于李.神经网络模型预测老年HBV相关原发性肝癌患者预后研究[J].转化医学杂志,2020,9(5):296-298
神经网络模型预测老年HBV相关原发性肝癌患者预后研究
A prediction model for outcome of elder patients with Primary Liver Cancer associated with hepatitis B virus based on artificial neural network
  
DOI:
中文关键词:  人工神经网络  老年  HBV相关原发性肝癌  中医证候  预后预测模型
英文关键词:Artificial neural network (ANN)  Elder  Primary Liver Cancer (PLC) associated with hepatitis B virus  TCM (Traditional Chinese Medicine) syndrome  Pprognosis prediction model
基金项目:
作者单位
于李 北京中医药大学第三附属医院特需老年病科 
摘要点击次数: 62
全文下载次数: 59
中文摘要:
      目的探讨人工神经网络(artificial neural network,ANN)智能预测方法结合中医证候特征预测老年乙型肝炎病毒(hepatitis B virus,HBV)相关原发性肝癌患者临床预后。方法以408例老年HBV相关原发性肝癌患者为研究对象,收集数据资料并进行中医证候辨证。以KNIME数据分析平台、数据特征提取算法进行筛选确认,以单时间点神经网络模型预测患者每年时间点(1~5年)预后。根据ROC曲线下面积及其精准度,评估神经网络模型在预测死亡概率方面与患者实际死亡率在各预测时间点的一致性。结果本研究获得该模型的敏感性和特异性分别为86.6%和92.7%。神经网络模型在预测患者死亡概率方面与患者实际死亡率在所有时间点高度一致。结论神经网络模型可提供简便快捷、准确的方法结合中医辨证判定老年HBV相关原发性肝癌患者的预后。
英文摘要:
      ObjectiveThis study aims to explore an outcome prediction model for elder patients with Primary Liver Cancer (PLC) associated with hepatitis B virus according to artificial intelligence, artificial neural network (ANN) and the classification of TCM (Traditional Chinese Medicine) syndromes. MethodsDemographic variables, and clinical data of 408 elderly patients with PLC associated with hepatitis B virus were collected. The Tranditional Chinese Medicine (TCM) syndromes of all patients were identified. Relevant variables were analyzed with KNIME data analysis platform utilizing data feature extraction algorithm for selection certification. The single time-point ANN models was developed to predict the outcome of patients after 1~5 years. According to the area under the ROC curve and its accuracy, the consistency of the neural network model in the predicting the mortality of death with the actual mortality of patients at each prediction time point was assessed. ResultsThe sensitivity and specificity of the model were 86.6% and 92.7%, respectively. The results of artificial neural network model based on TCM syndrome showed a high degree of consistency with the actual situation at all time points in predicting the probability of death in these elder patients with PLC. ConclusionOur results indicated that the artificial neural network model for outcome prediction of the elder patients with PLC associated with hepatitis B virus maybe provide a simple, quick, and accurate method for determining the prognosis of elder patients with PLC based on TCM syndrome differentiation.
查看全文  查看/发表评论  下载PDF阅读器
关闭