Machine Learning Approaches based on Wearable Devices for Respiratory Diseases Diagnosis

The respiratory system, a network of the most important processes of the human body, can easily be affected by different pulmonary diseases that have a great impact on a patient's health. Lung sound auscultation using different wearable devices has been one of the most used, cheap and easy methods to early detect respiratory diseases, but the lack of medical professionals that can put a correct diagnostic based on respiratory sounds has determined the implementation of machine learning and deep learning algorithms to classify and predict respiratory diseases. Therefore, the aim of this article is to present some related works that have been made in this field and the proposed method for classifying the International Conference on Biomedical and Health Informatics (ICBHI' 17) scientific challenge respiratory sound database. The method included the extraction of features using Mel-frequency cepstral coefficients (MFCC) and computing a Convolutional Neural Network (CNN) to classify the database. The results reveal that the proposed method serves an accuracy of 90.21% which provides a suitable method to faster classify any respiratory sounds collected from different devices.