摘要
With the development of computer hardware and the growth of clinical database,tremendous progress has been made in the application of deep learning to electrocardiographic data,which provides new ideas for the ex vivo cardiac electrical mapping of atrial fibrillation(AF)substrates.The AF mechanism and current status of AF substrate research are first summarized.Then,the advantages and limitations of cardiac electrophysiological mapping techniques are analyzed.Finally,the application of deep learning to electrocardiogram(ECG)data is reviewed,the problems with the ex vivo intelligent labeling of an AF substrate and the possible solutions are discussed,an outlook on future development is provided.
基金
Supported by the National Natural Science Foundation of China under Grants 62176208.
