目的:开发一种算法,根据我们机构的电子病历中包含的术前变量来预测术中红细胞(RBC)输血,以指导类型和屏幕订购为目标。
方法:基于回顾性单中心医院数据的机器学习模型开发。
方法:术前和手术室。
方法:该研究包括2019-2022年期间接受手术的≥18岁患者,并排除拒绝输血的患者。接受了紧急手术,或心脏或脑死亡后器官捐献手术。
方法:术中输血与输血的预测术中没有输血。
方法:结果变量为术中输血红细胞。预测变量是手术,外科医生,麻醉师,年龄,性别,身体质量指数,种族或民族,术前血红蛋白(g/dL),部分凝血活酶时间(s),血小板计数×109每升,和凝血酶原时间。我们比较了七种机器学习算法的性能。在2019-2021年数据集上进行培训和优化后,模型阈值设置为当前的机构绩效敏感性水平(93%)。要有资格进行比较,在预测2022年数据时,模型必须保持临床相关的敏感性(>90%);总体准确性是比较指标.
结果:在2019年至2021年符合研究标准的100,813例中,有5488例(5.4%)发生了术中输血。LightGBM模型是外部时间有效性实验中性能最高的算法,总体准确率为(76.1%)[95%置信区间(CI),75.6-76.5],同时保持临床相关敏感性(91.2%)[95%CI,89.8-92.5]。如果类型和屏幕是根据LightGBM模型订购的,预测类型和筛查输血比将从8.4提高到5.1。
结论:机器学习方法在根据术前变量预测术中输血方面是可行的,并且当被纳入电子健康记录时,可以改善术前类型和屏幕顺序操作。
To develop an algorithm to predict intraoperative Red Blood Cell (RBC) transfusion from preoperative variables contained in the electronic medical record of our institution, with the goal of guiding type and screen ordering.
Machine Learning model development on retrospective single-center hospital data.
Preoperative period and operating room.
The study included patients ≥18 years old who underwent surgery during 2019-2022 and excluded those who refused transfusion, underwent emergency surgery, or surgery for organ donation after cardiac or brain death.
Prediction of intraoperative transfusion vs. no intraoperative transfusion.
The outcome variable was intraoperative transfusion of RBCs. Predictive variables were surgery, surgeon, anesthesiologist, age, sex, body mass index, race or ethnicity, preoperative hemoglobin (g/dL), partial thromboplastin time (s), platelet count x 109 per liter, and prothrombin time. We compared the performances of seven machine learning algorithms. After training and optimization on the 2019-2021 dataset, model thresholds were set to the current institutional performance level of sensitivity (93%). To qualify for comparison, models had to maintain clinically relevant sensitivity (>90%) when predicting on 2022 data; overall accuracy was the comparative metric.
Out of 100,813 cases that met study criteria from 2019 to 2021, intraoperative transfusion occurred in 5488 (5.4%) of cases. The LightGBM model was the highest performing algorithm in external temporal validity experiments, with overall accuracy of (76.1%) [95% confidence interval (CI), 75.6-76.5], while maintaining clinically relevant sensitivity of (91.2%) [95% CI, 89.8-92.5]. If type and screens were ordered based upon the LightGBM model, the predicted type and screen to transfusion ratio would improve from 8.4 to 5.1.
Machine learning approaches are feasible in predicting intraoperative transfusion from preoperative variables and may improve preoperative type and screen ordering practices when incorporated into the electronic health record.