Notice of Publication of Related Article on Artificial Intelligence (AI)-based Diabetes Treatment Support System

Renascience has been developing the support system for diabetes treatment utilizing artificial intelligence (AI) in collaboration with Tohoku University. We are pleased to announce that an interview with Professor Hideki Katagiri of the Department of Diabetes and Metabolism, Tohoku University, who is one of the important collaborators on this project, was published in the November 25th issue of Kagaku Shimbun (The Science News).

(Outline of this project)

In Japan, there are approximately 10 million diabetes patients, of which about 1 million are on insulin therapy. Insulin therapy is the most important treatment for preventing diabetic complications, but its use by non-diabetologists is hampered by the large number of different types of the insulin products and the side effect of hypoglycemia. Furthermore, since diabetologists account for less than 2% of all physicians, many diabetes patients are treated by non-diabetologists, limiting the prevention of diabetes-related complications, and placing a heavy burden on both the individual patient and entire society.

In collaboration with Tohoku University, Renascience has been developing the support system for insulin therapy with AI as its core technology to assist non-diabetologists to perform diabetologist-level insulin therapy. We have been working on this task since 2019 using inpatient data from the Department of Diabetes and Metabolism at Tohoku University Hospital. The AI to be developed in this project is highly versatile because it predicts insulin dosage, leveraging the blood glucose levels obtained from self-measurement of blood glucose, which are widely available throughout Japan. Utilizing an AI algorithm based on deep learning, we have already succeeded in developing the AI that predicts insulin dosage with a one-unit error from diabetologists, and several specialists consider the AI can be used already in the clinical practice.

This project was adopted by the Japan Agency for Medical Research and Development (AMED) this fiscal year in a program of the Innovation in Medical Engineering. We plan to further improve the accuracy of the AI, verify its usefulness in clinical studies, and develop and commercialize the support system for the insulin therapy with this AI as its core technology, supported by AMED for three years.

This AI is expected to reduce chronic diabetic complications and adverse events during hospitalization. For family members of diabetes patients, the AI may relieve the burden on caregiving by suppressing the chronic diabetic complications. This AI is also expected to alleviate the burden on non-diabetologists and the medical institutions. Therefore, this project will create value for medical professionals, medical institutions, and families, and all together will make a significant contribution to the society, including a marked reduction in social security benefits.

Please see the following link for the article in the Science News.