Analysis course of the disease of type 2 diabetes patients using Markov chains and clustering methods.

The main idea of this research is the creation of a new approach to the prediction of chronic diabetes course. This approach is based on dividing patients into several clusters. We used Machine Learning methods for it. Next, we created a diagram of the course of disease in form Markov chains for each cluster of patients. Diagram in form directed graph. Nodes and edges are conditions of patients and transactions between these conditions. Edges values are the probability of transactions. The method based on 7000 histories of diabetes course. A medical specialist can use this method for creating a prediction of diabetes course for a patient.