Table 8_PRESCO: an online tool for predicting severe pulmonary complications and survival after cancer surgery.docx

BackgroundSevere pulmonary complications (SPCs) after cancer surgery have a significant impact on morbidity, mortality, and healthcare burden. Despite this, clinicians currently lack accurate and practical tools to predict the occurrence and survival outcomes of SPCs.
MethodsWe conducted a retrospective cohort study of cancer patients undergoing surgery at Hunan Cancer Hospital between June 2023 and June 2025. Two datasets were established (1): 434 patients (227 with SPCs and 207 controls) for predicting SPCs occurrence, and (2) 227 SPC patients with complete follow-up data for 28-day and 90-day survival prediction. Six supervised machine learning classifiers, including linear discriminant analysis (LDA), support vector machine (SVM), random forest (RDF), decision tree (DST), adaptive boosting (ADA), and extremely randomized trees (EXT), were developed. Hyperparameters were optimized using grid search with five-fold stratified cross-validation. Performance was assessed using testing sets by evaluating the Brier score, precision, recall, F1-score, and AUC. SHapley Additive exPlanations (SHAP) were used for model interpretability, and the finalized models were deployed as web-based applications.
ResultsFor SPC occurrence prediction, the EXT model demonstrated the best performance (AUC = 0.813). For 28-day mortality prediction, EXT achieved the highest discrimination (AUC = 0.921), whereas RDF performed best for 90-day mortality (AUC = 0.899). SHAP analysis identified preoperative hypertension, ECOG score, and intraoperative blood loss as the most influential predictors of SPC occurrence. Postoperative SOFA scores, APACHE II scores, and blood urea nitrogen (BUN) were key predictors of 28-day and 90-day mortality. PRESCO (https://presco.streamlit.app/), an online tool, provides real-time prediction of SPC and survival outcomes.
ConclusionsWe developed and validated machine learning models that accurately predict the occurrence and survival of SPCs in cancer patients after surgery. By deploying the online tool, clinicians can easily access it and utilize its functions to perform personalized risk stratification and guide perioperative decisions in oncology.

背景 癌症术后严重肺部并发症（Severe Pulmonary Complications, SPCs）对患者的发病率、死亡率及医疗负担均具有显著负面影响。尽管此类并发症的临床影响不容忽视，但目前临床医师仍缺乏精准且实用的工具，用以预测SPCs的发生风险与患者生存预后。
方法 本研究针对2023年6月至2025年6月期间于湖南省肿瘤医院接受手术治疗的癌症患者开展回顾性队列研究。共构建两类研究数据集：(1) 用于SPCs发生风险预测的队列，包含434例患者（其中227例合并术后严重肺部并发症，207例为对照个体）；(2) 用于28天与90天生存结局预测的队列，包含227例拥有完整随访数据的SPCs患者。本研究开发了六种监督式机器学习分类器，包括线性判别分析（Linear Discriminant Analysis, LDA）、支持向量机（Support Vector Machine, SVM）、随机森林（Random Forest, RDF）、决策树（Decision Tree, DST）、自适应提升算法（Adaptive Boosting, ADA）及极端随机树（Extremely Randomized Trees, EXT）。通过五折分层交叉验证结合网格搜索对模型超参数进行优化。模型性能通过测试集进行评估，评估指标包括布里尔分数（Brier score）、精确率、召回率、F1分数及受试者工作特征曲线下面积（Area Under the Curve, AUC）。采用SHapley加性解释（SHapley Additive exPlanations, SHAP）实现模型可解释性，并将最终训练完成的模型部署为基于网页的应用工具。
结果 针对SPCs发生风险预测任务，EXT模型展现出最优性能（AUC=0.813）。在28天死亡率预测任务中，EXT模型获得了最高的区分度（AUC=0.921）；而90天死亡率预测任务的最优模型为RDF（AUC=0.899）。SHAP分析结果显示，术前高血压、ECOG体能状态评分（Eastern Cooperative Oncology Group, ECOG）及术中失血量是影响SPCs发生的最关键预测因子。术后序贯器官衰竭评分（Sequential Organ Failure Assessment, SOFA）、急性生理学与慢性健康状况评分系统Ⅱ（Acute Physiology and Chronic Health Evaluation Ⅱ, APACHE Ⅱ）及血尿素氮（Blood Urea Nitrogen, BUN）则是预测28天与90天死亡率的核心指标。本研究开发的在线工具PRESCO（https://presco.streamlit.app/）可实时预测SPCs发生风险与患者生存结局。
结论 本研究开发并验证了机器学习模型，可精准预测癌症术后患者SPCs的发生风险与生存结局。通过部署该在线工具，临床医师可便捷获取并使用该系统，实现个性化风险分层，进而指导肿瘤围手术期临床决策。