Spectrogram Images of Breathing Sounds for COVID-19 and other Pulmonary Abnormalities
Published: 25 May 2021| Version 2 | DOI: 10.17632/pr7bgzxpgv.2
Contributors:
, Gokul Kv, , , , , Description
The dataset contains the spectrogram images of the real and generated human breathing sounds. The real breathing sounds are procured from Covid-19 patients as well as people not affected by Covid-19 (normal). The generated breathing sounds are procured from online medical repositories for educational purposes. The breathing sounds are segregated into 4 categories: Crackles-coarse, Crackles-fine, Wheezes, and Normal. The dataset comprises spectrogram images for - 420 coarse crackles (generated), - 375 fine crackles (generated), - 1143 normal (generated), - 322 wheezes (generated), - 1734 Covid-19 (real), - 2690 non Covid-19 (real)
Files
Institutions
Indian Institute of Science, PES University, M S Ramaiah Institute of Technology, Concordia University
Categories
Pneumonia, Pattern of Breathing, Sound, Flu Pandemic, Image Analysis, COVID-19