Errors-in-variables Modeling of Personalized Treatment-Response Trajectories
This addresses a practical issue in personalized medicine, such as diet monitoring, but is incremental in extending existing methods to handle measurement errors in treatment times.
The paper tackled the problem of estimating treatment effects on continuous temporal responses when covariates and treatment times have measurement errors, introducing a novel method that improved estimation accuracy and prediction in a diet impact on blood glucose case.
Estimating the effect of a treatment on a given outcome, conditioned on a vector of covariates, is central in many applications. However, learning the impact of a treatment on a continuous temporal response, when the covariates suffer extensively from measurement error and even the timing of the treatments is uncertain, has not been addressed. We introduce a novel data-driven method that can estimate treatment-response trajectories in this challenging scenario. We model personalized treatment-response curves as a combination of parametric response functions, hierarchically sharing information across individuals, and a sparse Gaussian process for the baseline trend. Importantly, our model considers measurement error not only in treatment covariates, but also in treatment times, a problem which arises in practice for example when treatment information is based on self-reporting. In a challenging and timely problem of estimating the impact of diet on continuous blood glucose measurements, our model leads to significant improvements in estimation accuracy and prediction.