文献速递烧伤创面修复

可解释机器学习驱动的烧伤患者皮肤移植失败风险评估工具的开发与验证

Development and Validation of The Skin Graft Loss Risk Estimator: An Interpretable Machine Learning-Driven Tool for Individualized Assessment in Burn Patients.
PubMed 文献 · J Burn Care Res · 2026-09-18

本研究基于美国烧伤协会2013-2022年数据开发并验证可解释机器学习工具,用于个体化预测烧伤患者自体皮肤移植失败风险。随机森林模型AUC-ROC为0.794,识别出烧伤面积≥40%等独立预测因素,并转化为交互式网络临床工具。

要点

  • 基于286479例烧伤患者中24046例自体皮肤移植者开发随机森林模型。
  • 模型预测皮肤移植失败的AUC-ROC为0.794,具有判别性能。
  • 烧伤面积≥40%总体表面积、酒精滥用史、主动吸烟等为独立预测因素。

核心结论

  • 皮肤移植失败风险评估工具提供透明、可解释且个体化的风险预测,以辅助临床判断、患者咨询和手术规划。

研究设计与样本

  • 队列研究,数据来自美国烧伤协会烧伤护理质量平台2013-2022年接受自体皮肤移植的患者。
  • 随机森林机器学习模型在独立的训练集(70%)、验证集(15%)和测试集(15%)上开发与验证。
  • 研究期间共286479例患者,其中24046例接受自体皮肤移植。

主要发现

  • 模型AUC-ROC为0.794。
  • 烧伤≥40%总体表面积(OR 8.16,p<0.001)、酒精滥用史(OR 1.86,p<0.001)、主动吸烟(OR 1.45,p=0.004)、年龄>60岁(OR 1.70,p=0.013)、高血压(OR 1.52,p=0.003)、男性(OR 0.74,p=0.009)、糖尿病(OR 1.45,p=0.030)和接触性烧伤(OR 1.70,p=0.006)为移植失败的独立预测因素。

临床怎么用

  • 模型已转化为交互式、可解释的网络临床工具,可用于增强临床判断、促进患者咨询和手术规划。

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

BACKGROUND: Skin graft failure is a significant complication in burn care that prolongs recovery, compromises surgical outcomes, and necessitates repeat operations. Accurate risk assessment is challenging due to a complex interplay of patient and injury factors, underscoring the need for objective, data-driven predictive tools. We developed and validated an interpretable machine learning (ML)-driven tool to provide individualized, data-driven risk assessment for skin graft loss in burn patients. METHODS: A cohort study of patients who had autologous skin grafting from the American Burn Association Burn Care Quality Platform was conducted from 2013-2022. A Random Forest ML model was developed and validated on independent training (70%), validation (15%), and testing (15%) sets. Model performance was assessed using the area under the receiver operating characteristic curve (AUC-ROC). RESULTS: Of 286,479 patients during the study period, 24,046 underwent autologous skin grafting. ML demonstrated discriminatory predictive performance with AUC-ROC of 0.794. Burns ≥40% total body burn surface area (OR 8.16, p<0.001), history of alcohol abuse (OR 1.86, p<0.001), active tobacco use (OR 1.45, p=0.004), age >60 years (OR 1.70, p=0.013), hypertension (OR 1.52, p=0.003), male sex (OR 0.74, p=0.009), diabetes mellitus (OR 1.45, p=0.030) and contact burns (OR 1.70, p=0.006) emerged as independent predictors of graft loss. The model was translated into an interactive, interpretable web-based clinical tool. CONCLUSIONS: The Skin Graft Loss Risk Estimator provides a transparent, interpretable and personalized risk assessment of skin graft loss to enhance clinical judgment, facilitate effective patient counselling and surgical planning in burn care.

来源:J Burn Care Res,查看原文。版权归原作者所有。

引用

  1. Hassan AM, Liu HY, Spera LJ, et al. Development and Validation of The Skin Graft Loss Risk Estimator: An Interpretable Machine Learning-Driven Tool for Individualized Assessment in Burn Patients. J Burn Care Res. 2026. PMID: 42757870. doi:10.1093/jbcr/irag166

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。