文献速递乳腺外科围术期管理与 ERAS

机器学习可准确预测乳房重建术后1年患者报告结局

Machine Learning Accurately Predicts Patient-reported Outcomes 1 Year After Breast Reconstruction.
PubMed 文献 · Ann Surg · 2025-03-05

该研究基于MSKCC 2010年1月至2024年3月期间2687例及外部验证队列2089例乳房重建患者数据,开发并验证五种机器学习算法预测BREAST-Q评分改善与否。模型在腹部躯体健康领域AUC达0.97,乳房满意度0.86,性健康0.79,胸部躯体健康0.78,心理社会健康0.74。术前BREAST-Q评分、放疗时机、BMI、年龄和重建技术是主要预测变量。

要点

  • 机器学习模型可术前预测乳房重建后BREAST-Q各领域改善情况,腹部躯体健康AUC最高达0.97。
  • 术前BREAST-Q评分、放疗时机、BMI、年龄和重建技术是跨领域最重要的预测变量。
  • 该数据驱动方法有望优化共享决策并促进以患者为中心的照护。

核心结论

  • 机器学习算法能在乳房重建术前准确预测患者报告结局,可能简化共享决策并提升以患者为中心的照护。

研究设计与样本

  • 回顾性收集MSKCC 2010年1月至2024年3月乳房重建患者数据,开发并验证五种机器学习算法预测BREAST-Q评分改善与否。
  • 使用Mastectomy Reconstruction Outcomes Consortium多中心数据进行外部验证,共纳入4776例患者(MSKCC 2687例,外部2089例)。
  • 模型评估指标包括受试者工作特征曲线下面积、敏感度、特异度和Brier评分。

主要发现

  • 机器学习算法预测BREAST-Q改善的AUC分别为:腹部躯体健康0.97、乳房满意度0.86、性健康0.79、胸部躯体健康0.78、心理社会健康0.74。
  • 对所有领域模型预测贡献最大的变量为术前BREAST-Q评分、放疗时机、体重指数、年龄和重建技术。

临床怎么用

  • 可利用术前BREAST-Q评分、放疗时机、BMI、年龄和重建技术等变量,通过机器学习模型预测乳房重建术后患者报告结局,辅助共享决策。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

OBJECTIVE: To develop and evaluate machine learning algorithms for predicting patient-reported outcomes following breast reconstruction. BACKGROUND: Machine learning may inform patient-reported outcomes in breast reconstruction, possibly enhancing shared decision-making and tailoring patient care. METHODS: Data on patient characteristics, reconstructive technique, and BREAST-Q scores from women undergoing breast reconstruction at Memorial Sloan Kettering Cancer Center (MSKCC) between January 2010 and March 2024 was retrospectively collected. Five machine learning algorithms were developed and validated on this data to predict improved versus not improved BREAST-Q scores after reconstruction. Models were externally validated models using multicenter data from the Mastectomy Reconstruction Outcomes Consortium. Models were evaluated using the area under the receiver operator curve, sensitivity, specificity, and Brier score. RESULTS: A total of 4776 patients (2687 from MSKCC, 2089 from Mastectomy Reconstruction Outcomes Consortium) were included in model development and validation. Machine learning algorithms demonstrated area under the received operator curves of 0.97 for physical well-being of the abdomen, 0.86 for satisfaction with the breast, 0.79 for sexual well-being, 0.78 for physical well-being of the chest, and 0.74 for psychosocial well-being. Variables that contributed the most to model predictions across all domains were preoperative BREAST-Q scores, timing of radiation, body mass index, age, and reconstructive technique. CONCLUSIONS: Machine learning algorithms can accurately predict patient-reported outcomes before breast reconstruction. Ultimately, this data-driven approach may streamline shared decision-making and enhance patient-centered care.

来源:Ann Surg,查看原文。版权归原作者所有。

引用

  1. Chen J, Gabay A, Boe LA, et al. Machine Learning Accurately Predicts Patient-reported Outcomes 1 Year After Breast Reconstruction. Ann Surg. 2025. PMID: 40040622. doi:10.1097/SLA.0000000000006688

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。