Linear classification algorithm for true and false cough sounds of patients with cervical spinal cord injury medium and equipment

A spinal cord injury and linear classification technology, applied in speech analysis, medical science, instruments, etc., can solve problems such as evaluation, difficulty in establishing a true and false cough recognition model, and limited number of cough sound samples

Active Publication Date: 2020-05-19
SOUTH CHINA UNIV OF TECH
View PDF5 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The current common cough recognition algorithm does not consider the difference between the above two sounds, and cannot be directly applied to the respiratory function evaluation of patients with cervical spinal cord injury
In addition, there is currently no publicly available cough sound sample data set for patients with cervical spinal cord injury, and the number of cough sound samples collected b

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
  • Linear classification algorithm for true and false cough sounds of patients with cervical spinal cord injury medium and equipment
  • Linear classification algorithm for true and false cough sounds of patients with cervical spinal cord injury medium and equipment
  • Linear classification algorithm for true and false cough sounds of patients with cervical spinal cord injury medium and equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0050] This embodiment is a linear classification algorithm for true and false cough sounds of patients with cervical spinal cord injury. A number of unvoiced and voiced samples are obtained, and the initial sequence of the unvoiced and voiced samples is intercepted and filtered; the zero-crossing rate and the maximum Autocorrelation coefficient; establish a linear classifier based on the zero-crossing rate and the maximum autocorrelation coefficient, and train the linear classifier with unvoiced samples and voiced samples as sample sets; identify the cervical spinal cord injury patients through the trained linear classifier True cough and shout-type pseudo-cough.

[0051] Its workflow is as follows figure 1 shown, including the following steps:

[0052] The first step, with f s Collect m unvoiced sound samples and m voiced sound samples for the sampling frequency, and intercept each sample with a length of N and a duration of The sequence of the unvoiced sequence x i (k)...

Embodiment 2

[0102] In this embodiment, a linear classification algorithm for true and false cough sounds of patients with cervical spinal cord injury, its workflow is as follows: Figure 5 As shown, the difference with Embodiment 1 is that in this embodiment, in the first step, f s Collect m unvoiced sound samples and m voiced sound samples for the sampling frequency, respectively perform endpoint detection processing on each unvoiced sound sample and voiced sound sample, and then intercept the length N from the initial moment after the endpoint detection, and the duration is The sequence of the unvoiced sequence x i (k) and voiced sequence y i (k), wherein, i=1,2,...,m, k=1,2,...,N, and m, N are positive integers;

[0103] Ninth step, with f s Obtain the sample to be tested for the sampling frequency, perform endpoint detection processing on the sample to be tested, and then intercept the length of N from the initial moment after the endpoint detection of the sample to be tested, and...

Embodiment 3

[0106] This embodiment is a storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the processor executes the cervical spinal cord injury patient's true and false described in Embodiment 1 or Embodiment 2. A linear classification algorithm for cough sounds.

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 provides a linear classification algorithm for true and false coughs of patients with cervical spinal cord injury. The algorithm is characterized by comprising the following steps: acquiring a plurality of unvoiced sound samples and voiced sound samples, intercepting initial segment sequences of the unvoiced sound samples and voiced sound samples, and filtering the initial segmentsequences; calculating a zero-crossing rate and a maximum autocorrelation coefficient of the obtained sequences; establishing a linear classifier by taking the zero-crossing rate and the maximum autocorrelation coefficient as characteristics, and training a linear classifier by taking the unvoiced sound samples and the voiced sound samples as a sample set; and identifying the true cough and the shouting type pseudocough of a patient with the cervical spinal cord injury through the trained linear classifier. According to the algorithm, the difficulty that a recognition model cannot be trained due to insufficient cough sound training samples of patients with cervical spinal cord injury is eliminated; the true and false coughs can be recognized only by utilizing unvoiced sound and voiced sound signals, and the interference of the false cough of a shouting sound type on the analysis of the cough intensity is eliminated; meanwhile, the method is simple in model, small in calculation amountand convenient to implement in wearable equipment.

Description

technical field [0001] The invention relates to the field of medical equipment and medical signal processing, and more specifically relates to a linear classification algorithm, medium and equipment for the true and false cough sounds of patients with cervical spinal cord injury. Background technique [0002] Cervical spinal cord injury has a high mortality rate, high disability rate, and many complications. Studies have shown that the quality of respiratory function often determines the patient's recovery and survival rate, and also determines whether to provide patients with assisted breathing devices. Cough is the original protective reflex of human beings. It is the perfect product of respiratory muscles and nervous system. It is also a concentrated expression of lung reserve function. The strength of cough sound can reflect the quality of respiratory function to a certain extent. [0003] However, some patients with cervical spinal cord injury cannot cough normally due...

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): G10L25/51G10L25/66G10L25/06G10L25/09A61B5/00A61B5/08
CPCG10L25/51G10L25/66G10L25/06G10L25/09A61B5/0823A61B5/7267
Inventor 莫鸿强章臻范潇田翔
Owner SOUTH CHINA UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap