Real World Data and Artificial Intelligence in Diabetology

Abstract

In the modern society in which we live, digitalization, big data and artificial intelligence (AI) are widely used in finance, e-commerce, manufacturing, and logistics. This trend is no exception in healthcare, and many healthcare professionals have the expectation that digital healthcare, including AI, which produce and utilize medical big data, can help doctors to improve the quality of healthcare services. In particular, in the endocrine area to which we belong, it can be seen that it is relatively easy to conduct AI research using medical big data, i.e., real world data, which is relatively well organized compared to other diseases. Already, AI technologies for diagnosing diabetic complication or vision recognition technologies for determining diabetic retinopathy have been studied for quite a long time and are also used in clinical practice. Hence, there is no doubt that medical big data will play an essential role in healthcare, especially in endocrinology and diabetology. However, there is a need to review the clinical implications of AI research results utilizing medical big data. Medical staff should clarify the purpose of AI to leverage medical big data. In addition, healthcare professionals must understand the precautions and benefits required to use medical big data when perform AI research. Therefore, in this manuscript, some studies are being conducted using real world data in the field of diabetology, and I would like to discuss the implications of these studies and future development directions. Keywords: Artificial intelligence; Endocrinology; Real world data