Abstract:
|
We have built a computational model for individual aging trajectories of health and survival that is conditioned on background baseline information. With a machine learning approach, we use an interpretable interaction network, where health variables are coupled by explicit linear pair-wise interactions within a stochastic dynamical system. Our model is scalable to large longitudinal data sets, is predictive of individual high-dimensional health trajectories and survival from baseline health states, and infers an interpretable network of directed interactions between the health variables. We use English Longitudinal Study of Aging (ELSA) data to train our model and show that it performs better than dedicated linear models for health outcomes and survival, and as well as more general deep-learning models that are less interpretable. Our model can be used to generate synthetic individuals that age realistically, to impute missing data, and to simulate future aging outcomes given arbitrary initial health states. We are continuing to develop simple, deep, and interpretable models of high-dimensional organismal aging and mortality.
|