Blind Source Separation in Biomedical Signals Using Variational Methods

TL;DR

This paper presents an unsupervised variational autoencoder-based method for blind source separation of heart and lung sounds in biomedical signals, enabling accurate, label-free reconstruction for clinical applications.

Contribution

It introduces a novel VAE-based approach that learns to separate overlapping biomedical signals without labeled data or prior source knowledge.

Findings

Distinct latent clusters for heart and lung sounds identified

Accurate reconstruction preserving spectral features achieved

Potential for portable diagnostic tools demonstrated

Abstract

This study introduces a novel unsupervised approach for separating overlapping heart and lung sounds using variational autoencoders (VAEs). In clinical settings, these sounds often interfere with each other, making manual separation difficult and error-prone. The proposed model learns to encode mixed signals into a structured latent space and reconstructs the individual components using a probabilistic decoder, all without requiring labeled data or prior knowledge of source characteristics. We apply this method to real recordings obtained from a clinical manikin using a digital stethoscope. Results demonstrate distinct latent clusters corresponding to heart and lung sources, as well as accurate reconstructions that preserve key spectral features of the original signals. The approach offers a robust and interpretable solution for blind source separation and has potential applications in portable diagnostic tools and intelligent stethoscope systems.