Abstract:Electrocardiogram (ECG) analysis is one of the earliest and the most successful fields of automated analysis using artificial intelligence (AI). The establishment of authoritative and opensourced or transplantable ECG database, and the generation of excellent algorithms require for the cooperation and runningin of disciplines including computer technology and biomedical engineering, and medical professionals. It also needs medical institutions, medical academic disciplines and administrative departments, and science and technology companies to work together; it also needs supports from a set of improved policies, laws and regulations to make a better promotion and application. This paper analyzes the demand, status of database and algorithm, and the existing key issues of ECG AI automatic analysis.