儿童持续性哮喘和过敏性鼻炎的早期预测因素：一项使用事后可解释性的机器学习研究。
Early predictors of persistent childhood asthma and allergic rhinitis: A machine learning study using post-hoc explainability.
文献信息
| PMID | 42851161 |
|---|---|
| 原文 | 在 PubMed 查看原文 ↗ |
| 发表日期 | 2026 |
| 作者 | Frederikke Rosenvinge Skov |
| 作者单位 | COPSAC, Copenhagen Prospective Studies on Asthma in Childhood, Copenhagen University Hospital-Herlev and Gentofte, Copenhagen, Denmark. |
| 期刊 | Pediatric allergy and immunology : official publication of the European Society of Pediatric Allergy and Immunology |
| SCI 分区 | Q1 |
| IF | 4.3 |
| 研究类型 | AI/ML · 临床 |
| 所属专科 | 鼻科 |
中文摘要
背景： 哮喘和过敏性鼻炎（AR）遵循异质性儿童轨迹，可靠地早期预测持续或缓解仍然具有挑战性。我们研究了应用于早期生活数据的机器学习模型是否能预测长期哮喘和AR结局。
方法： 我们分析了哥本哈根儿童哮喘前瞻性研究（COPSAC2000）队列中411名儿童的113个早期生活临床、遗传、环境、免疫和肺功能特征。训练了五种机器学习模型来预测四种结局：持续性哮喘与对照、持续性哮喘与缓解期哮喘、持续性AR与对照，以及青少年起病AR与对照。使用SHapley Additive exPlanations（SHAP）评估模型可解释性。
结果： 所有结局的预测性能有限，没有模型达到0.65以上的AUROC。在持续性AR与对照中观察到最高性能，基于树的模型表现相似，AUROC约为0.63。对于持续性哮喘与对照，随机森林和XGBoost均达到0.59的AUROC。在区分持续性哮喘与缓解期哮喘以及预测青少年起病AR方面性能较弱，最佳模型分别达到0.60和0.59的AUROC。对最佳性能模型的SHAP分析识别出生物学上合理的特征，包括持续性哮喘的气传过敏原致敏和遗传易感性，以及持续性AR的家族、免疫和生长相关特征，但这些信号并未转化为强区分性能。
结论： 在这个深度表型分析的高风险出生队列中，使用机器学习分析的早期生活数据不能可靠地预测长期哮喘或AR结局。这些发现表明，仅早期生活数据可能不足以进行个体预后分层，可能需要儿童期后期的重复测量来改善预测。
英文摘要
BACKGROUND: Asthma and allergic rhinitis (AR) follow heterogeneous childhood trajectories, and reliable early prediction of persistence or remission remains challenging. We investigated whether machine learning models applied to early-life data could predict long-term asthma and AR outcomes.
METHODS: We analyzed data from 411 children in the Copenhagen Prospective Studies on Asthma in Childhood (COPSAC2000) cohort using 113 early-life clinical, genetic, environmental, immunological, and lung function features. Five machine learning models were trained to predict four outcomes: persistent asthma versus controls, persistent asthma versus asthma in remission, persistent AR versus controls, and teen-onset AR versus controls. Model interpretability was assessed using SHapley Additive exPlanations (SHAP).
RESULTS: Predictive performance was limited across all outcomes, and no model achieved an AUROC above 0.65. The highest performance was observed for persistent AR versus controls, where the tree-based models performed similarly, with AUROCs around 0.63. For persistent asthma versus controls, Random Forest and XGBoost both achieved an AUROC of 0.59. Performance was weak for distinguishing persistent asthma from asthma in remission and for predicting teen-onset AR, with the best models reaching AUROCs of 0.60 and 0.59, respectively. SHAP analyses of the best-performing models identified biologically plausible features, including aeroallergen sensitization and genetic predisposition for persistent asthma, and familial, immunological, and growth-related features for persistent AR, but these signals did not translate into strong discriminative performance.
CONCLUSION: In this deeply phenotyped high-risk birth cohort, early-life data analyzed using machine learning did not reliably predict long-term asthma or AR outcomes. These findings suggest that early-life data alone may be insufficient for individual prognostic stratification and that repeated measurements later in childhood may be needed to improve prediction.