Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Respiratory Disease Classification and Biometric Analysis Using Biosignals from Digital Stethoscopes

Constantino Álvarez Casado, Manuel Lage Cañellas, Matteo Pedone, Xiaoting Wu, Le Nguyen, Miguel Bordallo López

TL;DR

The paper tackles the need for rapid, accurate diagnostic support for respiratory diseases by leveraging digital stethoscope biosignals. It introduces a multi-step methodology based on Empirical Mode Decomposition to extract Intrinsic Mode Functions and physiologically-relevant features, feeding a classifier and regression models. The authors demonstrate high performance on the ICBHI database, achieving 89% balanced accuracy for healthy vs diseased and 94% for COPD vs other conditions, and they pioneer age, BMI, and sex estimation from acoustic data. This work suggests significant potential for telemedicine and remote patient monitoring using portable auscultation devices. Overall, the approach advances non-invasive, remote diagnostic capabilities by combining advanced signal processing with machine learning on publicly available respiratory sound data.

Abstract

Respiratory diseases remain a leading cause of mortality worldwide, highlighting the need for faster and more accurate diagnostic tools. This work presents a novel approach leveraging digital stethoscope technology for automatic respiratory disease classification and biometric analysis. Our approach has the potential to significantly enhance traditional auscultation practices. By leveraging one of the largest publicly available medical database of respiratory sounds, we train machine learning models to classify various respiratory health conditions. Our method differs from conventional methods by using Empirical Mode Decomposition (EMD) and spectral analysis techniques to isolate clinically relevant biosignals embedded within acoustic data captured by digital stethoscopes. This approach focuses on information closely tied to cardiovascular and respiratory patterns within the acoustic data. Spectral analysis and filtering techniques isolate Intrinsic Mode Functions (IMFs) strongly correlated with these physiological phenomena. These biosignals undergo a comprehensive feature extraction process for predictive modeling. These features then serve as input to train several machine learning models for both classification and regression tasks. Our approach achieves high accuracy in both binary classification (89% balanced accuracy for healthy vs. diseased) and multi-class classification (72% balanced accuracy for specific diseases like pneumonia and COPD). For the first time, this work introduces regression models capable of estimating age and body mass index (BMI) based solely on acoustic data, as well as a model for sex classification. Our findings underscore the potential of intelligent digital stethoscopes to significantly enhance assistive and remote diagnostic capabilities, contributing to advancements in digital health, telehealth, and remote patient monitoring.

Respiratory Disease Classification and Biometric Analysis Using Biosignals from Digital Stethoscopes

TL;DR

The paper tackles the need for rapid, accurate diagnostic support for respiratory diseases by leveraging digital stethoscope biosignals. It introduces a multi-step methodology based on Empirical Mode Decomposition to extract Intrinsic Mode Functions and physiologically-relevant features, feeding a classifier and regression models. The authors demonstrate high performance on the ICBHI database, achieving 89% balanced accuracy for healthy vs diseased and 94% for COPD vs other conditions, and they pioneer age, BMI, and sex estimation from acoustic data. This work suggests significant potential for telemedicine and remote patient monitoring using portable auscultation devices. Overall, the approach advances non-invasive, remote diagnostic capabilities by combining advanced signal processing with machine learning on publicly available respiratory sound data.

Abstract

Respiratory diseases remain a leading cause of mortality worldwide, highlighting the need for faster and more accurate diagnostic tools. This work presents a novel approach leveraging digital stethoscope technology for automatic respiratory disease classification and biometric analysis. Our approach has the potential to significantly enhance traditional auscultation practices. By leveraging one of the largest publicly available medical database of respiratory sounds, we train machine learning models to classify various respiratory health conditions. Our method differs from conventional methods by using Empirical Mode Decomposition (EMD) and spectral analysis techniques to isolate clinically relevant biosignals embedded within acoustic data captured by digital stethoscopes. This approach focuses on information closely tied to cardiovascular and respiratory patterns within the acoustic data. Spectral analysis and filtering techniques isolate Intrinsic Mode Functions (IMFs) strongly correlated with these physiological phenomena. These biosignals undergo a comprehensive feature extraction process for predictive modeling. These features then serve as input to train several machine learning models for both classification and regression tasks. Our approach achieves high accuracy in both binary classification (89% balanced accuracy for healthy vs. diseased) and multi-class classification (72% balanced accuracy for specific diseases like pneumonia and COPD). For the first time, this work introduces regression models capable of estimating age and body mass index (BMI) based solely on acoustic data, as well as a model for sex classification. Our findings underscore the potential of intelligent digital stethoscopes to significantly enhance assistive and remote diagnostic capabilities, contributing to advancements in digital health, telehealth, and remote patient monitoring.
Paper Structure (14 sections, 2 figures, 3 tables)

This paper contains 14 sections, 2 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related work
  3. Proposed Methodology
  4. Preprocesssing
  5. Feature extraction
  6. Classification and Regression Models
  7. Experimental setup
  8. Benchmark Database
  9. Protocol and Metrics
  10. EXPERIMENTS AND ANALYSIS
  11. Classification for Respiratory Conditions
  12. Gender recognition
  13. Regression of Age and BMI
  14. Conclusion

Figures (2)

  • Figure 1: Multi-step methodology for automated diagnosis support for respiratory diseases using stethoscope acoustic signals.
  • Figure 2: Confusion matrices illustrating binary classification (left) and 4-class respiratory infection recognition (right), demonstrating model accuracy and differentiation among health statuses and specific infections. The matrices display normalized values in percentages and the distribution of subjects across each class, highlighting dataset imbalances.