Hierarchical deep learning for autonomous multi-label arrhythmia detection and classification on real-world wearable electrocardiogram data

Objective: Arrhythmia detection and classification are challenging because of the imbalanced ratio of normal heartbeats to arrhythmia heartbeats and the complicated combinations of arrhythmia types. Arrhythmia classification on wearable electrocardiogram monitoring devices poses a further unique challenge: unlike clinically used electrocardiogram monitoring devices, the environments in which wearable devices are deployed are drastically different from the carefully controlled clinical environment, leading to significantly more noise, thus making arrhythmia classification more difficult. Methods: We propose a novel hierarchical model based on CNN+BiLSTM with Attention to arrhythmia detection, consisting of a binary classification module between normal and arrhythmia heartbeats and a multi-label classification module for classifying arrhythmia events across combinations of beat and rhythm arrhythmia types. We evaluate our method on our proprietary dataset and compare it with various baselines, including CNN+BiGRU with Attention, ConViT, EfficientNet, and ResNet, as well as previous state-of-the-art frameworks. Results: Our model outperforms existing baselines on the proprietary dataset, resulting in an average accuracy, F1-score, and AUC score of 95%, 0.838, 0.906 for binary classification, and 88%, 0.736, 0.875 for multi-label classification. Conclusions: Our results validate the ability of our model to detect and classify real-world arrhythmia. Our framework could revolutionize arrhythmia diagnosis by reducing the burden on cardiologists, providing more personalized treatment, and achieving emergency intervention of patients by allowing real-time monitoring of arrhythmia occurrence.