Accelerated Predictive Healthcare Analytics with Pumas, A High Performance Pharmaceutical Modeling and Simulation Platform

Abstract

Pharmacometric modeling establishes causal quantitative relationship between administered dose, tissue exposures, desired and undesired effects and patient’s risk factors. These models are employed to de-risk drug development and guide precision medicine decisions. Recent technological advances rendered collecting real-time and detailed data easy. However, the pharmacometric tools have not been designed to handle heterogeneous, big data and complex models. The estimation methods are outdated to solve modern healthcare challenges. We set out to design a platform that facilitates domain specific modeling and its integration with modern analytics to foster innovation and readiness to data deluge in healthcare.New specialized estimation methodologies have been developed that allow dramatic performance advances in areas that have not seen major improvements in decades. New ODE solver algorithms, such as coefficient-optimized higher order integrators and new automatic stiffness detecting algorithms which are robust to frequent discontinuities, give rise to up to 4x performance improvements across a wide range of stiff and non-stiff systems seen in pharmacometric applications. These methods combine with JIT compiler techniques and further specialize the solution process on the individual systems, allowing statically-sized optimizations and discrete sensitivity analysis via forward-mode automatic differentiation, to further enhance the accuracy and performance of the solving and parameter estimation process. We demonstrate that when all of these techniques are combined with a validated clinical trial dosing mechanism and non-compartmental analysis (NCA) suite, real applications like NLME parameter estimation see run times halved while retaining the same accuracy. Meanwhile in areas with less prior optimization of software, like optimal experimental design, we see orders of magnitude performance enhancements. Together we show a fast and modern domain specific modeling framework which lays a platform for innovation via upcoming integrations with modern analytics.

Population pharmacokinetics and pharmacodynamics of Tranexamic acid in women undergoing caesarean delivery

“…The population approach was adopted using Pumas v1.0.5 (http://www.pumas.ai). 13 Nonlinear mixed effects modelling with first‐order conditional estimation with interaction approach was applied to characterize TXA concentration and its effect on ML. Models were built hierarchically: a base model containing a structural component and a variability component was determined first, followed by the exploration of covariate models to explain the variability in the parameters.…”

Pharmacokinetics of high-titer anti–SARS-CoV-2 human convalescent plasma in high-risk children

“…Pharmacokinetic analysis: A one-compartment model with a first-order elimination was used to analyze the time course of the antibody titers using Pumas version 2.0 (Pumas-AI, Baltimore, MD, USA)(37). Clearance and volume of distribution of anti-spike IgG and anti-RBD IgG were estimated with an additive residual error.…”

Efficient Precision Dosing Under Estimated Uncertainties via Koopman Expectations of Bayesian Posteriors with Pumas

“…Common probabilistic programming languages, such as Stan [7] or Turing [10], can be utilized to determine these posterior distributions using Markov Chain Monte Carlo (MCMC) techniques such as Hamiltonian Monte Carlo [1, 2, 15]. We will demonstrate how estimated distributions of NLME parameters computed using the Pumas ® software [20] can be reinterpreted as probability distributions via a kernel density estimate (KDE) [16] and used within a Koopman expectation framework to greatly accelerate probabilistic estimates and optimization of clinical choices with respect to uncertainty.…”

“…To demonstrate the utility of this method for performing dosage optimization under uncertainty, we created a high-performance implementation of the Koopman expectation in Pumas ® . Equation 11 was implemented by utilizing DifferentialEquations.jl [21] to calculate U S g given a Pumas ® [20] specification of a dynamical system and the multidimensional integral was calculated using Quadrature.jl, a wrapper library over common quadrature methods such as Cuba [13] and Cubature [12, 6]. This integration implementation allows for a batch solve that parallelizes the computation over the quadrature points, allowing for multithreaded, distributed, and GPU acceleration of the quadrature.…”