Robust clinical marker identification for diabetic kidney disease with ensemble feature selection
Robust clinical marker identification for diabetic kidney disease with ensemble feature selection
复制标题
通过整体特征选择对糖尿病肾病进行稳健的临床标志物识别
DOI：
10.1093/jamia/ocy165
复制
发表时间：
2019-03-01
影响因子：
6.4
通讯作者：
Liu, Mei
中科院分区：
文献类型：
--
作者：
Song, Xing;Waitman, Lemuel R.;Liu, Mei
目的
糖尿病肾病（DKD）是糖尿病中最常见的并发症之一，与高发病率和死亡率相关。为了加速DKD风险因素的发现，我们提出了一个集成的特征选择方法，以确定一个强大的一组判别因素，使用电子病历（EMR）。
材料和方法
我们确定了一个包含15645例成人2型糖尿病患者的回顾性队列，排除了既往存在肾脏疾病的患者，并在建模中使用了所有可用的临床数据类型。我们比较了3种基于机器学习的嵌入式特征选择方法与6种特征集成技术，以在对数据扰动的鲁棒性和DKD发作的可预测性方面选择排名靠前的特征。
结果
采用加权平均秩特征集成技术的梯度增强机（GBM）实现了最佳性能，内部验证的AUC为0.82 [95%CI，0.81-0.83]，外部时间验证的AUC为0.71 [95%CI，0.68-0.73]。总体模型从84872个独特的临床特征中识别出一组440个特征，这些特征既可预测DKD发作，又可抵抗数据扰动，包括191个实验室、51个访视详情（主要是生命体征）、39个药物、34个医嘱、30个诊断和95个其他临床特征。
讨论
许多排名靠前的特征尚未纳入最新的DKD预测模型中，但现有文献中已提出了它们与肾功能的关系。
结论
我们的集成特征选择框架提供了一个选项，用于从EMR数据中无偏地识别一个强大而简约的特征集，这有效地帮助了DKD风险因素的知识发现。
Objective
Diabetic kidney disease (DKD) is one of the most frequent complications in diabetes associated with substantial morbidity and mortality. To accelerate DKD risk factor discovery, we present an ensemble feature selection approach to identify a robust set of discriminant factors using electronic medical records (EMRs).
Material and Methods
We identified a retrospective cohort of 15 645 adult patients with type 2 diabetes, excluding those with pre-existing kidney disease, and utilized all available clinical data types in modeling. We compared 3 machine-learning-based embedded feature selection methods in conjunction with 6 feature ensemble techniques for selecting top-ranked features in terms of robustness to data perturbations and predictability for DKD onset.
Results
The gradient boosting machine (GBM) with weighted mean rank feature ensemble technique achieved the best performance with an AUC of 0.82 [95%-CI, 0.81-0.83] on internal validation and 0.71 [95%-CI, 0.68-0.73] on external temporal validation. The ensemble model identified a set of 440 features from 84 872 unique clinical features that are both predicative of DKD onset and robust against data perturbations, including 191 labs, 51 visit details (mainly vital signs), 39 medications, 34 orders, 30 diagnoses, and 95 other clinical features.
Discussion
Many of the top-ranked features have not been included in the state-of-art DKD prediction models, but their relationships with kidney function have been suggested in existing literature.
Conclusion
Our ensemble feature selection framework provides an option for identifying a robust and parsimonious feature set unbiasedly from EMR data, which effectively aids in knowledge discovery for DKD risk factors.