COPD screening using time–frequency features of self-recorded respiratory sounds

Chronic obstructive pulmonary disease (COPD) is the third leading cause of death worldwide, with up to 70% of cases remaining undiagnosed. This paper proposes a COPD screening tool based on time-frequency representation features of self-recorded respiratory sounds. Respiratory sound samples (breath and cough sounds) were extracted from COPD and asymptomatic non-COPD volunteers using a large, scientific-purpose database. We analysed 39 time-frequency representation features of breath and cough sounds, combined with age, sex, and smoking status, using Autoencoder neural networks and random forest algorithms. We compared the performance of different breath and cough random forest models built to detect COPD: based exclusively on sound features, based exclusively on sociodemographic characteristics, and based on sound features and sociodemographic characteristics. Models including breathing features outperformed models exclusively based on sociodemographic characteristics. Specifically, the..., , # Data from: COPD screening using time–frequency features of self-recorded respiratory sounds

Dataset DOI: [10.5061/dryad.v41ns1s8g](10.5061/dryad.v41ns1s8g)

## Description of the data and file structure

This dataset, **record_data.csv**, is a new collection of time-frequency features derived from cough and respiration sound samples. The original sound samples were obtained from the proprietary **Covid-19 Sounds database**, a resource developed by Professor Cecilia Mascolo and her team in the Department of Computer Science and Technology at Cambridge University. We were granted permission to compute these features and share the resulting dataset. While the original sound data and associated sociodemographic information cannot be shared here due to proprietary rights held by Cambridge University, the data included in this repository can be used freely. For additional information about the original Covid-19 Sounds database or to request access to the raw sound data, please refer to the..., All human subject data included in this dataset have been de-identified in accordance with relevant privacy regulations. Identifying information has been removed or masked. No data in this dataset can reasonably be used to identify individuals.