Derivation and validation of a machine learning risk score using biomarker and electronic patient data to predict progression of diabetic kidney disease.
Derivation and validation of a machine learning risk score using biomarker and electronic patient data to predict progression of diabetic kidney disease.
复制标题
DOI：
10.1007/s00125-021-05444-0
复制
发表时间：
2021-07
期刊：
影响因子：
8.2
通讯作者：
Damrauer SM
中科院分区：
文献类型：
--
作者：
Chan L;Nadkarni GN;Fleming F;McCullough JR;Connolly P;Mosoyan G;El Salem F;Kattan MW;Vassalotti JA;Murphy B;Donovan MJ;Coca SG;Damrauer SM
预测糖尿病肾病（DKD）的进展对于改善结局至关重要。我们试图开发/验证一种结合电子健康记录（EHR）和生物标志物的机器学习预后风险评分（KidneyIntelX™）。这是一项在来自两个EHR相关生物库的流行DKD/库存血浆患者中进行的观察性队列研究。训练随机森林模型，并将性能（AUC、阳性和阴性预测值[PPV/NPV]和净重新分类指数[NRI]）与临床模型和肾脏疾病：改善全球结局（KDIGO）类别的性能进行比较，以预测5年内eGFR下降≥5 ml/min/年、≥40%持续下降或肾衰竭的复合结局。在1146例患者中，中位年龄为63岁，51%为女性，基线eGFR为54 ml min−1 [1.73 m]−2，尿白蛋白/肌酐比值（uACR）为6.9 mg/mmol，随访时间为4.3年，21%有复合终点。在推导的交叉验证中（n = 686），KidneyIntelX的AUC为0.77（95% CI 0.74，0.79）。在验证中（n = 460），AUC为0.77（95% CI 0.76，0.79）。相比之下，临床模型的AUC在推导中为0.62（95% CI 0.61，0.63），在验证中为0.61（95% CI 0.60，0.63）。使用推导截止值，KidneyIntelX将46%、37%和17%的验证队列分别分层为复合肾脏终点的低、中和高风险组。高危组肾功能进行性下降的PPV为61%，KidneyIntelX组为40%，KDIGO分类的最高风险分层为40%（p < 0.001）。只有10%的被KidneyIntelX评分为低风险的患者经历了进展（即，净现值90%）。高危组的NRI事件为41%（p < 0.05）。KidneyIntelX改善了对早期DKD个体的KDIGO和临床模型的肾脏结局预测。在线版本包含同行评审但未经编辑的补充材料，可通过10.1007/s 00125 -021-05444-0获得。
Predicting progression in diabetic kidney disease (DKD) is critical to improving outcomes. We sought to develop/validate a machine-learned, prognostic risk score (KidneyIntelX™) combining electronic health records (EHR) and biomarkers. This is an observational cohort study of patients with prevalent DKD/banked plasma from two EHR-linked biobanks. A random forest model was trained, and performance (AUC, positive and negative predictive values [PPV/NPV], and net reclassification index [NRI]) was compared with that of a clinical model and Kidney Disease: Improving Global Outcomes (KDIGO) categories for predicting a composite outcome of eGFR decline of ≥5 ml/min per year, ≥40% sustained decline, or kidney failure within 5 years. In 1146 patients, the median age was 63 years, 51% were female, the baseline eGFR was 54 ml min−1 [1.73 m]−2, the urine albumin to creatinine ratio (uACR) was 6.9 mg/mmol, follow-up was 4.3 years and 21% had the composite endpoint. On cross-validation in derivation (n = 686), KidneyIntelX had an AUC of 0.77 (95% CI 0.74, 0.79). In validation (n = 460), the AUC was 0.77 (95% CI 0.76, 0.79). By comparison, the AUC for the clinical model was 0.62 (95% CI 0.61, 0.63) in derivation and 0.61 (95% CI 0.60, 0.63) in validation. Using derivation cut-offs, KidneyIntelX stratified 46%, 37% and 17% of the validation cohort into low-, intermediate- and high-risk groups for the composite kidney endpoint, respectively. The PPV for progressive decline in kidney function in the high-risk group was 61% for KidneyIntelX vs 40% for the highest risk strata by KDIGO categorisation (p < 0.001). Only 10% of those scored as low risk by KidneyIntelX experienced progression (i.e., NPV of 90%). The NRIevent for the high-risk group was 41% (p < 0.05). KidneyIntelX improved prediction of kidney outcomes over KDIGO and clinical models in individuals with early stages of DKD. The online version contains peer-reviewed but unedited supplementary material available at 10.1007/s00125-021-05444-0.
登录
查看更多内容
影响因子：
2
作者：
Pencina, Michael J.;D'Agostino, Ralph B., Sr.;Steyerberg, Ewout W.
通讯作者：
Steyerberg, Ewout W.
影响因子：
13.6
作者：
Niewczas, Monika A.;Gohda, Tomohito;Krolewski, Andrzej S.
通讯作者：
Krolewski, Andrzej S.
影响因子：
120.7
作者：
Damrauer, Scott M.;Chaudhary, Kumardeep;Do, Ron
通讯作者：
Do, Ron
影响因子：
5.5
作者：
Levin, Adeera;Stevens, Paul E.
通讯作者：
Stevens, Paul E.
影响因子：
120.7
作者：
Nelson, Robert G.;Grams, Morgan E.;Woodward, Mark
通讯作者：
Woodward, Mark