ITMO University, St. Petersburg, Russia.
Russian Scientific Research Institute of Traumatology and Orthopedics "R.R. Vreden", St. Petersburg, Russia.
Stud Health Technol Inform. 2021 Oct 27;285:106-111. doi: 10.3233/SHTI210581.
In this paper, we present a framework, which aims at facilitating the choice of the best strategy related to the treatment of periprosthetic joint infection (PJI). The framework includes two models: a detailed non-Markovian model based on the decision tree approach, and a general Markov model, which captures the most essential states of a patient under treatment. The application of the framework is demonstrated on the dataset provided by Russian Scientific Research Institute of Traumatology and Orthopedics "R.R. Vreden", which contains records of patients with PJI occurred after total hip arthroplasty. The methods of cost-effectiveness analysis of treatment strategies and forecasting of individual treatment outcomes depending on the selected strategy are discussed.
在本文中，我们提出了一个框架，旨在为与假体周围关节感染（PJI）治疗相关的最佳策略选择提供便利。该框架包括两个模型：一个基于决策树方法的详细非马尔可夫模型，以及一个通用的马尔可夫模型，该模型捕捉了治疗中患者的最基本状态。该框架的应用是基于俄罗斯创伤和矫形科学研究所（R.R. Vreden）提供的数据集进行的，该数据集包含全髋关节置换术后发生 PJI 的患者记录。讨论了治疗策略的成本效益分析方法和根据所选策略预测个体治疗结果的方法。
