Electrocardiography comprehensive sorting method based on deep learning algorithm

The invention discloses an electrocardiography comprehensive sorting method based on deep learning algorithm. The method comprises the following steps of acquiring original electrocardiography waveform data and electrocardiography additional information as well as electrocardiography rhythm information, representative PQRST waveform data, conducting waveform classification to the related information via trained first deep learning algorithm to achieve a first sorting result, conducting trained second deep learning algorithm to related information to achieve P wave, QRS wave and T wave data and calculating representative PQRST wave featured data and inputting the above into a traditional electrocardiography computer for automatic sorting algorithm to achieve a second sorting result, and adding weight to adjust the sorting results and designating a sorting result having the maximum grade value as a final sorting result. Characteristics of electrocardiography classification are rationally combined; the deep learning method is trained via the above steps and waveform classification is conducted via the deep learning method, so accuracy of sorting result of the electrocardiography explanation can be improved.