基于电子健康记录数据的机器学习算法预测术后并发症并在移动平台报告的表现
这项预后研究纳入 2014 年 6 月 1 日至 2020 年 9 月 20 日所有 18 岁及以上接受住院手术的患者,共 74 417 例手术、58 236 例患者;用 2014 年 6 月 1 日至 2018 年 11 月 27 日的 52 117 例回顾性数据开发模型,用 2018 年 11 月 28 日至 2020 年 9 月 20 日前瞻性收集的 22 300 例数据验证。采用广义加性模型和随机森林模型,以受试者工作特征曲线下面积(AUROC)评价预测性能。随机森林模型(135 个输入特征)预测急性肾损伤 AUROC 0.82(95%CI 0.82-0.83)、心血管并发症 0.81(95%CI 0.81-0.82)、神经系统并发症(包括谵妄)0.87(95%CI 0.87-0.88)、ICU 延长停留 0.89(95%CI 0.88-0.89)、机械通气延长 0.91(95%CI 0.90-0.91)、脓毒症 0.86(95%CI 0.85-0.87)、静脉血栓栓塞 0.82(95%CI 0.81-0.83)、伤口并发症 0.78(95%CI 0.78-0.79)、30 天死亡率 0.84(95%CI 0.82-0.86)、90 天死亡率 0.84(95%CI 0.82-0.85),准确性与外科医生预测相似。移动设备应用较原网页门户加载数据快约 10 倍,并支持指纹登录
要点
- 前瞻性验证队列含 22 300 例住院手术、19 132 例患者,随机森林模型 135 个输入特征预测术后并发症 AUROC 0.78–0.91。
- 机械通气延长 AUROC 0.91(95%CI 0.90-0.91)、ICU 延长停留 0.89(95%CI 0.88-0.89)、脓毒症 0.86(95%CI 0.85-0.87)。
- 30 天死亡率 AUROC 0.84(95%CI 0.82-0.86)、90 天死亡率 0.84(95%CI 0.82-0.85),准确性与外科医生预测相似。
- 输入特征数量越多,模型性能越稳定或改善;随机森林模型 135 个输入特征时 AUROC 最高。
- 移动设备应用较原网页门户加载数据快约 10 倍,输出显示患者信息、并发症风险、每种并发症前 3 位危险因素及外科医生个人并发症模式。
适用人群与场景
- 所有 18 岁及以上、因任何类型住院手术而入院并接受手术的患者(74 417 例手术、58 236 例患者)。
- 验证队列为 2018 年 11 月 28 日至 2020 年 9 月 20 日前瞻性收集的 22 300 例住院手术、19 132 例患者。
研究设计与关键结果
- 研究类型为预后研究,使用自动电子健康记录数据输入和机器学习算法预测术后并发症,并通过网页门户和移动设备应用向外科医生提供预测。
- 模型开发使用 2014 年 6 月 1 日至 2018 年 11 月 27 日的 52 117 例回顾性住院手术数据;验证使用 2018 年 11 月 28 日至 2020 年 9 月 20 日前瞻性收集的 22 300 例住院手术数据。
- 验证队列 22 300 例手术涉及 19 132 例患者,平均(SD)年龄 58(17)岁,9672 例(50.6%)男性;2765 例(14.5%)为黑人或非裔美国人,14 777 例(77.2%)为白人,1235 例(6.5%)为其他种族,355 例(1.9%)种族未知(数据缺失);979 例(5.1%)为西班牙裔,17 663 例(92.3%)为非西班牙裔,490 例(2.6%)族裔未知(数据缺失)。
- 随机森林模型(135 个输入特征)验证 AUROC:急性肾损伤 0.82(95%CI 0.82-0.83);心血管并发症 0.81(95%CI 0.81-0.82);神经系统并发症(包括谵妄)0.87(95%CI 0.87-0.88);ICU 延长停留 0.89(95%CI 0.88-0.89);机械通气延长 0.91(95%CI 0.90-0.91);脓毒症 0.86(95%CI 0.85-0.87);静脉血栓栓塞 0.82(95%CI 0.81-0.83);伤口并发症 0.78(95%CI 0.78-0.79);30 天死亡率 0.84(95%CI 0.82-0.86);90 天死亡率 0.84(95%CI 0.82-0.85),准确性与外科医生预测相似。
- 输入特征数量越多,模型性能越稳定或改善;随机森林模型 135 个输入特征时 AUROC 最高。
- 移动设备应用较原网页门户可实现高效指纹登录,加载数据快约 10 倍;应用输出显示患者信息、术后并发症风险、每种并发症前 3 位危险因素,以及单个外科医生与同事相比的并发症模式。
证据类型、等级与局限
- 证据类型:预后研究,包含模型开发(回顾性数据)和前瞻性验证。
- 局限:原文未报告局限性。
对床旁实践的改变
- 原文结论认为,自动实时预测术后并发症并通过移动设备输出,在前瞻性验证中于临床环境表现良好,与外科医生预测准确性相当。
提炼说明
- 本条依据论文摘要整理;摘要未涉及「诊断标准与关键阈值」「处置流程与药物剂量」「红旗征与不能漏的情况」「指南或共识出处」,故未列出这些栏目;全文请点「阅读原文」。
英文原文(摘要)· 点击展开
IMPORTANCE: Predicting postoperative complications has the potential to inform shared decisions regarding the appropriateness of surgical procedures, targeted risk-reduction strategies, and postoperative resource use. Realizing these advantages requires that accurate real-time predictions be integrated with clinical and digital workflows; artificial intelligence predictive analytic platforms using automated electronic health record (EHR) data inputs offer an intriguing possibility for achieving this, but there is a lack of high-level evidence from prospective studies supporting their use.
OBJECTIVE: To examine whether the MySurgeryRisk artificial intelligence system has stable predictive performance between development and prospective validation phases and whether it is feasible to provide automated outputs directly to surgeons' mobile devices.
DESIGN, SETTING, AND PARTICIPANTS: In this prognostic study, the platform used automated EHR data inputs and machine learning algorithms to predict postoperative complications and provide predictions to surgeons, previously through a web portal and currently through a mobile device application. All patients 18 years or older who were admitted for any type of inpatient surgical procedure (74 417 total procedures involving 58 236 patients) between June 1, 2014, and September 20, 2020, were included. Models were developed using retrospective data from 52 117 inpatient surgical procedures performed between June 1, 2014, and November 27, 2018. Validation was performed using data from 22 300 inpatient surgical procedures collected prospectively from November 28, 2018, to September 20, 2020.
MAIN OUTCOMES AND MEASURES: Algorithms for generalized additive models and random forest models were developed and validated using real-time EHR data. Model predictive performance was evaluated primarily using area under the receiver operating characteristic curve (AUROC) values.
RESULTS: Among 58 236 total adult patients who received 74 417 major inpatient surgical procedures, the mean (SD) age was 57 (17) years; 29 226 patients (50.2%) were male. Results reported in this article focus primarily on the validation cohort. The validation cohort included 22 300 inpatient surgical procedures involving 19 132 patients (mean [SD] age, 58 [17] years; 9672 [50.6%] male). A total of 2765 patients (14.5%) were Black or African American, 14 777 (77.2%) were White, 1235 (6.5%) were of other races (including American Indian or Alaska Native, Asian, Native Hawaiian or Pacific Islander, and multiracial), and 355 (1.9%) were of unknown race because of missing data; 979 patients (5.1%) were Hispanic, 17 663 (92.3%) were non-Hispanic, and 490 (2.6%) were of unknown ethnicity because of missing data. A greater number of input features was associated with stable or improved model performance. For example, the random forest model trained with 135 input features had the highest AUROC values for predicting acute kidney injury (0.82; 95% CI, 0.82-0.83); cardiovascular complications (0.81; 95% CI, 0.81-0.82); neurological complications, including delirium (0.87; 95% CI, 0.87-0.88); prolonged intensive care unit stay (0.89; 95% CI, 0.88-0.89); prolonged mechanical ventilation (0.91; 95% CI, 0.90-0.91); sepsis (0.86; 95% CI, 0.85-0.87); venous thromboembolism (0.82; 95% CI, 0.81-0.83); wound complications (0.78; 95% CI, 0.78-0.79); 30-day mortality (0.84; 95% CI, 0.82-0.86); and 90-day mortality (0.84; 95% CI, 0.82-0.85), with accuracy similar to surgeons' predictions. Compared with the original web portal, the mobile device application allowed efficient fingerprint login access and loaded data approximately 10 times faster. The application output displayed patient information, risk of postoperative complications, top 3 risk factors for each complication, and patterns of complications for individual surgeons compared with their colleagues.
CONCLUSIONS AND RELEVANCE: In this study, automated real-time predictions of postoperative complications with mobile device outputs had good performance in clinical settings with prospective validation, matching surgeons' predictive accuracy.
参考来源
- Ren Y, Loftus TJ, Datta S, et al. Performance of a Machine Learning Algorithm Using Electronic Health Record Data to Predict Postoperative Complications and Report on a Mobile Platform. JAMA Netw Open. 2022. PMID: 35576007. doi:10.1001/jamanetworkopen.2022.11973
供执业医师抢救与诊疗参考,不替代临床判断;剂量请结合体重、肝肾功能和本院药品规格复核,以指南原文为准。