Medical Treatment Graph and Bayesian Network for Modeling the Therapy of Patients with Type 1 Diabetes
Rafał Deja, Grażyna Deja · Procedia Computer Science · 2024
In this paper we present the Medical Treatment Graph (MTG) and Bayesian network to model the therapy of patients with type 1 diabetes at onset. Both approaches graphically represent the clinical pathways and allow to plan the therapy and adapt medical decisions based on the current state of a patient and the progress of the treatment. We investigate the dependencies between the medical events and the probability distribution of their transition. We infer the Bayesian network from real medical health records and we show that the inferred network is a good complement to the MTG and that the two models can be used together to improve the treatment of patients.