Timely detection of modifiable risk factors for postoperative pulmonary complications (PPCs) could inform ventilation strategies that attenuate lung injury. We sought to develop, validate, and internally test machine learning models that use intraoperative respiratory features to predict PPCs. We analysed perioperative data from a cohort comprising patients aged 65 yr and older at an academic medical centre from 2019 to 2023. Two linear and four nonlinear learning models were developed and compared with the current gold-standard risk assessment tool ARISCAT (Assess Respiratory Risk in Surgical Patients in Catalonia Tool). The Shapley additive explanation of artificial intelligence was utilised to interpret feature importance and interactions. Perioperative data were obtained from 10 284 patients who underwent 10 484 operations (mean age [range] 71 [65–98] yr; 42% female). An optimised XGBoost model that used preoperative variables and intraoperative respiratory variables had area under the receiver operating characteristic curves (AUROCs) of 0.878 (0.866–0.891) and 0.881 (0.879–0.883) in the validation and prospective cohorts, respectively. These models outperformed ARISCAT (AUROC: 0.496–0.533). The intraoperative dynamic features of respiratory dynamic system compliance, mechanical power, and driving pressure were identified as key modifiable contributors to PPCs. A simplified model based on XGBoost including 20 variables generated an AUROC of 0.864 (0.852–0.875) in an internal testing cohort. This has been developed into a web-based tool for further external validation (). These findings suggest that real-time identification of surgical patients' risk of postoperative pulmonary complications could help personalise intraoperative ventilatory strategies and reduce postoperative pulmonary complications.

中文翻译：

---

利用术中呼吸动态特征开发和验证术后肺部并发症的可解释机器学习模型

及时检测术后肺部并发症（PPC）的可改变危险因素可以为减轻肺损伤的通气策略提供信息。我们寻求开发、验证和内部测试机器学习模型，使用术中呼吸特征来预测 PPC。我们分析了 2019 年至 2023 年学术医疗中心的 65 岁及以上患者队列的围手术期数据。开发了两个线性和四个非线性学习模型，并与当前的金标准风险评估工具 ARISCAT（评估患者呼吸风险）进行了比较。加泰罗尼亚手术患者工具）。人工智能的沙普利附加解释被用来解释特征重要性和相互作用。围手术期数据来自接受 10 484 次手术的 10 284 名患者（平均年龄[范围] 71 [65-98] 岁；42% 为女性）。使用术前变量和术中呼吸变量的优化 XGBoost 模型在验证队列和前瞻性队列中的受试者工作特征曲线下面积 (AUROC) 分别为 0.878 (0.866–0.891) 和 0.881 (0.879–0.883)。这些模型的性能优于 ARISCAT（AUROC：0.496–0.533）。术中呼吸动态系统顺应性、机械功率和驱动压力的动态特征被确定为 PPC 的关键可修改因素。基于 XGBoost 的简化模型（包括 20 个变量）在内部测试队列中生成的 AUROC 为 0.864 (0.852–0.875)。这已被开发成一个基于网络的工具，用于进一步的外部验证（）。这些发现表明，实时识别手术患者术后肺部并发症的风险有助于个性化术中通气策略并减少术后肺部并发症。