基于Levenberg-Marquardt算法的胃脘痛BP神经网络辨证模型研究

Research on BP Neural Network Syndrome Differentiation Model of Stomach Pain Based on LM Algorithm

目的:本文提出了一种基于Levenberg-Marquardt算法的用于胃脘痛的BP神经网络辨证模型,用于提高胃脘痛智能辨证的准确率。方法:以"中医数字化诊疗平台"的门诊临床电子病历数据作为数据集,采用Matlab作为模型仿真平台,运用Levenberg-Marquardt算法构建了胃脘痛中医智能辨证的双隐含层BP神经网络模型。结果:实验结果显示,网络模型预测"肝胃不和"和"胃阳虚"的证型准确率和诊断准确率非常高,都在95%以上。结论:该智能辨证模型能有效利用BP神经网络的自主学习能力,充分逼近中医辨证的真实面貌,表现出优秀的辨证预测能力。而且,每天在"中医数字化诊疗平台"中都有新的中医临床真实数据上传,若利用这些数据完善该智能辨证模型,有望推动中医智能辨证在中医临床辅助诊断中大规模应用。

Objective： A BP neural network syndrome differentiation model of stomach pain based on Levenberg-Marquardt （LM） algorithm is put forward in this paper, and it is used to improve the accuracy of intelligent syndrome differentiation in stomach pain. Methods ： The clinical electronic medical record data of ＇ digital diagnosis and treatment platform of TCM＇ is used as a data set, and Matlab is used as the model simulation platform, and the LM algorithm is used to build the double hidden layer BP neural network model of TCM intelligent syndrome differentiation for stomach pain. Results： The testing results shows that the syndrome accuracy and diagnosis accuracy of ＇ liver-stomach disharmony＇ and ＇ stomach yang deficiency＇ are very high, above 95%, by using network pre- dictive model. Conclusion ： The intelligent syndrome differentiation model of TCM can fully approach the real side of syndrome differentiation by effectively using the autonomous learning ability of BP neural network, and shows excellent predicted ability of syndrome differentiation. Furthermore, new reliable clinical data of TCM are uploaded every day in the ＇ digital diagnosis and treatment platform of TCM＇ . If these data are used to improve the intelligent syndrome differentiation model, it is expected to promote the large-scale ap- plication of TCM intelligent syndrome differentiation in clinical auxiliary diagnosis of TCM.