Artificial Intelligence and Machine Learning in Electrophysiology—a Short Review

Purpose of review To summarize the expanding role of artificial intelligence (AI) in cardiac electrophysiology. Recent findings AI is uniquely powered to integrate variable data-streams and consider complex non-linear relationships. Deep learning algorithms can consider aspects in data with unappreciated relevance in order to produce results that are impossible with other methods. The wide adoption of wearable technologies necessitated the development of accurate algorithms to identify cardiac rhythms. Similarly, algorithms use electrocardiograms to make arrhythmic diagnosis, localize arrhythmias, and uncover pathologies such as contractile dysfunction or valvular disease. AI use in imaging and intracardiac electrogram interpretation may enhance efficiency and reproducibility. AI dramatically improves prognostication including for sudden cardiac death, response to catheter ablations, and cardiac resynchronization therapy. AI also holds promise to potentially guide catheter ablation of the future. Summary AI may improve availability, accuracy, and efficiency of electrophysiologic treatments as well as aid in translational research. Ethical and legal challenges will need to be addressed.