Duke Clinical Research Institute, Durham, NC, USA.
Division of Cardiology, Department of Medicine, Duke University Medical Center, 2301 Erwin Rd, DUMC 3845, Durham, NC 27710, USA.
Europace. 2020 Nov 1;22(11):1635-1644. doi: 10.1093/europace/euaa172.
Prediction models for outcomes in atrial fibrillation (AF) are used to guide treatment. While regression models have been the analytic standard for prediction modelling, machine learning (ML) has been promoted as a potentially superior methodology. We compared the performance of ML and regression models in predicting outcomes in AF patients.
The Outcomes Registry for Better Informed Treatment of Atrial Fibrillation (ORBIT-AF) and Global Anticoagulant Registry in the FIELD (GARFIELD-AF) are population-based registries that include 74 792 AF patients. Models were generated from potential predictors using stepwise logistic regression (STEP), random forests (RF), gradient boosting (GB), and two neural networks (NNs). Discriminatory power was highest for death [STEP area under the curve (AUC) = 0.80 in ORBIT-AF, 0.75 in GARFIELD-AF] and lowest for stroke in all models (STEP AUC = 0.67 in ORBIT-AF, 0.66 in GARFIELD-AF). The discriminatory power of the ML models was similar or lower than the STEP models for most outcomes. The GB model had a higher AUC than STEP for death in GARFIELD-AF (0.76 vs. 0.75), but only nominally, and both performed similarly in ORBIT-AF. The multilayer NN had the lowest discriminatory power for all outcomes. The calibration of the STEP modelswere more aligned with the observed events for all outcomes. In the cross-registry models, the discriminatory power of the ML models was similar or lower than the STEP for most cases.
When developed from two large, community-based AF registries, ML techniques did not improve prediction modelling of death, major bleeding, or stroke.
房颤（AF）结局预测模型用于指导治疗。虽然回归模型一直是预测模型分析的标准，但机器学习（ML）已被推广为一种潜在的优越方法。我们比较了 ML 和回归模型在预测 AF 患者结局中的性能。
Outcomes Registry for Better Informed Treatment of Atrial Fibrillation（ORBIT-AF）和 Global Anticoagulant Registry in the FIELD（GARFIELD-AF）是基于人群的登记处，包含 74792 例 AF 患者。使用逐步逻辑回归（STEP）、随机森林（RF）、梯度提升（GB）和两个神经网络（NN）从潜在预测因子生成模型。死亡的判别能力最高[ORBIT-AF 中 STEP 曲线下面积（AUC）=0.80，GARFIELD-AF 中为 0.75]，所有模型中卒中的判别能力最低（ORBIT-AF 中 STEP AUC=0.67，GARFIELD-AF 中为 0.66）。对于大多数结局，ML 模型的判别能力与 STEP 模型相似或较低。GB 模型在 GARFIELD-AF 中的死亡 AUC 高于 STEP（0.76 对 0.75），但仅略有优势，在 ORBIT-AF 中表现相似。多层 NN 对所有结局的判别能力最低。STE P 模型的校准与所有结局的观察事件更一致。在跨登记处模型中，对于大多数病例，ML 模型的判别能力与 STEP 相似或较低。
当从两个大型、基于社区的 AF 登记处开发时，ML 技术并未提高死亡、大出血或卒中的预测模型构建。
