Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Method and system for classification and detection of sleep snoring

A technology for classification detection and snoring, applied in speech analysis, instruments, etc., can solve the problems of untreated patients, high labor consumption, and high cost.

Active Publication Date: 2017-11-17
SOUTH CHINA UNIV OF TECH
View PDF6 Cites 20 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Polysomography (PSG) is the gold standard for diagnosing OSAHS, but PSG diagnosis takes a long time, consumes a lot of manpower and is expensive, and many patients in need around the world cannot receive timely treatment

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method and system for classification and detection of sleep snoring
  • Method and system for classification and detection of sleep snoring
  • Method and system for classification and detection of sleep snoring

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0142] Such as figure 1 As shown, the present embodiment provides a sleep snoring classification detection method, the method includes the following steps:

[0143] S1. Pick up the sleep snoring sound of the subject throughout the night, and extract each sleep snoring signal according to the sleep snoring sound signal of the subject throughout the night;

[0144] S101. Detect potential snoring segment

[0145] The sleep snoring signal picked up by the microphone is pre-processed with pre-emphasis and frame division, and the noise reduction processing is performed on the sleep snoring signal based on the spectral subtraction of the power spectrum. Subtract the noise power spectrum with reference from the signal to obtain the sleep snoring signal after spectrum subtraction;

[0146] Calculate the effective value of the sleep snoring signal after spectrum subtraction, and determine the effective value signal threshold according to the effective value contour of the sleep snorin...

Embodiment 2

[0265] Such as image 3 As shown, the present embodiment provides a sleep snoring classification detection system, which can be realized by computer software, mobile phone app or hardware module with a digital signal processor, including a signal extraction module, a calculation module, an identification module and a statistical prediction module , the specific functions of each module are as follows:

[0266] The signal extraction module is used to pick up the sleep snoring sound of the subject throughout the night, and extracts each sleep snoring signal according to the sleep snoring sound signal of the subject throughout the night; the module is as follows: Figure 4 As shown, it includes a potential snoring segment detection unit, a potential snoring segment feature extraction unit, and a snoring automatic detection unit. The functions of each unit are as follows:

[0267] The potential snoring segment detection unit is used to detect the potential snoring segment, such a...

Embodiment 3

[0279] This embodiment is a specific application example. Six cases of OSAHS patients who were diagnosed as moderate and severe OSAHS by PSG in the First Affiliated Hospital of Guangzhou Medical University were selected. Before and after the event, there were 878 snoring sounds (accounting for 6.27% of the total snoring sounds), 262 snoring sounds during apnea (accounting for 1.87%), 691 snoring sounds during hypopnea (accounting for 4.94%), and 12169 common snoring sounds (accounting for 86.92%) . 300 common snoring clips were randomly selected from the normal snoring sounds of 6 patients throughout the night, and 878 snoring sounds before and after respiratory disturbance events, 262 snoring sounds during apnea, 691 snoring sounds during hypoventilation and 1800 common snoring sounds constituted the implementation sample set.

[0280] Perform principal component analysis on the characteristics of the snoring sound sample, and select 27 features such as spectrum centroid, spe...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a method and system for classification and detection of sleep snoring. The method includes the steps of picking up the sleep snoring of a patient all night long, and extracting each snoring signal according to the sleep snoring signals of the patient all night long; calculating related features of fourth types of snoring including the snoring before and after a breathing disorder event, apnea snoring, hypopnea snoring and general snoring in the sleep snoring all night long; performing feature dimension reduction using principal component analysis (PCA), classifying the sleep snoring all night long respectively according to the snoring before and after the breathing disorder event, apnea snoring, hypopnea snoring and general snoring through a multi-class support vector machine (SVM), and realizing the recognition of the four types of snoring; and conducting statistics on the snoring signals all night long to obtain statistical results of the number of times of the four types of snoring, and predicting an AHI value according to the statistical results. According to the invention, the automatic classification of four types of snoring is realized accurately, the number of breathing disorder events all night long is determined by using the classification of snoring and the types of snoring before and after to predict the AHI value, and a data reference is provided for the patient with the OSAHS.

Description

technical field [0001] The invention relates to a sleep snoring sound detection method, in particular to a sleeping snoring sound classification detection method and system, belonging to the field of audio signal processing and technology. Background technique [0002] Obstructive sleep apnea-hypopneasy syndrome (OSAHS) is a common sleep-related breathing disorder, which can easily lead to neurocognitive dysfunction, metabolic dysfunction, cardiovascular disease, respiratory failure and In addition to complications such as pulmonary heart disease, loud and deep snoring at night often disturbs the sleep of companions and affects the sleep quality of others. Due to poor sleep quality, OSAHS patients are prone to fatigue and drowsiness during the day, which greatly increases the possibility of traffic accidents and other accidents , seriously endangering the health and safety of patients and others. [0003] Polysomnography (PSG) is the gold standard for diagnosing OSAHS, but ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G10L25/18G10L25/21G10L25/27G10L25/45G10L25/51G10L25/66
CPCG10L25/18G10L25/21G10L25/27G10L25/45G10L25/51G10L25/66
Inventor 彭健新王璨
Owner SOUTH CHINA UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products