人工神经网络在HIV/AIDS患者主要虚证诊断中的应用

目的:探讨人工神经网络技术在HIV/AIDS患者主要中医虚证诊断中的应用。方法:利用Clementine中的特征选择节点筛选142例脾气虚弱及肺脾气虚证HIV/AIDS患者的主要实验室指标,四诊信息中的主要症状和舌象进行网络训练,建立脾气虚弱和肺脾气虚的人工神经网络模型,将样本按3:1的比例分为训练集和测试集,训练集用于建立模型,测试集用于验证模型的正确率。结果:筛选出了发热、咳嗽咳痰、神疲乏力等10项对模型构建重要的指标,模型训练集诊断的正确率为87.25%,测试集诊断的正确率为80.00%。结论:人工神经网络模型能较好地诊断艾滋病患者脾气虚弱和肺脾气虚证型,在中医证型的诊断方面具有一定的应用潜力。

Application of artificial neural network technology to diagnose deficiency syndromes of HIV/AIDS patients

Objective: To study the application of artificial neural network(ANN) to diagnose deficiency syndromes of HIV/AIDS patients.Methods: The churn of Clementine was used to screen the major lab indicators,main symptoms and tongue manifestation of 142 HIV/AIDS patients.ANN model for deficiency of the spleen and lung and spleen qi deficiency was developed,the sample was divided into training set and testing set with the ratio of 3:1,the training set was used to build the model and the testing set was used to test the accurate rate of model.Results: Altogether,10 invariables were screened out,including fever,cough and expectoration,spiritlessness and weakness.The correct rate of model was 87.25% in the training set and 80.00% in the testing set.Conclusion: ANN model can better diagnose syndrome of the spleen and lung deficiency and spleen qi deficiency of HIV/AIDS.It has great potential application in the diagnosis of syndromes of HIV/AIDS patients.