Harvard Institute for Applied Computational Science, Harvard University, Cambridge, MA.
Computational Health Informatics Program, Boston Children's Hospital, Boston, MA.
Pediatr Crit Care Med. 2021 Apr 1;22(4):392-400. doi: 10.1097/PCC.0000000000002626.
To create a machine-learning model identifying potentially avoidable blood draws for serum potassium among pediatric patients following cardiac surgery.
Retrospective cohort study.
Tertiary-care center.
All patients admitted to the cardiac ICU at Boston Children's Hospital between January 2010 and December 2018 with a length of stay greater than or equal to 4 days and greater than or equal to two recorded serum potassium measurements.
None.
We collected variables related to potassium homeostasis, including serum chemistry, hourly potassium intake, diuretics, and urine output. Using established machine-learning techniques, including random forest classifiers, and hyperparameter tuning, we created models predicting whether a patient's potassium would be normal or abnormal based on the most recent potassium level, medications administered, urine output, and markers of renal function. We developed multiple models based on different age-categories and temporal proximity of the most recent potassium measurement. We assessed the predictive performance of the models using an independent test set. Of the 7,269 admissions (6,196 patients) included, serum potassium was measured on average of 1 (interquartile range, 0-1) time per day. Approximately 96% of patients received at least one dose of IV diuretic and 83% received a form of potassium supplementation. Our models predicted a normal potassium value with a median positive predictive value of 0.900. A median percentage of 2.1% measurements (mean 2.5%; interquartile range, 1.3-3.7%) was incorrectly predicted as normal when they were abnormal. A median percentage of 0.0% (interquartile range, 0.0-0.4%) critically low or high measurements was incorrectly predicted as normal. A median of 27.2% (interquartile range, 7.8-32.4%) of samples was correctly predicted to be normal and could have been potentially avoided.
Machine-learning methods can be used to predict avoidable blood tests accurately for serum potassium in critically ill pediatric patients. A median of 27.2% of samples could have been saved, with decreased costs and risk of infection or anemia.
创建一种机器学习模型，用于识别小儿心脏手术后患者中可能避免的血清钾抽取。
回顾性队列研究。
三级医疗中心。
2010 年 1 月至 2018 年 12 月期间在波士顿儿童医院心脏重症监护病房（ICU）住院，住院时间大于或等于 4 天，且至少记录两次血清钾值的所有患者。
无。
我们收集了与钾稳态相关的变量，包括血清化学、每小时钾摄入量、利尿剂和尿量。使用已建立的机器学习技术，包括随机森林分类器和超参数调整，我们创建了基于最近一次钾水平、给药的药物、尿量和肾功能标志物，预测患者的钾是正常还是异常的模型。我们根据最近一次钾测量的不同年龄类别和时间接近程度，开发了多个模型。我们使用独立的测试集评估模型的预测性能。在纳入的 7269 例（6196 例患者）住院患者中，平均每天测量血清钾 1 次（中位数，0-1 次）。大约 96%的患者至少接受了一次静脉利尿剂治疗，83%的患者接受了某种形式的钾补充。我们的模型预测正常钾值的中位阳性预测值为 0.900。当异常时，有 2.1%（中位数 2.5%；中位数，1.3-3.7%）的测量值被错误地预测为正常。有 0.0%（中位数，0.0-0.4%）的临界低值或高值测量值被错误地预测为正常。中位数为 27.2%（中位数，7.8-32.4%）的样本被正确预测为正常，可能可以避免。
机器学习方法可用于准确预测危重症儿科患者的血清钾是否可避免的采血。中位数为 27.2%的样本可以被节省，同时降低成本和感染或贫血的风险。
