99)
【分类号】：R737.9;TP391.7

【参考文献】