文献速递流行病学与生物统计医学人工智能与信息学

用于空间疾病制图的分层多变量有向无环图自回归模型

Hierarchical multivariate directed acyclic graph autoregressive models for spatial diseases mapping.
PubMed 文献 · Stat Med · 2022-04-06

本文提出分层多变量有向无环图自回归模型,用于区域聚合数据中多种疾病的空间依赖与疾病间关联的分解。模型通过桥接采样实现贝叶斯模型选择与跨顺序平均,以消除建模顺序依赖。模拟研究比较了竞争方法,并应用于监测、流行病学和最终结果规划的多癌种制图数据。

要点

  • 提出分层多变量有向无环图自回归模型,同时刻画空间自相关与疾病间依赖。
  • 利用桥接采样实现贝叶斯模型选择与跨顺序平均,缓解建模顺序依赖。
  • 通过模拟研究比较竞争方法,并应用于监测、流行病学和最终结果规划的多癌种制图。

核心结论

  • 分层多变量有向无环图自回归模型能灵活、可解释地分解多种疾病间的关联与各自的空间自相关,但依赖癌症建模顺序,可用桥接采样进行贝叶斯模型选择与跨顺序平均来消除该依赖。

研究设计与样本

  • 方法学研究提出分层多变量有向无环图自回归模型,用于区域聚合数据的多疾病空间制图。
  • 通过模拟研究将本方法与竞争方法比较。
  • 应用监测、流行病学和最终结果规划的数据进行多癌种制图。

主要发现

  • 分层构造赋予模型灵活性、丰富性、空间自相关与疾病间关系的可解释性以及计算便利性。
  • 模型依赖癌症建模顺序,但可通过桥接采样实现贝叶斯模型选择与跨顺序平均。

临床怎么用

  • 原文未据此给出推荐。

局限性

  • 模型依赖癌症建模顺序。

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

Disease mapping is an important statistical tool used by epidemiologists to assess geographic variation in disease rates and identify lurking environmental risk factors from spatial patterns. Such maps rely upon spatial models for regionally aggregated data, where neighboring regions tend to exhibit similar outcomes than those farther apart. We contribute to the literature on multivariate disease mapping, which deals with measurements on multiple (two or more) diseases in each region. We aim to disentangle associations among the multiple diseases from spatial autocorrelation in each disease. We develop multivariate directed acyclic graphical autoregression models to accommodate spatial and inter-disease dependence. The hierarchical construction imparts flexibility and richness, interpretability of spatial autocorrelation and inter-disease relationships, and computational ease, but depends upon the order in which the cancers are modeled. To obviate this, we demonstrate how Bayesian model selection and averaging across orders are easily achieved using bridge sampling. We compare our method with a competitor using simulation studies and present an application to multiple cancer mapping using data from the Surveillance, Epidemiology, and End Results program.

来源:Stat Med,查看原文。版权归原作者所有。

引用

  1. Gao L, Datta A, Banerjee S, et al. Hierarchical multivariate directed acyclic graph autoregressive models for spatial diseases mapping. Stat Med. 2022. PMID: 35708210. doi:10.1002/sim.9404

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。