Center for Applied Mathematics - Ecole Polytechnique, Palaiseau, France; UMR S1138, University of Paris, INRIA, INSERM, Sorbonne University, Paris, France.
UMR S1138, University of Paris, INRIA, INSERM, Sorbonne University, Paris, France; Department of Statistics, Medical Informatics and Public Health, Hôpital Européen Georges-Pompidou, AP-HP, Paris, France.
Artif Intell Med. 2022 Apr;126:102258. doi: 10.1016/j.artmed.2022.102258. Epub 2022 Feb 23.
Population monitoring is a challenge in many areas such as public health and ecology. We propose a method to model and monitor population distributions over space and time, in order to build an alert system for spatio-temporal data changes. Assuming that mixture models can correctly model populations, we propose a new version of the Expectation-Maximization (EM) algorithm to better estimate the number of clusters and their parameters at the same time. This algorithm is compared to existing methods on several simulated datasets. We then combine the algorithm with a temporal statistical model, allowing for the detection of dynamical changes in population distributions, and call the result a spatio-temporal mixture process (STMP). We test STMPs on synthetic data, and consider several different behaviors of the distributions, to fit this process. Finally, we validate STMPs on a real data set of positive diagnosed patients to coronavirus disease 2019. We show that our pipeline correctly models evolving real data and detects epidemic changes.
人口监测在公共卫生和生态学等许多领域都是一个挑战。我们提出了一种方法来对人口的时空分布进行建模和监测，以便为时空数据变化构建一个警报系统。假设混合模型可以正确地对人口进行建模，我们提出了一种新的期望最大化（EM）算法版本，以便更好地同时估计聚类的数量及其参数。该算法在几个模拟数据集上与现有方法进行了比较。然后，我们将该算法与一个时间统计模型相结合，允许检测人口分布的动态变化，并将结果称为时空混合过程（STMP）。我们在合成数据上测试了 STMP，并考虑了分布的几种不同行为，以适应这个过程。最后，我们将 STMP 应用于 2019 年冠状病毒病的真实阳性诊断患者数据集进行验证。我们表明，我们的流水线正确地对演变的真实数据进行建模并检测到了疫情的变化。
