IncepSE: Leveraging InceptionTime's performance with Squeeze and Excitation mechanism in ECG analysis
This work addresses ECG analysis for medical diagnostics, but it is incremental as it builds on existing architectures with known modifications.
The authors tackled the challenge of improving ECG analysis by modifying InceptionTime with a Squeeze and Excitation mechanism, resulting in a model that consistently outperforms InceptionTime and other state-of-the-art methods, with a 0.013 AUROC score improvement in the 'all' task.
Our study focuses on the potential for modifications of Inception-like architecture within the electrocardiogram (ECG) domain. To this end, we introduce IncepSE, a novel network characterized by strategic architectural incorporation that leverages the strengths of both InceptionTime and channel attention mechanisms. Furthermore, we propose a training setup that employs stabilization techniques that are aimed at tackling the formidable challenges of severe imbalance dataset PTB-XL and gradient corruption. By this means, we manage to set a new height for deep learning model in a supervised learning manner across the majority of tasks. Our model consistently surpasses InceptionTime by substantial margins compared to other state-of-the-arts in this domain, noticeably 0.013 AUROC score improvement in the "all" task, while also mitigating the inherent dataset fluctuations during training.