JMIR Publications

ABSTRACT

Background:

Respiratory sounds have been recognized as a possible indicator of behavior and health. Computer analysis of these sounds can indicate of characteristic sound changes caused by COVID-19 and can be used for diagnosis of this illness

Objective:

The communication aim is development of fast remote computer-assistance diagnosis methods for COVID-19, based on analysis of respiratory sounds

Methods:

Fast Fourier transform (FFT) was applied for computer analysis of respiratory sounds recorded near the mouth of 14 COVID-19 patients (age 18-80) and 17 healthy volunteers (age from 5 to 48). Sampling rate was from 44 to 96 kHz. Unlike usual computer-assistance methods of diagnostics of illness, based on respiratory sound analysis, we propose to test the high frequency part of the FFT spectrum (2000-6000 Hz).

Results:

Comparing FFT spectrums of the respiratory sounds of the patients and volunteers we developed computer-assistance methods of COVID 19 diagnostics and determined numerical healthy-ill criterions. These criterions are independent of gender and age of the tested person.

Conclusions:

The proposed computer methods, based on analysis of the FFT spectrums of respiratory sounds of the patients and volunteers, allows one to automatically diagnose COVID-19 with sufficiently high diagnostic values. These methods can be applied to develop noninvasive self-testing kits for COVID-19.