Deep sequential neural network models improve stratification of suicide attempt risk among US veterans.
Deep sequential neural network models improve stratification of suicide attempt risk among US veterans.
复制标题
深度序列神经网络模型改善了美国退伍军人自杀未遂风险的分层。
DOI：
10.1093/jamia/ocad167
复制
发表时间：
2023
期刊：
影响因子：
--
通讯作者：
Beckham,JeanC
中科院分区：
文献类型：
--
作者：
Martinez,Carianne;Levin,Drew;Jones,Jessica;Finley,PatrickD;McMahon,Benjamin;Dhaubhadel,Sayera;Cohn,Judith;MillionVeteranProgram;MVPSuicideExemplarWorkgroup;Oslin,DavidW;Kimbrel,NathanA;Beckham,JeanC
目的将深度神经网络(DNN)应用于纵向EHR数据，以预测退伍军人自杀未遂风险。使用局部解释技术为每个预测提供解释，最终改善外展和干预努力。材料和方法DNN将人口统计信息与诊断、处方和程序代码融合在一起。模型基于约50万名美国退伍军人的电子健康记录数据进行了培训和测试：所有退伍军人在2005年4月1日至2016年1月1日期间有自杀未遂记录，每个退伍军人与5名同龄退伍军人配对，这些退伍军人没有试图自杀。计算Shapley附加解释(Shap)值以解释DNN的预测。结果DNN在预测自杀企图方面优于Logistic和线性回归模型。在调整抽样技术后，卷积神经网络(CNN)模型对风险最高0.1%的退伍军人在12个月内自杀企图的阳性预测值(PPV)为0.54。探讨与结论本研究所采用的深度学习方法有可能显著提升现有的退伍军人自杀风险模型。这些方法也可以为探索长期和短期干预策略的相对价值提供重要线索。此外，这里使用的解释方法也可以用来向临床医生传达增加特定退伍军人企图自杀风险的关键特征。
ObjectiveTo apply deep neural networks (DNNs) to longitudinal EHR data in order to predict suicide attempt risk among veterans. Local explainability techniques were used to provide explanations for each prediction with the goal of ultimately improving outreach and intervention efforts.Materials and methodsThe DNNs fused demographic information with diagnostic, prescription, and procedure codes. Models were trained and tested on EHR data of approximately 500 000 US veterans: all veterans with recorded suicide attempts from April 1, 2005, through January 1, 2016, each paired with 5 veterans of the same age who did not attempt suicide. Shapley Additive Explanation (SHAP) values were calculated to provide explanations of DNN predictions.ResultsThe DNNs outperformed logistic and linear regression models in predicting suicide attempts. After adjusting for the sampling technique, the convolutional neural network (CNN) model achieved a positive predictive value (PPV) of 0.54 for suicide attempts within 12 months by veterans in the top 0.1% risk tier. Explainability methods identified meaningful subgroups of high-risk veterans as well as key determinants of suicide attempt risk at both the group and individual level.Discussion and conclusionThe deep learning methods employed in the present study have the potential to significantly enhance existing suicide risk models for veterans. These methods can also provide important clues to explore the relative value of long-term and short-term intervention strategies. Furthermore, the explainability methods utilized here could also be used to communicate to clinicians the key features which increase specific veterans’ risk for attempting suicide.