Determinants of In-Hospital Mortality After Percutaneous Coronary Intervention: A Machine Learning Approach
Determinants of In-Hospital Mortality After Percutaneous Coronary Intervention: A Machine Learning Approach
复制标题
DOI：
10.1161/jaha.118.011160
复制
发表时间：
2019-03-05
影响因子：
5.4
通讯作者：
Minutello, Robert M.
中科院分区：
文献类型：
--
作者：
Al'Aref, Subhi J.;Singh, Gurpreet;Minutello, Robert M.
准确预测经皮冠状动脉介入治疗后院内死亡发生率的能力对于临床决策非常重要。我们试图利用纽约经皮冠状动脉介入治疗报告系统，以阐明纽约州接受经皮冠状动脉介入治疗的患者住院死亡率的决定因素。方法和结果-我们检查了2004年至2012年期间接受经皮冠状动脉介入治疗的479804例患者，利用传统和先进的机器学习算法来确定住院死亡率的最重要预测因素。整个数据被随机分为训练集（80%）和测试集（20%）。使用调整的超参数生成训练模型，同时在绘制接受者-操作者特征曲线并使用曲线下面积（AUC）的输出测量值和相关95%CI后，在测试集上独立评价模型的性能。平均年龄为65.2 ± 11.9岁，68.5%为女性。患者人群中有2549例院内死亡。增强集成算法（AdaBoost）具有最佳区分度，AUC为0.927（95%CI 0.923-0.929），而XGBoost的AUC为0.913（95%CI 0.906-0.919，P=0.02），随机森林的AUC为0.892（95%CI 0.889-0.896，P = 0.02）。
Background-The ability to accurately predict the occurrence of in-hospital death after percutaneous coronary intervention is important for clinical decision-making. We sought to utilize the New York Percutaneous Coronary Intervention Reporting System in order to elucidate the determinants of in-hospital mortality in patients undergoing percutaneous coronary intervention across New York State.Methods and Results-We examined 479 804 patients undergoing percutaneous coronary intervention between 2004 and 2012, utilizing traditional and advanced machine learning algorithms to determine the most significant predictors of in-hospital mortality. The entire data were randomly split into a training (80%) and a testing set (20%). Tuned hyperparameters were used to generate a trained model while the performance of the model was independently evaluated on the testing set after plotting a receiver-operator characteristic curve and using the output measure of the area under the curve (AUC) and the associated 95% CIs. Mean age was 65.2 +/- 11.9 years and 68.5% were women. There were 2549 in-hospital deaths within the patient population. A boosted ensemble algorithm (AdaBoost) had optimal discrimination with AUC of 0.927 (95% CI 0.923-0.929) compared with AUC of 0.913 for XGBoost (95% CI 0.906-0.919, P=0.02), AUC of 0.892 for Random Forest (95% CI 0.889-0.896, P