Challenging common bolus advisor for self-monitoring type-I diabetes patients using Reinforcement Learning

Frédéric Logé (1, 2) , Erwan Le Pennec (3, 1) , Habiboulaye Amadou-Boubacar (2)

1 CMAP - Centre de Mathématiques Appliquées - Ecole Polytechnique
2 Air Liquide, Centre de Recherche Claude-Delorme, Paris-Saclay, France.
3 XPOP - Modélisation en pharmacologie de population

Patients with diabetes who are self-monitoring have to decide right before each meal how much insulin they should take. A standard bolus advisor exists, but has never actually been proven to be optimal in any sense. We challenged this rule applying Reinforcement Learning techniques on data simulated with T1DM, an FDA-approved simulator developed by Kovatchev et al. modeling the gluco-insulin interaction. Results show that the optimal bolus rule is fairly different from the standard bolus advisor, and if followed can actually avoid hypoglycemia episodes.

Mots clés: Reinforcement learning; Consumer health

Domaines: Machine Learning [stat.ML]

Frédéric Logé, Erwan Le Pennec, Habiboulaye Amadou-Boubacar. Challenging common bolus advisor for self-monitoring type-I diabetes patients using Reinforcement Learning. 2020 KDD Workshop on Applied Data Science for Healthcare, Aug 2020, San Diego, United States. ⟨hal-02903045⟩