Estimating Counterfactual Treatment Outcomes over Time Through Adversarially Balanced Representations
This work addresses the challenge of determining optimal treatment timing and selection for patients in medical settings, representing an incremental improvement over existing methods.
The paper tackles the problem of estimating treatment effects over time from observational data by introducing the Counterfactual Recurrent Network (CRN), which uses adversarial training to balance representations and reduce bias from time-varying confounders, achieving lower error in counterfactual estimation and treatment selection compared to state-of-the-art methods on a simulated tumor growth model.
Identifying when to give treatments to patients and how to select among multiple treatments over time are important medical problems with a few existing solutions. In this paper, we introduce the Counterfactual Recurrent Network (CRN), a novel sequence-to-sequence model that leverages the increasingly available patient observational data to estimate treatment effects over time and answer such medical questions. To handle the bias from time-varying confounders, covariates affecting the treatment assignment policy in the observational data, CRN uses domain adversarial training to build balancing representations of the patient history. At each timestep, CRN constructs a treatment invariant representation which removes the association between patient history and treatment assignments and thus can be reliably used for making counterfactual predictions. On a simulated model of tumour growth, with varying degree of time-dependent confounding, we show how our model achieves lower error in estimating counterfactuals and in choosing the correct treatment and timing of treatment than current state-of-the-art methods.