Expectations surrounding the use of AI in medicine have been increasing, and there are hopes for application in the treatment of diabetes, regarding prevention and treatment efficiency. A Japanese collaboration between Sapporo Medical University, Fujitsu, and Fujitsu Hokuriku Systems will begin development of an AI machine learning model to optimise prescription of oral hypoglycaemic medicines in the treatment of diabetes.
This R&D project aims to use AI to predict the effects of diabetes medicines on diabetic patients, by applying machine learning to patient test values and the prescription data of oral hypoglycaemic medicines. This is done in order to keep glycated haemoglobin (HbA1c) levels below 7%, which is a common target to prevent complications in diabetes.
Based on the clinical insights, dataset creation technology, and AI technology of a research group of clinicians and AI engineers, the three organisations will create a dataset from the large volumes of clinical data. The information will be from systems such as electronic medical record systems and clinical data warehouses.
This technology is expected to make it possible to select and prescribe oral hypoglycaemic medicines that are optimised for patients who do not need insulin preparation, taking combinations of medicines and other factors into account.
Going forward, the three organisations will further improve the accuracy and general applicability of the results of this project, and will conduct R&D applications to transfer the technology to other diseases.
Blood glucose levels
Diabetes requires continual control to keep blood glucose levels within normal boundaries, with the treatment goal being maintaining HbA1c levels of less than 7%. To achieve this, it is important to appropriately provide drug treatment using oral hypoglycaemic medicines or insulin preparation, in addition to general treatment methods, including diet and exercise. In long-term treatment, however, coexisting illnesses can lead to complexities in a patient's condition, and from among the many different oral hypoglycaemic medicines, there is currently no established method for determining prescriptions for oral hypoglycaemic medicines in light of considerations like selection, combinations, sequencing, and side effects of medicines.
