文献速递牙周病

人工智能构建牙周炎进展预测模型:纵向队列研究

Developing Predictive Models for Periodontitis Progression Using Artificial Intelligence: A Longitudinal Cohort Study.
PubMed 文献 · J Clin Periodontol · 2025-08-19

本研究利用机器学习对多中心纵向队列中牙周健康者(n=113)和牙周炎患者(n=302)的基线数据进行建模,受试者每两个月检查一次、持续12个月且未接受治疗。概率图模型(PGM)综合临床指标、唾液IL-1β、年龄和性别后表现最佳(AUROC=0.88)。

要点

  • PGM模型AUROC达0.88,优于逻辑回归(0.72)和多层感知机(0.58)。
  • PGM敏感度0.55、特异度0.81,多层感知机敏感度为0。
  • 深牙周袋数量是PGM和多层感知机预测的关键驱动特征。

核心结论

  • 机器学习模型可预测牙周炎进展,支持早期发现策略。
  • 结合临床数据与唾液IL-1β等生物标志物的整合方法提高了预测准确性。

研究设计与样本

  • 多中心纵向队列研究,牙周健康者113例和牙周炎患者302例,每两个月检查一次、持续12个月且未治疗。
  • 采用逻辑回归、多层感知机和概率图模型建模,输入临床与人口学参数及10种唾液分析物水平。
  • 通过AUROC比较14个模型性能,并用SHAP评估特征重要性。

主要发现

  • PGM模型(临床指标、唾液IL-1β、年龄、性别)总体表现最佳,AUROC=0.88。
  • 逻辑回归AUROC=0.72,多层感知机AUROC=0.58。
  • 多层感知机Brier分数较低(0.12),但敏感度为0,限制其临床实用性。
  • PGM敏感度0.55、特异度0.81,表现较为均衡。
  • 特征重要性分析显示深牙周袋数量是PGM和多层感知机预测的关键驱动因素。

关键数值与推荐

  • PGM模型AUROC=0.88,敏感度0.55,特异度0.81。
  • 逻辑回归AUROC=0.72;多层感知机AUROC=0.58,Brier分数0.12,敏感度为0。

临床怎么用

  • 可将临床指标联合唾液IL-1β、年龄和性别输入PGM模型辅助预测牙周炎进展。
  • 关注深牙周袋数量作为模型预测的关键特征。

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

AIM: To construct predictive models of periodontitis progression by applying Machine Learning (ML) to baseline data from a study of periodontitis progression. MATERIALS AND METHODS: Logistic regression (LR), multi-layer perceptron (MLP) and probabilistic graphic model (PGM) were utilised on data from a multi-centre longitudinal study in which periodontally healthy (n = 113) and periodontitis participants (n = 302) were examined bi-monthly for 12 months without treatment. Periodontal examination was performed, and salivary levels of 10 analytes were determined. Clinical and demographic parameters and analytes levels were input into the model. The performance of 14 models was compared using the area under the receiver operating characteristic curve (AUROC), and feature importance was assessed using SHapley Additive exPlanations (SHAP). RESULTS: The PGM model (Clinical measures, saliva IL-1β, age, sex) demonstrated the best overall performance (AUROC = 0.88), compared to LR (AUROC = 0.72) and MLP (AUROC = 0.58). Although MLP had a lower Brier score (0.12), its sensitivity was 0, limiting its clinical utility. In contrast, PGM achieved a balanced sensitivity (0.55) and specificity (0.81). Feature importance analyses highlighted the number of deep periodontal pockets as a key driver of model predictions in both PGM and MLP. CONCLUSIONS: ML models can predict periodontitis progression, supporting early detection strategies. Our integrative approach, combining clinical data with salivary biomarkers such as IL-1β, improved predictive accuracy.

来源:J Clin Periodontol,查看原文。版权归原作者所有。

引用

  1. Furquim CP, Caruth L, Chandrasekaran G, et al. Developing Predictive Models for Periodontitis Progression Using Artificial Intelligence: A Longitudinal Cohort Study. J Clin Periodontol. 2025. PMID: 40830987. doi:10.1111/jcpe.14194

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。