Recent developments in medical technology and their global spread have extended human life expectancy. Even acute illnesses that once had very low survival rates can now be treated quickly and minimally invasively. On the other hand, prognosis and quality of life are becoming increasingly important. For example, cardiogenic cerebral infarction (CCI) is one of the diseases for which the life-saving rate of patients has risen sharply, but the decline in quality of life due to physical and mental impairment in the prognosis and the costs associated with the care of such patients have become a problem. Therefore, early detection and treatment of atrial fibrillation (AF), which is considered to be a cause of cardiogenic stroke, is attracting attention as a means of stroke prevention. However, early detection of AF requires electrocardiographic recording and diagnosis by a physician at the time of AF onset. We have therefore analysed data from different electrocardiographs and developed an AI that can determine the presence or absence of AF with high accuracy. This AI has been designed to operate in the cloud and can receive, analyse and judge data from electrocardiographs worn by people at the onset of atrial fibrillation, whether they are in or out of the hospital. This makes it possible to obtain electrocardiographic data from people in their daily lives, easily determine whether or not they have developed AF and, if it is determined that they have to visit a medical facility and receive treatment at an early stage. As a result, the prevention rate of cardiogenic stroke can be increased, reducing the quality of life due to its sequelae and the costs required to care for the patient. This AI and related system construction will play a major role in addressing the problems associated with a society with a long life expectancy, which is likely to increase in the future.