支持向量机在高血压病中医证候诊断中的应用

Application of support vector machine in the diagnosis of hypertension in TCM syndrome

目的:探索支持向量机方法在高血压病中医证候诊断中的应用,为高血压病中医证候诊断合理应用提供理论依据。方法:基于MATLAB 7.0环境,采用支持向量机(SVM)学习算法,以常见的21个症状、舌苔及舌体、脉象的量化数据为输入,高血压病证型为输出,建立基于SVM的高血压病中医证候诊断模型,并使用419例样本进行训练,130例样本进行测试。结果:总体准确率为90.0%。结论:基于SVM建立高血压病中医证候诊断模型具有较高准确率和方法学上的可行性。

Objective: To explore the application of support vector machine method in the diagnosis of hypertension in TCM syndrome, providing theoretical basis for the reasonable application of TCM syndrome diagnosis of hypertension. Methods: Based on the MATLAB 7.0 environment, support vector machine(SVM) learning algorithm was used, with 21 common symptoms, tongue coating, tongue body and quantitative data of pulse as input as well as the hypertension syndromes as output, to establish the hypertension TCM syndrome diagnosis model which was based on SVM, using 419 samples for training and 130 for testing. Results: The overall accuracy rate was 90.0%. Conclusion: Establishing the hypertension TCM syndrome diagnosis model based on SVM has a high accuracy and the methodological feasibility.