RespireNet: A Deep Neural Network for Accurately Detecting Abnormal Lung Sounds in Limited Data Setting

TL;DR

RespireNet introduces a CNN-based approach with novel techniques to improve lung sound classification accuracy in limited data scenarios, advancing tele-diagnosis capabilities.

Contribution

The paper presents RespireNet, a novel CNN model with techniques like device-specific fine-tuning and data augmentation for small datasets.

Findings

Achieved 2.2% improvement over state-of-the-art in 4-class classification

Demonstrated effective use of small respiratory datasets

Enhanced tele-screening potential for lung diseases

Abstract

Auscultation of respiratory sounds is the primary tool for screening and diagnosing lung diseases. Automated analysis, coupled with digital stethoscopes, can play a crucial role in enabling tele-screening of fatal lung diseases. Deep neural networks (DNNs) have shown a lot of promise for such problems, and are an obvious choice. However, DNNs are extremely data hungry, and the largest respiratory dataset ICBHI has only 6898 breathing cycles, which is still small for training a satisfactory DNN model. In this work, RespireNet, we propose a simple CNN-based model, along with a suite of novel techniques -- device specific fine-tuning, concatenation-based augmentation, blank region clipping, and smart padding -- enabling us to efficiently use the small-sized dataset. We perform extensive evaluation on the ICBHI dataset, and improve upon the state-of-the-art results for 4-class classification by 2.2%