摘要
目的应用人工神经网络系统,建立基于人工神经网络的涂阴肺结核诊断模型并研究其诊断性能。方法将全部研究对象随机分为训练样本和测试样本,利用训练样本筛选出对诊断涂阴肺结核有意义的单项参数指标,构建人工神经网络诊断模型,用ROC曲线下面积、灵敏度及特异度评价其诊断性能。结果得到人工神经网络模型结构为(29-9-1)-BP型。模型在训练样本和测试样本ROC曲线下面积分别为0.984±0.004和0.955±0.018,诊断的正确率、灵敏度及特异度分别为93.30%和91.07%,90.78%和90.91%,95.45%和91.30%。结论(29-9-1)-BP型网络模型可作为涂阴肺结核的诊断工具,有良好的诊断性能,值得进一步探讨。
Objective To establish a model based on artificial neural network for diagnosis of smear negative pulmonary tuberculosis and evaluating its performance. Methods All original data was randomly divided into training sample and testing sample. training samples were used to screen out significant single meter and develop the diagnostic model of smear negative pulmonary tuberculosis based on artificial neural network. The area under the receiver operator characteristic curve, sensitivity and specificity were used to evaluate the model＇s performance in both training and testing samples. Results The structure of artificial neural network is （29 - 9 - 1） - BP. When the model was applied to the training sample and the testing sample, the area under the receiver operator characteristic curve, accuracy, sensitivity and specificity were 0.984 ±0.004 and 0.955 ± 0.018, 91.07% and 93.10%, 90.91% and 88.89% ,91.30% and 100% respectively. Condusion Artificial neural network model is capable of diagnosing smear negative pulmonary tubereulosis and it can be used as a tool in routine diagnostic practice.
出处
《中国热带医学》
CAS
2007年第5期661-663,678,共4页
China Tropical Medicine
基金
山东省卫生厅立项课题(2005HW012)
关键词
涂阴
结核
肺
人工神经网络
诊断
Smear negative
Tuberculosis
Lung
Artificial neural network
Diagnosis
