Project: Longitudinal Data Imputation under Dependent Censoring for Organ Allocation
Description
At any given time, multiple patients are registered on a waitlist for organ transplantation. The new organs are allocated to the patients based on a certain policy. When a new allocation policy is considered, it is first tested on a discrete event simulator. This simulator needs a realistic set of patients and their disease trajectories. We want to construct such a set from observed patient data.
We have the observed disease trajectories of past patients either until they were transplanted or until they died in case no organ became available for them. When a patient was transplanted their natural disease trajectory was effectively censored. We are interested in imputing the disease trajectories for the patients that were transplanted in order to construct a complete set of patients. In other words we want to model how their disease would have progressed if they had not received a transplantation.
The difficulty of this task lies in the fact that patients have been selected for transplantation based on an expectation of their future disease course. Thus, the censoring is dependent on the outcome and standard imputation techniques cannot be applied. We want to develop a technique for longitudinal data imputation in a situation with dependent censoring. Starting point for this research is: https://doi.org/10.1002/sim.7283
Details
- Supervisor
-
Hilde Weerts
- Secondary supervisor
-
Kalina Bakardzhieva
- Interested?
- Get in contact