# Reduced-Lead ECG Classifier Model Trained with DivideMix and Model Ensemble

arXiv: https://arxiv.org/abs/2109.12063 · HTML: https://mlsanity.com/paper/2109.12063/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 21 references that are in the map.
- First posted on arXiv 24 Sep 2021.

## Builds on (most cited first, 7 of 7)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [MixMatch: A Holistic Approach to Semi-Supervised Learning](https://mlsanity.com/paper/1905.02249/) (2019)
- [Averaging Weights Leads to Wider Optima and Better Generalization](https://mlsanity.com/paper/1803.05407/) (2018)
- [EfficientNetV2: Smaller Models and Faster Training](https://mlsanity.com/paper/2104.00298/) (2021)
- [A Closer Look at Memorization in Deep Networks](https://mlsanity.com/paper/1706.05394/) (2017)
- [DivideMix: Learning with Noisy Labels as Semi-supervised Learning](https://mlsanity.com/paper/2002.07394/) (2020)
- [Deep learning with noisy labels: exploring techniques and remedies in medical image analysis](https://mlsanity.com/paper/1912.02911/) (2019)

## Abstract (arXiv)

Automatic diagnosis of multiple cardiac abnormalities from reduced-lead electrocardiogram (ECG) data is challenging. One of the reasons for this is the difficulty of defining labels from standard 12-lead data. Reduced-lead ECG data usually do not have identical characteristics of cardiac abnormalities because of the noisy label problem. Thus, there is an inconsistency in the annotated labels between the reduced-lead and 12-lead ECG data. To solve this, we propose deep neural network (DNN)-based ECG classifier models that incorporate DivideMix and stochastic weight averaging (SWA). DivideMix was used to refine the noisy label by using two separate models. Besides DivideMix, we used a model ensemble technique, SWA, which also focuses on the noisy label problem, to enhance the effect of the models generated by DivideMix. Our classifiers (ami_kagoshima) received scores of 0.49, 0.47, 0.48, 0.47, and 0.47 (ranked 9th, 10th, 10th, 11th, and 10th, respectively, out of 39 teams) for the 12-lead, 6-lead, 4-lead, 3-lead, and 2-lead versions, respectively, of the hidden test set with the challenge evaluation metric. We obtained the scores of 0.701, 0.686, 0.693, 0.693, and 0.685 on the 10-fold cross validation, and 0.623, 0.593, 0.606, 0.612, and 0.601 on the hidden validation set for each lead combination.