文献速递鼻科

基于机器学习的慢性鼻窦炎伴鼻息肉术后预后预测模型开发与验证

[Development and validation of a machine learning-based prognostic model for chronic rhinosinusitis with nasal polyps].
PubMed 文献 · Zhonghua Er Bi Yan Hou Tou Jing Wai Ke Za Zhi · 2026-01-01

前瞻性纳入333例接受内镜鼻窦手术的CRSwNP患者,按7∶3分为训练集和测试集,利用12种特征选择算法构建113种机器学习模型预测术后1年预后不良风险。glmBoost+GBM集成模型表现最佳,训练集平均AUC为0.894,独立测试集AUC为0.813(95%CI:0.728-0.899),校准良好,具有临床风险分层价值。

要点

  • CRSwNP术后1年预后不良发生率为36.34%。
  • glmBoost+GBM模型预测性能最佳,测试集AUC为0.813。
  • 关键预测变量包括嗅裂息肉、高龄、组织嗜酸粒细胞高、组织中中性粒细胞低等。

核心结论

  • glmBoost+GBM集成学习模型结合特征筛选可有效预测CRSwNP患者术后预后,兼具高预测性能和良好可解释性。

研究设计与样本

  • 前瞻性纳入武汉大学人民医院耳鼻咽喉头颈外科2021年3月至2024年7月接受内镜鼻窦手术的333例CRSwNP患者,按7∶3比例随机分为训练集(n=233)和测试集(n=100),收集术前临床资料及随访1年数据,根据EPOS 2020评估并分组,利用12种特征选择算法构建113种机器学习模型。

主要发现

  • 333例患者随访1年,预后不良者占36.34%。
  • 预后不良组年龄较高,嗅裂息肉发生率、既往饮酒史、外周血及组织嗜酸粒细胞计数、Lund-Mackay CT评分、VAS评分、嗅觉评分及IL-2、IL-6、IL-10、IL-17、TNF-α、IFN-γ水平较高(P值均<0.05)。
  • 预后不良组外周血及组织中中性粒细胞计数、IgM及补体C3水平较低(P值均<0.05)。
  • glmBoost+GBM模型在训练集平均AUC为0.894,独立测试集AUC为0.813(95%CI:0.728-0.899),校准良好,决策曲线显示具有临床应用价值。
  • 主要预测变量包括嗅裂息肉存在、高龄、组织中嗜酸粒细胞高、组织中中性粒细胞低、基线高VAS及基线嗅觉减退等。

关键数值与推荐

  • 根据EPOS 2020评估用药、症状和内镜结果,将患者分为预后良好与不良两组。
  • 模型性能评估采用训练集10折交叉验证的AUC,测试集验证可靠性,SHAP值排序特征重要性。

临床怎么用

  • 该模型可用于CRSwNP患者术后预后风险分层,辅助临床决策。

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 predictive models based on multiple machine learning (ML) algorithms, to systematically compare the predictive performance of these models in assessing the risk of poor postoperative prognosis in patients with chronic rhinosinusitis with nasal polyps (CRSwNP), and to further explore their potential clinical application value. Methods: A total of 333 CRSwNP patients who underwent endoscopic sinus surgery were prospectively enrolled from the Department of Otorhinolaryngology and Head and Neck Surgery, Renmin Hospital of Wuhan University, from March 2021 to July 2024. All patients were randomly allocated into a training set (n=233) and a test set (n=100) at a 7∶3 ratio. Preoperative clinical data and 1-year follow-up data were collected. Medication status, clinical symptoms and endoscopic findings were assessed according to the European Position Paper on Rhinosinusitis and Nasal Polyps 2020 (EPOS 2020), and patients were classified into favorable and poor prognosis groups. Differential analysis was conducted to screen independent prognostic factors. A total of 113 machine learning models were established using 12 feature selection algorithms, with prognostic factors as feature variables for model training. Model performance was assessed using the area under the curve (AUC) derived from 10-fold cross-validation in the training set. Model reliability was validated in the test set, and SHAP values were used to rank the importance of predictive features. Results: At the 1-year follow-up of 333 CRSwNP patients, the incidence of poor postoperative prognosis was 36.34%. Compared with patients in the favorable prognosis group, patients in the poor prognosis group were older, with a higher incidence of olfactory cleft polyps and a higher rate of prior alcohol consumption. They also had elevated peripheral blood and tissue eosinophil counts, higher Lund-Mackay CT scores, Visual Analogue Scale (VAS) scores and olfactory scores, as well as elevated levels of interleukin (IL)-2, IL-6, IL-10, IL-17, tumor necrosis factor α (TNF-α) and interferon γ (IFN-γ). In contrast, peripheral blood and tissue neutrophil counts, serum immunoglobulin M (IgM) and complement C3 levels were markedly decreased (all P<0.05). Among the 113 machine learning models constructed using 12 algorithms, the ensemble model of generalized linear model boosting plus gradient boosting machine (glmBoost+GBM) yielded the optimal performance, with a mean AUC of 0.894 in the training set. Key predictive features included olfactory cleft polyps, advanced age, elevated local eosinophils, reduced tissue neutrophils, elevated baseline VAS scores, and baseline olfactory impairment. In an independent test set, the model achieved an AUC of 0.813 (95%CI: 0.728-0.899) with good calibration. Decision curve analysis further verified its favorable clinical application value. Conclusions: The glmBoost+GBM ensemble learning model combined with feature selection can effectively predict postoperative prognosis in CRSwNP patients, with good predictive efficacy and interpretability. This model provides a reliable tool for clinical risk stratification of CRSwNP patients. 目的: 基于多种机器学习算法构建预测模型,系统比较不同模型在评估慢性鼻窦炎伴鼻息肉(CRSwNP)术后预后不良风险中的预测性能,并探索其潜在的临床应用价值。 方法: 前瞻性纳入武汉大学人民医院耳鼻咽喉头颈外科自2021年3月至2024年7月接受内镜鼻窦手术的333例CRSwNP患者,并按7∶3比例随机划分为训练集和测试集。收集术前临床资料及随访1年数据,根据欧洲鼻窦炎和鼻息肉意见书2020(EPOS 2020)评估用药、症状和内镜结果,将患者分为预后良好与不良两组,并通过差异分析筛选预后影响因素。利用12种特征选择算法构建113种机器学习模型,以预后影响因素为特征变量训练模型,并在训练集中通过10折交叉验证计算受试者工作特征曲线下面积(AUC)评估性能,在测试集验证模型可靠性,使用SHAP值进行特征重要性排序。 结果: 333例CRSwNP患者随访1年,预后不良者占36.34%。预后不良组患者年龄较高,嗅裂息肉发生率、既往饮酒史、外周血及组织中嗜酸粒细胞计数、Lund-Mackay CT评分、视觉模拟量表(VAS)评分、嗅觉评分及白细胞介素(IL)-2、IL-6、IL-10、IL-17、肿瘤坏死因子-α(TNF-α)、干扰素-γ(IFN-γ)水平较高,外周血及组织中中性粒细胞计数、免疫球蛋白M(IgM)及补体C3水平较低(P值均<0.05)。基于12类算法构建113种机器学习模型,其中广义线性模型提升算法联合梯度提升机(generalized linear model boosting+gradient boosting machine,glmBoost+GBM)表现最佳,训练集平均AUC为0.894。主要预测变量包括嗅裂息肉存在、高龄、组织中嗜酸粒细胞高、组织中中性粒细胞低、基线高VAS及基线嗅觉减退等。在独立测试集中,该模型AUC为0.813(95%CI:0.728~0.899),校准良好,决策曲线显示具有临床应用价值。 结论: glmBoost+GBM集成学习模型结合特征筛选可有效预测CRSwNP术后预后,兼具高预测性能和良好可解释性,为临床风险分层提供了可靠工具。.

来源:Zhonghua Er Bi Yan Hou Tou Jing Wai Ke Za Zhi,查看原文。版权归原作者所有。

引用

  1. Huang JY, Qin YR, Chu JJ, et al. [Development and validation of a machine learning-based prognostic model for chronic rhinosinusitis with nasal polyps]. Zhonghua Er Bi Yan Hou Tou Jing Wai Ke Za Zhi. 2026. PMID: 42270300. doi:10.3760/cma.j.cn115330-20251201-00640

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。