Patient Domain Supervised Contrastive Learning for Lung Sound Classification Using Mobile Phone

TL;DR

This study introduces Patient Domain Supervised Contrastive Learning (PD-SCL) combined with Audio Spectrogram Transformer to enhance lung sound classification from smartphone recordings, addressing variability and style differences for more accessible diagnosis.

Contribution

The paper presents a novel PD-SCL method integrated with AST, significantly improving lung sound classification accuracy from smartphone data over existing models.

Findings

Performance improved by 2.4% with PD-SCL and AST.

Smartphones can effectively diagnose lung sounds.

Addresses variability in patient data and audio style.

Abstract

Auscultation is crucial for diagnosing lung diseases. The COVID-19 pandemic has revealed the limitations of traditional, in-person lung sound assessments. To overcome these issues, advancements in digital stethoscopes and artificial intelligence (AI) have led to the development of new diagnostic methods. In this context, our study aims to use smartphone microphones to record and analyze lung sounds. We faced two major challenges: the difference in audio style between electronic stethoscopes and smartphone microphones, and the variability among patients. To address these challenges, we developed a method called Patient Domain Supervised Contrastive Learning (PD-SCL). By integrating this method with the Audio Spectrogram Transformer (AST) model, we significantly improved its performance by 2.4\% compared to the original AST model. This progress demonstrates that smartphones can effectively diagnose lung sounds, addressing inconsistencies in patient data and showing potential for broad use beyond traditional clinical settings. Our research contributes to making lung disease detection more accessible in the post-COVID-19 world.