Updating Markov models to integrate cross-sectional and longitudinal studies
作者：
Highlights：
• Built realistic disease ‘trajectories’ from cross-sectional data.
• Learnt pseudo time-series models from these trajectories.
• Used Baum–Welch algorithm to integrate real longitudinal data into the models to calibrate them.
• Tested extensively on simulated data and data from glaucoma patients.
• Demonstrated that this method offers the best of both cross-sectional and longitudinal data.
摘要
Highlights•Built realistic disease ‘trajectories’ from cross-sectional data.•Learnt pseudo time-series models from these trajectories.•Used Baum–Welch algorithm to integrate real longitudinal data into the models to calibrate them.•Tested extensively on simulated data and data from glaucoma patients.•Demonstrated that this method offers the best of both cross-sectional and longitudinal data.
论文关键词：Disease progression,Cross-sectional studies,Markov models
论文评审过程：Received 28 February 2017, Available online 9 March 2017, Version of Record 21 March 2017.
论文官网地址：https://doi.org/10.1016/j.artmed.2017.03.005
