Respiratory tract symptom detection method based on smart phone audio perception in driving environment

Abstract

The invention discloses a respiratory tract symptom detection method based on smart phone audio perception in a driving environment. According to the method, a loudspeaker of a smart phone is used forcollecting sounds in a vehicle, vehicle driving noise is filtered out based on a self-adaptive sub-band spectral entropy method, then acoustic characteristics of the denoised sound are extracted andsent to a trained neural network, and whether respiratory tract symptoms such as cough, sneezing and nasal aspiration exist in the collected sound or not is judged. And the frequency of related respiratory tract symptoms is recorded. The method does not depend on various professional medical devices erected in advance, is low in cost and high in anti-interference performance, does not have the privacy leakage problem, and is suitable for the detection environment where driving noise is stable and the distance between a driver and passengers is short. According to the method, the influence of various driving noises is eliminated by adopting a denoising method based on the adaptive sub-band spectral entropy, so that the robustness of the system to environmental noises is relatively high, andthe detection and classification of three typical respiratory tract symptoms can be accurately and efficiently realized.

Description

technical field [0001] The present invention relates to a method for detecting respiratory symptoms, in particular to a method for detecting respiratory symptoms based on the audio perception capabilities of smart phone audio sensors, i.e. speakers and microphones, in a driving environment. The three typical respiratory symptoms of sneezing and sniffing belong to the technical field of mobile computing applications. Background technique [0002] Among the respiratory symptoms closely related to human health, coughing, sneezing and sniffing are the most common respiratory symptoms in daily life. Although these respiratory symptoms may seem insignificant, they are actually associated with more than 100 diseases, such as common diseases such as colds, flu, and allergies, and more serious respiratory diseases such as pneumonia, asthma, and chronic lung disease. Most of these respiratory diseases are curable, but early detection is still needed, especially the infectious respira...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to solve the problem of high cost and low anti-interference ability in detecting the respiratory symptoms of...

Images