System and method for lung rale recognition
A rale and lung technology, applied in the field of machine learning classification, can solve the problem of low accuracy of lung rale recognition
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
specific Embodiment approach 1
[0026] Specific implementation mode one: as figure 1 shown. A system for pulmonary rale recognition described in this embodiment, the system for pulmonary rale recognition includes an input module, a signal preprocessing and feature extraction module, a neural network module, and an output module;
[0027] The input module is used to input the original breath sound signal to the signal preprocessing and feature extraction module;
[0028] The signal preprocessing and feature extraction module is used to preprocess the original breath sound signal, and calculate the features of the original breath sound signal; input the calculated feature into the neural network module, and the neural network module obtains the recognition result according to the input feature;
[0029] The output module is used to output the recognition result of the neural network module.
[0030] In this embodiment, the original breath sound signal comes from two parts: data collected from volunteers and ...
specific Embodiment approach 2
[0031] Specific embodiment 2: The difference between this embodiment and specific embodiment 1 is that the signal preprocessing and feature extraction module is used to preprocess the original breath sound signal and calculate the features of the original breath sound signal. include:
[0032] Zero-crossing rate before filtering, zero-crossing rate after filtering, average energy before filtering, energy ratio after filtering, average amplitude before filtering, amplitude ratio after filtering, power spectrum energy density before filtering, and power spectrum after filtering of the original breath sound signal Energy density, spectrum energy ratio before and after filtering, and index of the highest energy position in the filtered spectrum.
specific Embodiment approach 3
[0033] Specific embodiment three: the difference between this embodiment and specific embodiment two is: the zero-crossing rate Z of the original breath sound signal before filtering n is calculated as:
[0034]
[0035] Among them, x(n) is the original breath sound signal before filtering, n represents the moment, n=1,2,...,L, x(n-1) is the point at the previous moment of x(n), L is the point before filtering The length of the original breath sound signal, sgn[ ] is a sign function, and the sign function is defined as:
[0036]
[0037] Filter the original breath sound signal x(n) to obtain the filtered signal x'(n);
[0038] The filtering of the original breath sound signal x(n) adopts a Chebyshev band-pass filter, and the Chebyshev band-pass filter H a The expression of (jΩ) is:
[0039]
[0040] Among them, Ω is the frequency of the original breath sound signal, Ω pu is the upper cut-off frequency of the passband, Ω pl is the cut-off frequency under the passb...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More