Bean: A Multibranch Network for 12 Leads Electrocardiogram Multilabel Classification Based on Ensemble Learning and Attention

Cardiovascular disease (CVD) patients may suffer from multiple conditions, and in clinical diagnosis, doctors observe the waveform features of electrocardiogram (EEG) signals in each lead and then assess disease abnormalities based on the positions of the leads. Therefore, it is essential to learn multilabel features and lead weights. In addition, EEG is a time series physiological signal, and the semantic relationships between features must be considered. This article proposes a branch network model based on ensemble learning and attention, which includes multibranch networks and a lead ensemble (LE) module. Each branch network embeds a global semantic attention (GSA) module and a class-specific classifier (CSC). First, the GSA in the branch network encodes the global semantic relationships of EEG feature data; the CSC extracts semantic vectors for each class in the feature space, guided by the labels. Finally, the LE module adaptively updates the lead weights and integrates the outputs from all leads for classification. Extensive experiments were conducted on two benchmark datasets (PTB-XL and CPSC2018), and the results demonstrate the superiority of our proposed method. Case analysis further confirms that the features used by the proposed method for EEG classification align with clinical diagnostic rules.