SUN-LB110 The Nutrition Education Using a Health Care Application With Artificial Intelligence in Patients With Diabetes Mellitus

Abstract Background: Diet control is the basis of the treatment of type 2 diabetes. However, the education and practice of diet control for the patients with type 2 diabetes mellitus (T2DM) need a lot of manpower and time. In 2009, we have developed a telemedicine model that nutritionists analyze photos of T2DM patients’ meal and supervise them remotely. Our system resulted in the improvement of glycemic control of T2DM patients. Recently, the image analysis technology using the artificial intelligence (AI) progresses rapidly. The smart device application “Asken” has an AI-powered photo analysis system which analyzes the photo of the entire meal and identifies the frame of each item as well as its menu and serving amount. In addition, this application delivers individualized dietary messages and feedbacks. Case reports: We report two T2DM cases who conducted nutrient intervention by this application. One case was a 72-year-old man whose HbA1c decreased from 7.2% to 6.6% and weighed from 58.7kg to 57.5kg in 4 months. However, his total cholesterol increased from 119mg/dl to 200mg/dl, and low-density lipoprotein cholesterol (LDL) also increased from 47mg/dl to 106mg/dl. Another case is a 60-year-old man whose HbA1c improved from 7.0% to 6.6% and his weight decreased from 78.0kg to 76.0kg in 3 months. Total cholesterol was 140mg/dl to 128mg/dl, and LDL-cholesterol was from 65mg/dl to 54mg/dl. Conclusion: Using this application might be useful for diet control of T2DM patients. The effects of AI-supported nutrient intervention using application like this should be further clarified in the large number of patients.