A Convolutional Neural Network model using audio features achieved 92.4% accuracy and an AUC of 97% in detecting and classifying respiratory conditions, outperforming traditional ML techniques.
Does Deep Learning (CNN) improve the detection and classification accuracy of respiratory conditions from audio recordings compared to traditional ML techniques?
Deep learning models, specifically CNNs, demonstrate high accuracy in detecting and classifying respiratory diseases from non-invasive audio recordings.
Effect estimate: AUC 97%
This paper presents a comprehensive approach for detecting respiratory diseases using IoMTs and Machine Learning (ML) algorithms, leveraging audio recordings from multiple sensor locations on the body. By capturing, extracting, and analyzing diverse audio features, such as MFCC, STFT, and Mel-spectrogram, we aim to detect and classify six respi-ratory conditions: Bronchiectasis, Bronchiolitis, COPD, Healthy, Pneumonia, and URTI. We used a publicly annotated dataset to conduct the experiment and analyze the performance of our proposed approach. This dataset underwent preprocessing, which included feature extraction, removal of rare diseases, data flattening, and encoding for model training. Our findings demonstrate that Deep Learning (DL), such as the Convolutional Neural Network (CNN) model achieved the highest accuracy of 92.4 % and an AU C of 97 %, highlighting its potential in audio-based diagnostics. Our experimental results prove that DL, particularly CNN, outperforms traditional ML techniques in detection accuracy, which makes them a good choice in developing non-invasive, efficient, and cost-effective solutions for respiratory disease detection.
Ea et al. (Mon,) conducted a other in Respiratory diseases. Convolutional Neural Network (CNN) model vs. Traditional ML techniques was evaluated on Detection and classification accuracy (AUC 97%). A Convolutional Neural Network model using audio features achieved 92.4% accuracy and an AUC of 97% in detecting and classifying respiratory conditions, outperforming traditional ML techniques.