Current status and future perspective of AI-based diagnosis of arrhythmias

As indicated in the proposal of “Society 5.0” by the cabinet office of the Japanese government, the recent development of artificial intelligence (AI) technology has been vastly changing our society. Also in the field of cardiac arrhythmias, the development of machine learning and deep learning may give us a paradigm shift in the analytical approach. They are applicable to analyzing the simple electrophysiological data and several kinds of imaging data. The AI technology also will be useful for integrated analysis combining multilayer information such as the “big data” in the registry study.

In this symposium, we expect to know the current status of the application of machine learning and deep learning in electrophysiology. It might include the AI analysis in basic research, and also the AI analysis of surface electrocardiogram, intracardiac electrocardiogram, electroanatomical mapping in a clinical electrophysiological study. Another promising AI-based approach is the novel analysis over the multilayer dataset to predict the onset of disease, the efficacy of treatment, the risk of complications, and so on. Since the limitation of AI-based diagnosis machine learning and deep learning in this field is an important issue, the comparisons between these AI-based analyses and conventional analyses are also welcome.

By gathering the innovative trials, we expect this session will get a clue to advance AI-based analysis in multicenter and multimodal approach, combining with the conventional analyses.