Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Combining Scatter Transform and Deep Neural Networks for Multilabel Electrocardiogram Signal Classification

Maximilian P Oppelt, Maximilian Riehl, Felix P Kemeth, Jan Steffan

TL;DR

The paper addresses multilabel ECG classification by integrating a theory-informed scatter transform with a ResNet encoder, followed by self-attention and a differentiable loss aligned with the challenge metric. This hybrid approach leverages fixed, Lipschitz-stable features while retaining learnable components to capture inter-channel interactions, achieving efficient training and better generalization. On the PhysioNet CinC2020 data, the method yields a holdout score of 0.724 compared with 0.682 for a standard ResNet, using fewer parameters and demonstrating faster convergence. These results underscore the value of incorporating inductive bias via scatter transforms for domain-specific time-series classification and suggest practical gains for large-scale, multilabel ECG analysis.

Abstract

An essential part for the accurate classification of electrocardiogram (ECG) signals is the extraction of informative yet general features, which are able to discriminate diseases. Cardiovascular abnormalities manifest themselves in features on different time scales: small scale morphological features, such as missing P-waves, as well as rhythmical features apparent on heart rate scales. For this reason we incorporate a variant of the complex wavelet transform, called a scatter transform, in a deep residual neural network (ResNet). The former has the advantage of being derived from theory, making it well behaved under certain transformations of the input. The latter has proven useful in ECG classification, allowing feature extraction and classification to be learned in an end-to-end manner. Through the incorporation of trainable layers in between scatter transforms, the model gains the ability to combine information from different channels, yielding more informative features for the classification task and adapting them to the specific domain. For evaluation, we submitted our model in the official phase in the PhysioNet/Computing in Cardiology Challenge 2020. Our (Team Triage) approach achieved a challenge validation score of 0.640, and full test score of 0.485, placing us 4th out of 41 in the official ranking.

Combining Scatter Transform and Deep Neural Networks for Multilabel Electrocardiogram Signal Classification

TL;DR

The paper addresses multilabel ECG classification by integrating a theory-informed scatter transform with a ResNet encoder, followed by self-attention and a differentiable loss aligned with the challenge metric. This hybrid approach leverages fixed, Lipschitz-stable features while retaining learnable components to capture inter-channel interactions, achieving efficient training and better generalization. On the PhysioNet CinC2020 data, the method yields a holdout score of 0.724 compared with 0.682 for a standard ResNet, using fewer parameters and demonstrating faster convergence. These results underscore the value of incorporating inductive bias via scatter transforms for domain-specific time-series classification and suggest practical gains for large-scale, multilabel ECG analysis.

Abstract

An essential part for the accurate classification of electrocardiogram (ECG) signals is the extraction of informative yet general features, which are able to discriminate diseases. Cardiovascular abnormalities manifest themselves in features on different time scales: small scale morphological features, such as missing P-waves, as well as rhythmical features apparent on heart rate scales. For this reason we incorporate a variant of the complex wavelet transform, called a scatter transform, in a deep residual neural network (ResNet). The former has the advantage of being derived from theory, making it well behaved under certain transformations of the input. The latter has proven useful in ECG classification, allowing feature extraction and classification to be learned in an end-to-end manner. Through the incorporation of trainable layers in between scatter transforms, the model gains the ability to combine information from different channels, yielding more informative features for the classification task and adapting them to the specific domain. For evaluation, we submitted our model in the official phase in the PhysioNet/Computing in Cardiology Challenge 2020. Our (Team Triage) approach achieved a challenge validation score of 0.640, and full test score of 0.485, placing us 4th out of 41 in the official ranking.

Paper Structure

This paper contains 12 sections, 2 equations, 2 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Methodology
  3. Preprocessing
  4. Augmentation
  5. Scatter Transform
  6. Model Architecture
  7. ResNet
  8. Scatter blocks
  9. Attention
  10. Cost-function
  11. Training
  12. Discussion

Figures (2)

  • Figure 1: Low-pass filter ($\phi$, left) and high-pass filter ($\psi$, right) with the numerical values given in Table \ref{['tab:wav_coeff']}.
  • Figure 2: The scatter blocks use scatter layers in place of convolutions with stride 2. In the reference bottleneck blocks there is a convolution with kernel size 1 and stride 2 in the skip connection when downsampling is performed.