Predicting the causative pathogen among children with osteomyelitis using Bayesian networks – improving antibiotic selection in clinical practice
作者：
Highlights：
• Establish a generalisable methodological framework to help improve our understanding of the epidemiology of bone infections in children.
• Model the relationship between unobserved infecting pathogens, observed culture results, and clinical and demographic variables.
• Expert knowledge plays a critical role in building the model of paediatric osteomyelitis, building on what the data provides.
• Illustrate the use of utility function in translating probabilistic model outputs to implementable recommendations for antibiotic selection.
• This approach can be applied broadly to antibiotic decision-making under imperfect information - a critical challenge in clinical medicine.
摘要
•Establish a generalisable methodological framework to help improve our understanding of the epidemiology of bone infections in children.•Model the relationship between unobserved infecting pathogens, observed culture results, and clinical and demographic variables.•Expert knowledge plays a critical role in building the model of paediatric osteomyelitis, building on what the data provides.•Illustrate the use of utility function in translating probabilistic model outputs to implementable recommendations for antibiotic selection.•This approach can be applied broadly to antibiotic decision-making under imperfect information - a critical challenge in clinical medicine.
论文关键词：Bone infection,Infectious disease,Bayesian belief network,Clinical decision support,Causal diagram,Probabilistic graph model
论文评审过程：Received 27 September 2019, Revised 19 May 2020, Accepted 29 May 2020, Available online 3 June 2020, Version of Record 12 June 2020.
论文官网地址：https://doi.org/10.1016/j.artmed.2020.101895
