The invention relates to a
breathing machine man-
machine asynchronous classification method and
system, a terminal and a storage medium. The method comprises the steps: collecting the multi-channel
breathing data of a simulated flow channel, a
tidal volume channel,
airway pressure, alveolar pressure,
pleural cavity internal pressure, cardiac pressure, a corrugated
pipe position channel and the like under a man-
machine asynchronous event simulated by a simulated
lung and a
breathing machine; carrying out replacement deviation index (PDI)
feature extraction on the
respiration data, and labeling the
respiration data according to the extracted PDI features; inputting the labeled
respiration data into a
network model for training to obtain a trained man-machine asynchronous classification model; and classifying the breathing machine man-machine asynchronous events through the trained man-machine asynchronous classification model. According to the embodiment of the invention, the acquired respiration data is small in interference and convenient to acquire,
difference analysis of the respiration data of the adjacent channels is carried out by using the PDI features, and the accuracy of man-machine asynchronous classification is improved.