There are approximately 10 million diabetes patients in Japan, and more than 1 million are on insulin therapy. Diabetes mellitus causes various complications (retinopathy, nephropathy, neuropathy, stroke, myocardial infarction, arteriosclerosis obliterans of the lower extremities, etc.), and the development of the complications significantly reduces QOL and life expectancy of the patients. Proper blood glucose control is essential to prevent complications, and therefore diet and exercise therapy and drug treatment (hypoglycemic agents and insulin) are important. It is important to mirror the physiological blood insulin levels of healthy people for adequate glycemic control of diabetics, and a typical means for this purpose is the intensive insulin therapy (multiple daily insulin injection regimen) with additional insulin and basal insulin supplementation.

Image of insulin action

Characteristics of Our Solution

We are developing DM-SAiL (Diabetes Skill Acquisition Learning, SAiL), a programmed medical device based on artificial intelligence (AI) to enable non-diabetes specialists (users) to provide appropriate insulin therapy to hospitalized diabetic patients (target patients), in collaboration with Tohoku University, NEC Corporation (NEC), and NEC Solution Innovators, Ltd. (NES). DM-SAiL assists intensive insulin therapy by non-specialists, and presents recommended dosage units for ultra-rapid-acting insulin preparations and long-acting insulin preparations when implementing intensive insulin therapy.

Progress to date

In April 2022, this project was selected for AMED’s “Medical Engineering Collaboration Innovation Promotion Project (Development and Commercialization Project) (our company is the lead research institution)” and completed analysis work based on patient data of approximately 1,000 people (approximately 1,080,000 clinical parameters) hospitalized at Tohoku University Hospital, developing an AI that predicts insulin dosages prescribed by specialists with an error of about 2 units.

In August 2023, we began clinical performance test for regulatory approval in collaboration with universities and medical institutions such as Tohoku University Hospital, Yamaguchi University Hospital, Sendai City Hospital, Osaki City Hospital, South Miyagi Medical Center, and Tohoku Rosai Hospital, and completed test for the target number of cases, 130. As a result of the analysis, the accuracy rate (average) obtained was 85.46%, which is expected to exceed the target accuracy rate of 80% for the primary endpoint set by the beginning by 5%, proving the non-inferiority (equivalent) of AI predictions to specialists.