Original research
Evaluating warfarin dosing models on multiple datasets with a novel software framework and evolutionary optimisation

https://doi.org/10.1016/j.jbi.2020.103634Get rights and content
Under an Elsevier user license
open archive

Highlights

  • Supervised machine learning can be used to predict warfarin maintenance dose.

  • Warfit-learn Python framework enables reproducible warfarin dose estimation research.

  • Linear and support vector regression predict warfarin dose well on two datasets.

  • Stacked ensemble algorithms reduce error when using pharmacogenetic parameters.

  • Genetic programming can automatically evolve accurate prediction models.

Abstract

Warfarin is an effective preventative treatment for arterial and venous thromboembolism, but requires individualised dosing due to its narrow therapeutic range and high individual variation. Many machine learning techniques have been demonstrated in this domain. This study evaluated the accuracy of the most promising algorithms on the International Warfarin Pharmacogenetics Consortium dataset and a novel clinical dataset of South African patients. Support vectors and linear regression were amongst the top performers in both datasets and performed comparably to recent stacked ensemble approaches, whilst neural networks were one of the worst performers in both datasets. We also introduced genetic programming to automatically optimise model architectures and hyperparameters without human guidance. Remarkably, the generated models were found to match the performance of the best models hand-crafted by human experts. Finally, we present a novel software framework (Warfit-learn) for warfarin dosing research. It leverages the most successful techniques in preprocessing, imputation, and parallel evaluation, with the goal of accelerating research and making results in this domain more reproducible.

Keywords

Warfarin
Machine learning
Genetic programming
Python
Supervised learning
Anticoagulant
Pharmacogenetics
Software

Cited by (0)