Method, system and non-transitory computer- readable recording medium for detecting and classifying beat in electrocardiogram signal

1 . A method performed in a system for detecting and classifying a beat in an electrocardiogram (ECG) signal, the system comprising one or more processors and the method comprising the steps of:

(a) by the one or more processors, detecting a QRS waveform in an ECG signal using a waveform detection model, and detecting a class of a cardiac event capable of being derived from the QRS waveform; and

(b) by the one or more processors, detecting a unique R-peak in the QRS waveform using a regression model,

wherein the waveform detection model is formed in a structure in which a plurality of encoding blocks and a plurality of decoding blocks are successively connected,

wherein the plurality of encoding blocks include a CBR block and a plurality of residual blocks, the plurality of residual blocks being successively connected at an end of the CBR block, and

wherein the plurality of decoding blocks include a CBR block and a squeeze-and-excitation (SE) block, the SE block being successively connected at an end of the CBR block.

2 . The method of claim 1 , wherein the waveform detection model is a model for performing semantic segmentation on the ECG signal.

3 . The method of claim 1 , wherein the regression model includes a plurality of encoding blocks formed in the same structure as encoding blocks included in the waveform detection model, and a plurality of fully connected (FC) layers.

4 . The method of claim 1 , wherein in step (b), a normalized location of an R-peak is determined for data obtained by resampling the QRS waveform to a predetermined length, and the normalized location of the R-peak is transformed into a physical location in the QRS waveform.

5 . A non-transitory computer-readable recording medium having stored thereon a computer program for executing the method of claim 1 .

6 . A system for detecting and classifying a beat in an ECG signal, the system comprising one or more processors configured to:

detect a QRS waveform in an ECG signal using a waveform detection model, and detect a class of a cardiac event capable of being derived from the QRS waveform; and

detect a unique R-peak in the QRS waveform using a regression model,

wherein the waveform detection model is formed in a structure in which a plurality of encoding blocks and a plurality of decoding blocks are successively connected,

wherein the plurality of encoding blocks include a CBR block and a plurality of residual blocks, the plurality of residual blocks being successively connected at an end of the CBR block, and

wherein the plurality of decoding blocks include a CBR block and a squeeze-and-excitation (SE) block, the SE block being successively connected at an end of the CBR block.

7 . The system of claim 6 , wherein the waveform detection model is a model for performing semantic segmentation on the ECG signal.

8 . The system of claim 6 , wherein the regression model includes a plurality of encoding blocks formed in the same structure as encoding blocks included in the waveform detection model, and a plurality of fully connected (FC) layers.

9 . The system of claim 6 , wherein the one or more processors are configured to determine a normalized location of an R-peak for data obtained by resampling the QRS waveform to a predetermined length, and transform the normalized location of the R-peak into a physical location in the QRS waveform.