mlsanityData to 7 Oct 2026

Which Augmentation Should I Use? An Empirical Investigation of Augmentations for Self-Supervised Phonocardiogram Representation Learning

arXiv 2312.00502 ↗First posted 1 Dec 2023cs.LG cs.SD eess.AScited by 2

Not on the map. Why?

Abstract (arXiv)

Despite recent advancements in deep learning, its application in real-world medical settings, such as phonocardiogram (PCG) classification, remains limited. A significant barrier is the lack of high-quality annotated datasets, which hampers the development of robust, generalizable models that can perform well on newly collected, out-of-distribution (OOD) data. Self-Supervised Learning (SSL) contrastive learning, has shown promise in mitigating the issue of data scarcity by using unlabeled data to enhance model robustness. Even though SSL methods have been proposed and researched in other domains, works focusing on the impact of data augmentations on model robustness for PCG classification are limited. In particular, while augmentations are a key component in SSL, selecting the most suitable policy during training is highly challenging. Improper augmentations can lead to substantial performance degradation and even hinder a network's ability to learn meaningful representations. Addressing this gap, our research aims to explore and evaluate a wide range of audio-based augmentations and uncover combinations that enhance SSL model performance in PCG classification. We conduct a comprehensive comparative analysis across multiple datasets, assessing the impact of various augmentations on model performance. Our findings reveal that depending on the training distribution, augmentation choice significantly influences model robustness, with fully-supervised models experiencing up to a 32% drop in effectiveness when evaluated on unseen data, while SSL models demonstrate greater resilience, losing only 10% or even improving in some cases. This study also highlights the most promising and appropriate augmentations for PCG signal processing, by calculating their effect size on training. These insights equip researchers with valuable guidelines for developing reliable models in PCG signal processing.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 31 of its 82 references that are in the map. First posted on arXiv 1 Dec 2023.

Builds on

31 of 82 references.

References · 31

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  4. 2018Representation Learning with Contrastive Predictive Codingcited by 7,662
  5. 2019Do ImageNet Classifiers Generalize to ImageNet?cited by 1,507
  6. 2019A Comprehensive Survey on Transfer Learningcited by 1,154
  7. 2019Self-Supervised Learning of Pretext-Invariant Representationscited by 806
  8. 2021Domain Generalization: A Surveycited by 760
  9. 2020Self-supervised Learning: Generative or Contrastivecited by 693
  10. 2019Self-supervised Visual Feature Learning with Deep Neural Networks: A Surveycited by 656
  11. 2018Deep Visual Domain Adaptation: A Surveycited by 634
  12. 2023A Survey on Self-supervised Learning: Algorithms, Applications, and Future Trendscited by 205
19 more
  1. 2021BENDR: using transformers and a contrastive self-supervised learning task to learn from massive amounts of EEG datacited by 194
  2. 2020CLOCS: Contrastive Learning of Cardiac Signals Across Space, Time, and Patientscited by 170
  3. 2021Self-Supervised Representation Learning: Introduction, Advances and Challengescited by 147
  4. 2020Uncovering the structure of clinical EEG signals with self-supervised learningcited by 109
  5. 2020Self-supervised ECG Representation Learning for Emotion Recognitioncited by 99
  6. 2021Self-supervised representation learning from 12-lead ECG datacited by 80
  7. 2021Unleashing the Power of Contrastive Self-Supervised Visual Models via Contrast-Regularized Fine-Tuningcited by 50
  8. 2022Lead-agnostic Self-supervised Learning for Local and Global Representations of Electrocardiogramcited by 38
  9. 20213KG: Contrastive Learning of 12-Lead Electrocardiograms using Physiologically-Inspired Augmentationscited by 36
  10. 2021Intra-Inter Subject Self-supervised Learning for Multivariate Cardiac Signalscited by 35
  11. 2020SelfAugment: Automatic Augmentation Policies for Self-Supervised Learningcited by 32
  12. 2019Self-supervised representation learning from electroencephalography signalscited by 24
  13. 2022Transformer-Based Self-Supervised Learning for Emotion Recognitioncited by 16
  14. 2023Towards Domain Generalization for ECG and EEG Classification: Algorithms and Benchmarkscited by 14
  15. 2020Domain Generalization in Biosignal Classificationcited by 8
  16. 2022A Domain Generalization Approach for Out-Of-Distribution 12-lead ECG Classification with Convolutional Neural Networkscited by 5
  17. 2022Listen2YourHeart: A Self-Supervised Approach for Detecting Murmur in Heart-Beat Soundscited by 5
  18. 2022Beyond Heart Murmur Detection: Automatic Murmur Grading from Phonocardiogramcited by 4
  19. 2023Multi-Scale and Multi-Layer Contrastive Learning for Domain Generalizationcited by 4

Built on by

Cited by year

Citing papers per year: 2025: 2.

2 papers.

  1. 2025Domain Knowledge is Power: Leveraging Physiological Priors for Self Supervised Representation Learning in Electrocardiographycited by 1
  2. 2025SS-DPPN: A self-supervised dual-path foundation model for the generalizable cardiac audio representationcited by 0