Constraint reasoning in deep biomedical models

Deep biomedical models are often expressed by means of differential equations. Despite their expressive power, they are difficult to reason about and make decisions, given their non-linearity and the important effects that the uncertainty on data may cause. For this reason traditional numerical simulations may only provide a likelihood of the results obtained. In contrast, we propose in this paper the use of a constraint reasoning framework able to make safe decision notwithstanding some degree of uncertainty, and illustrate this approach in the diagnosis of diabetes and the tuning of drug design.