Generative Deep Learning and Signal Processing for Data Augmentation of Cardiac Auscultation Signals: Improving Model Robustness Using Synthetic Audio

2024-10-14

arXiv.org

AI Summary

To address label scarcity, class imbalance, and insufficient model robustness in phonocardiogram (PCG) classification, this paper proposes a hybrid data augmentation framework integrating conventional audio augmentations (e.g., noise injection, time-stretching, filtering) with conditional generative diffusion modeling. We are the first to adapt WaveGrad and DiffWave for high-fidelity, class-conditional synthesis of PCG signals, thereby constructing an enriched training dataset. Furthermore, we introduce a multi-metric robustness evaluation framework centered on the Matthews Correlation Coefficient (MCC), jointly assessing in-distribution accuracy and out-of-distribution (OOD) generalization. Extensive experiments across multiple public PCG datasets demonstrate that our method significantly improves CNN-based classifiers’ accuracy, balanced accuracy, and MCC—effectively mitigating data bias and enhancing model stability under realistic clinical acoustic noise conditions.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Computer Vision: Diffusion Models for Vision
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Accurately interpreting cardiac auscultation signals plays a crucial role in diagnosing and managing cardiovascular diseases. However, the paucity of labelled data inhibits classification models' training. Researchers have turned to generative deep learning techniques combined with signal processing to augment the existing data and improve cardiac auscultation classification models to overcome this challenge. However, the primary focus of prior studies has been on model performance as opposed to model robustness. Robustness, in this case, is defined as both the in-distribution and out-of-distribution performance by measures such as Matthew's correlation coefficient. This work shows that more robust abnormal heart sound classifiers can be trained using an augmented dataset. The augmentations consist of traditional audio approaches and the creation of synthetic audio conditionally generated using the WaveGrad and DiffWave diffusion models. It is found that both the in-distribution and out-of-distribution performance can be improved over various datasets when training a convolutional neural network-based classification model with this augmented dataset. With the performance increase encompassing not only accuracy but also balanced accuracy and Matthew's correlation coefficient, an augmented dataset significantly contributes to resolving issues of imbalanced datasets. This, in turn, helps provide a more general and robust classifier.

Problem

• Addressing scarcity of labeled cardiac auscultation data
• Improving robustness of heart sound classifiers
• Enhancing performance with synthetic audio augmentation

Innovation

• Generative deep learning for synthetic audio creation
• Signal processing combined with WaveGrad and DiffWave
• Augmented dataset improves CNN classifier robustness