Variational Autoencoders for Anomaly Detection in Respiratory Sounds

TL;DR

This paper introduces a weakly-supervised variational autoencoder approach for automatic respiratory disease detection, aiming to provide an accessible and effective diagnostic tool with competitive accuracy using limited data.

Contribution

It presents a novel application of variational autoencoders for respiratory anomaly detection, enabling effective diagnosis with small datasets and simplified training pipelines.

Findings

Achieves 57% accuracy in respiratory disease detection

Uses variational autoencoders for weakly-supervised learning

Provides a practical, accessible diagnostic tool

Abstract

This paper proposes a weakly-supervised machine learning-based approach aiming at a tool to alert patients about possible respiratory diseases. Various types of pathologies may affect the respiratory system, potentially leading to severe diseases and, in certain cases, death. In general, effective prevention practices are considered as major actors towards the improvement of the patient's health condition. The proposed method strives to realize an easily accessible tool for the automatic diagnosis of respiratory diseases. Specifically, the method leverages Variational Autoencoder architectures permitting the usage of training pipelines of limited complexity and relatively small-sized datasets. Importantly, it offers an accuracy of 57 %, which is in line with the existing strongly-supervised approaches.