Bridging a translational gap: using machine learning to improve the prediction of PTSD.
Bridging a translational gap: using machine learning to improve the prediction of PTSD.
复制标题
DOI：
10.1186/s12888-015-0399-8
复制
发表时间：
2015-03-16
期刊：
影响因子：
4.4
通讯作者：
members of Jerusalem Trauma Outreach and Prevention Study (J-TOPS) group
中科院分区：
文献类型：
--
作者：
Karstoft KI;Galatzer-Levy IR;Statnikov A;Li Z;Shalev AY;members of Jerusalem Trauma Outreach and Prevention Study (J-TOPS) group
预测创伤后应激障碍（PTSD）是有针对性的预防的先决条件。目前的研究已经确定了群体层面的风险指标，其中许多（例如，头部创伤，接受阿片类药物）的关注，但幸存者的一个子集。确定可互换的风险指标可提高早期风险评估的效率。该研究的目标是使用有监督的机器学习（ML）来发现早期风险指标的可互换的、最大限度的预测组合。收集了957名创伤幸存者在艾德入院后10天内的事件特征、急诊科（艾德）记录和早期症状的数据变量（特征），并用于预测随后15个月内的PTSD症状轨迹。目标信息等价算法（TIE*）识别出所有最小特征集（马尔可夫边界; MB），这些特征集在集成到支持向量机（SVM）中时最大化非缓解性PTSD症状轨迹的预测。在重复的10倍交叉验证中评价每组预测因子的预测准确度，并表示为所有验证试验的受试者操作特征曲线下面积（AUC）的平均值。每次交叉验证的MB平均数为800。MB的平均AUC为0.75（95%范围：0.67-0.80）。每MB的平均特征数为18（范围：12-32），超过75%的集合中存在13个特征。我们的研究结果支持假设存在多个和可互换的风险指标集，同样和详尽地预测非缓解性PTSD。ML增加预测多功能性的能力是朝着开发创伤后精神病理学的算法，基于知识的个性化预测迈出的有希望的一步。本文的在线版本（doi：10.1186/s12888-015-0399-8）包含补充材料，可供授权用户使用。
Predicting Posttraumatic Stress Disorder (PTSD) is a pre-requisite for targeted prevention. Current research has identified group-level risk-indicators, many of which (e.g., head trauma, receiving opiates) concern but a subset of survivors. Identifying interchangeable sets of risk indicators may increase the efficiency of early risk assessment. The study goal is to use supervised machine learning (ML) to uncover interchangeable, maximally predictive combinations of early risk indicators. Data variables (features) reflecting event characteristics, emergency department (ED) records and early symptoms were collected in 957 trauma survivors within ten days of ED admission, and used to predict PTSD symptom trajectories during the following fifteen months. A Target Information Equivalence Algorithm (TIE*) identified all minimal sets of features (Markov Boundaries; MBs) that maximized the prediction of a non-remitting PTSD symptom trajectory when integrated in a support vector machine (SVM). The predictive accuracy of each set of predictors was evaluated in a repeated 10-fold cross-validation and expressed as average area under the Receiver Operating Characteristics curve (AUC) for all validation trials. The average number of MBs per cross validation was 800. MBs’ mean AUC was 0.75 (95% range: 0.67-0.80). The average number of features per MB was 18 (range: 12–32) with 13 features present in over 75% of the sets. Our findings support the hypothesized existence of multiple and interchangeable sets of risk indicators that equally and exhaustively predict non-remitting PTSD. ML’s ability to increase prediction versatility is a promising step towards developing algorithmic, knowledge-based, personalized prediction of post-traumatic psychopathology. The online version of this article (doi:10.1186/s12888-015-0399-8) contains supplementary material, which is available to authorized users.
登录
查看更多内容
影响因子：
--
作者：
KESSLER, RC;SONNEGA, A;NELSON, CB
通讯作者：
NELSON, CB
影响因子：
120.7
作者：
Binder, Elisabeth B.;Bradley, Rebekah G.;Ressler, Kerry J.
通讯作者：
Ressler, Kerry J.
影响因子：
3.2
作者：
Boscarino JA;Erlich PM;Hoffman SN;Zhang X
通讯作者：
Zhang X
影响因子：
8
作者：
Bradley, AP
通讯作者：
Bradley, AP
影响因子：
4.1
作者：
Forbes, D;Creamer, M;Biddle, D
通讯作者：
Biddle, D