InVAErt networks for amortized inference and identifiability analysis of lumped parameter hemodynamic models
This work addresses parameter estimation in hemodynamic models for healthcare applications, presenting a novel method but with incremental improvements in handling non-identifiability.
The paper tackled the challenge of estimating cardiovascular model parameters from electronic health records by addressing non-identifiability issues, using inVAErt networks to enable flexible inference from synthetic to real data with missing components.
Estimation of cardiovascular model parameters from electronic health records (EHR) poses a significant challenge primarily due to lack of identifiability. Structural non-identifiability arises when a manifold in the space of parameters is mapped to a common output, while practical non-identifiability can result due to limited data, model misspecification, or noise corruption. To address the resulting ill-posed inverse problem, optimization-based or Bayesian inference approaches typically use regularization, thereby limiting the possibility of discovering multiple solutions. In this study, we use inVAErt networks, a neural network-based, data-driven framework for enhanced digital twin analysis of stiff dynamical systems. We demonstrate the flexibility and effectiveness of inVAErt networks in the context of physiological inversion of a six-compartment lumped parameter hemodynamic model from synthetic data to real data with missing components.