US 12,476,004 B2
Deep neural network pre-training method for classifying electrocardiogram (ECG) data
Byeongtak Lee, Seoul (KR); Youngjae Song, Anyang-si (KR); Woong Bae, Seoul (KR); and Oyeon Kwon, Seoul (KR)
Assigned to VUNO INC., Seoul (KR)
Filed by VUNO INC., Seoul (KR)
Filed on Sep. 2, 2021, as Appl. No. 17/464,685.
Claims priority of application No. 10-2020-0118669 (KR), filed on Sep. 15, 2020.
Prior Publication US 2022/0084679 A1, Mar. 17, 2022
Int. Cl. G16H 50/20 (2018.01); A61B 5/353 (2021.01); A61B 5/355 (2021.01); A61B 5/36 (2021.01); A61B 5/366 (2021.01); G06N 20/00 (2019.01); G16H 50/70 (2018.01)
CPC G16H 50/20 (2018.01) [A61B 5/353 (2021.01); A61B 5/355 (2021.01); A61B 5/36 (2021.01); A61B 5/366 (2021.01); G06N 20/00 (2019.01); G16H 50/70 (2018.01); G06F 2218/08 (2023.01); G06F 2218/12 (2023.01)] 13 Claims
OG exemplary drawing
 
1. A method for learning an electrocardiogram (ECG) label output model by a computing device comprising:
receiving at least one unlabeled ECG signal and at least one labeled ECG signal;
performing self-supervised learning of an ECG feature extraction model including an encoder and a rule-based feature extractor based on the at least one unlabeled ECG signal; and
performing supervised learning of the ECG label output model based on the at least one labeled ECG signal,
wherein the ECG label output model is configured to include the encoder pre-trained through the self-supervised learning of the ECG feature extraction model and a classifier configured to classify a diagnostic label;
wherein the self-supervised learning of the ECG feature extraction model includes:
extracting at least one first feature from the at least one unlabeled ECG signal using the rule-based feature extractor;
extracting at least one second feature from the at least one unlabeled ECG signal using the encoder;
obtaining at least one output value by mapping the at least one second feature to at least one of a regression function and a classification function; and
performing the self-supervised learning of the ECG feature extraction model using a loss function based on the at least one output value and the at least one first feature.