@Article{info:doi/10.2196/26256,作者=“Yeh, Marvin Chia-Han and Wang, yu - xiangang and Yang, Hsuan-Chia and Bai, Kuan-Jen and Wang, Hsiao-Han and Li, Yu-Chuan Jack”,标题=“基于人工智能的肺癌风险预测:基于非成像电子病历的深度学习方法”,期刊=“J Med Internet Res”,年=“2021”,月=“8”,日=“3”,卷=“23”,数=“8”,页=“e26256”,关键词=“人工智能”;肺癌筛查;背景:人工智能方法可以整合复杂的特征,并可用于预测患者患肺癌的风险,从而减少不必要和昂贵的诊断干预的需要。目的:本研究的目的是利用电子病历对有患肺癌风险的患者进行预筛查。方法:从1999 - 2013年台湾全民健康保险研究数据库中随机抽取200万名参保者。我们用神经网络建立了一个预测肺癌筛查模型,该模型使用2012年前的数据进行训练和验证,并对2012年后的数据进行了前瞻性测试。一个年龄和性别匹配的亚组比原来的肺癌组大10倍,用来评估电子病历的预测能力。鉴别(接收机工作特征曲线下面积[AUC])和校准分析。结果:纳入11617例肺癌患者和1423154例对照患者。该模型在总体人群中的auc为0.90,在≥55岁的患者中auc为0.87。 The AUC in the matched subgroup was 0.82. The positive predictive value was highest (14.3{\%}) among people aged ≥55 years with a pre-existing history of lung disease. Conclusions: Our model achieved excellent performance in predicting lung cancer within 1 year and has potential to be deployed for digital patient screening. Convolution neural networks facilitate the effective use of EMRs to identify individuals at high risk for developing lung cancer. ", issn="1438-8871", doi="10.2196/26256", url="//www.mybigtv.com/2021/8/e26256", url="https://doi.org/10.2196/26256", url="http://www.ncbi.nlm.nih.gov/pubmed/34342588" }
Baidu
map