A method, device and product for monitoring the health status of cetaceans based on breathing sound characteristics

Abstract

The present invention discloses a method, device and product for monitoring the health status of cetaceans based on breathing sound characteristics, which relates to the technical field of health status monitoring. The method comprises obtaining a standard sample library and determining the attributes of the sample; obtaining the breathing signal of the sample in the standard sample library and extracting the feature of the breathing signal; recording and extracting the feature of the breathing signal of the sample to be evaluated; comparing the feature extracted from the breathing signal of the sample to be evaluated with the feature extracted from the breathing signal of the healthy sample and the diseased sample in the standard sample library; and determining the health status of the sample to be evaluated according to the comparison result. The present invention has the characteristics of non-destructive monitoring and strong universality.

Description

Technical Field

[0001] The present invention relates to the technical field of health status monitoring, and in particular to a method, device and product for monitoring the health status of cetaceans based on breathing sound characteristics. Background Art

[0002] Cetaceans are top creatures in the ecosystem and play a vital role. They not only play a key role in the ecosystem, but also play an important role in the global carbon cycle. They have great ecological and tourism value. Compared with artificial sonar, the sonar system of cetaceans is far ahead in terms of accuracy and advancement. Therefore, they have important bionic value.

[0003] However, with global climate change and intensified human activities, cetaceans are increasingly disturbed and threatened. Globally, the number of cetaceans is declining at an alarming rate. For example, in the North Pacific, the population of large whales has dropped by 86% from the mid-1990s to the early 21st century. Therefore, the protection of cetaceans has become more and more urgent, and protection measures have become more and more stringent.

[0004] How to evaluate the health status of cetaceans in the wild and in captivity in oceanariums is an important basis for their effective protection.

[0005] Currently, the evaluation of the health status of cetaceans is usually based on (1) physiological indicators [monitoring physiological parameters such as weight, body length, heart rate, etc.], (2) behavioral observation [observing the activity level, swimming pattern, social behavior, etc. of cetaceans] and (3) blood and tissue sample analysis [assessing hormone levels, disease markers, nutritional status, etc. by collecting blood, skin or fecal samples]. In addition, a method for non-invasively evaluating the health status of cetaceans has not yet been fully developed. Therefore, it is urgent to provide a method and system for monitoring the health status of cetaceans based on respiratory sound feature analysis to solve the above problems. Summary of the invention

[0006] The purpose of the present invention is to provide a method, device and product for monitoring the health status of cetaceans based on breathing sound characteristics, which has the characteristics of non-destructive monitoring and strong universality.

[0007] To achieve the above object, the present invention provides the following solutions:

[0008] A method for monitoring the health status of cetaceans based on breathing sound characteristics, the method comprising:

[0009] Obtain a standard sample library and determine the attributes of the samples; the standard sample library includes: a set number of healthy samples and diseased samples; the attributes of the samples include: species, age group, gender and weight of the samples;

[0010] Obtain the breathing signal of the sample in the standard sample library, and extract the features of the breathing signal; the extracted features include: time feature parameters, sound intensity feature parameters and frequency feature parameters;

[0011] Record the respiratory signals and extract the features of the samples to be evaluated;

[0012] Compare the features extracted from the respiratory signals of the sample to be evaluated with the features extracted from the respiratory signals of healthy samples and diseased samples in the standard sample library;

[0013] The health status of the sample to be evaluated is determined based on the comparison results.

[0014] Optionally, obtaining a respiratory signal of a sample in a standard sample library specifically includes:

[0015] The sample respiratory signal is obtained by using a sample respiratory signal recording and storage module; the sample respiratory signal recording and storage module includes: a data acquisition device, a signal processor, a data recorder, a storage medium and a data management system;

[0016] Wherein, the data acquisition device is a device for collecting breathing signals of cetaceans;

[0017] The signal processor is used to process the acquired respiratory signal; the processing includes: signal filtering, amplification, and conversion between analog signal and digital signal;

[0018] The data recorder is used to record the processed respiratory signal;

[0019] The storage medium is used to store the recorded respiratory signals.

[0020] Optionally, feature extraction is performed on the respiratory signal, specifically including:

[0021] Extracting time characteristic parameters according to the time domain signal in the respiratory signal; the time characteristic parameters include: signal duration, phonation interval and signal rhythm;

[0022] Extracting sound intensity characteristic parameters according to the time domain signal in the breathing signal; the sound intensity characteristic parameters include the loudness of the signal;

[0023] Fourier transform is performed on the time domain signal in the breathing signal, and frequency characteristic parameters are extracted; the frequency characteristic parameters include: timbre and pitch.

[0024] Optionally, comparing the features extracted from the respiratory signal of the sample to be evaluated with the features extracted from the respiratory signals of healthy samples and diseased samples in the standard sample library, specifically includes:

[0025] Determine a first correlation coefficient between the feature extracted from the respiratory signal of each healthy sample and the feature extracted from the respiratory signal of the sample to be evaluated;

[0026] Determine a second correlation coefficient between the feature extracted from the respiratory signal of each diseased sample and the feature extracted from the respiratory signal of the sample to be evaluated; the first correlation coefficient and the second correlation coefficient both include: a correlation coefficient of a time characteristic parameter, a correlation coefficient of a sound intensity characteristic parameter, and a correlation coefficient of a frequency characteristic parameter;

[0027] Determine the magnitude of the first correlation coefficient and the second correlation coefficient.

[0028] Optionally, the step of comparing the features extracted from the respiratory signals of the sample to be evaluated with the features extracted from the respiratory signals of healthy samples and diseased samples in a standard sample library specifically includes: a machine learning module; the machine learning module includes: a support vector machine, a k-nearest neighbor algorithm, and a neural network pattern recognition model.

[0029] A computer device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a method for monitoring the health status of cetaceans based on breathing sound characteristics.

[0030] Optionally, the memory is a computer-readable storage medium.

[0031] A computer program product includes a computer program, which, when executed by a processor, implements a method for monitoring the health status of cetaceans based on breathing sound characteristics.

[0032] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0033] The present invention provides a method, device and product for monitoring the health status of cetaceans based on breathing sound characteristics. The method, device and product can monitor the breathing signals of cetaceans to study the health status of cetaceans. The method is applicable to all types of cetaceans, including toothed whales and baleen whales, and has good universality. The method of monitoring the breathing signals of cetaceans to study the health status of cetaceans is a non-destructive monitoring technology, which has stronger biological safety for aquatic organisms in water bodies. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0035] Figure 1 A schematic flow chart of a method for monitoring the health status of cetaceans based on breathing sound characteristics provided in Example 1 of the present invention;

[0036] Figure 2 This is a schematic diagram of the overall process of a method for monitoring the health status of cetaceans based on breathing sound characteristics provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] The purpose of the present invention is to provide a method, device and product for monitoring the health status of cetaceans based on breathing sound characteristics, which has the characteristics of non-destructive monitoring and strong universality.

[0039] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0040] Example 1

[0041] like Figure 1 and Figure 2 As shown, the present invention provides a method for monitoring the health status of cetaceans based on breathing sound characteristics, the method comprising:

[0042] S101, obtain a standard sample library and determine the attributes of the sample; the standard sample library includes: a set number of healthy samples and diseased samples; that is, the standard sample library includes a healthy sample library H i (i = 1, 2, ..., N; N represents the total number of individuals in the healthy sample library) and the diseased sample library S i (i = 1, 2, ..., M; M represents the total number of individuals in the diseased sample library), healthy cetacean individuals are selected under the conditions of natural populations in the wild and artificial breeding environments to form a healthy sample library H i (i = 1, 2, ..., N; N represents the total number of individuals in the healthy sample library), select sick cetaceans to form the sick sample library S i (i=1, 2, ..., M; M represents the total number of individuals in the diseased sample library) to carry out the research.

[0043] Health Sample Database H iThe species, age group, gender and weight attributes of each individual in the healthy sample library (i = 1, 2, ..., N, N represents the total number of individuals in the healthy sample library) are marked, and then the attribute library H of each sample individual in the healthy sample library is obtained. i P j (i = 1, 2, ..., N; j = 1, 2, ..., J; N represents the total number of individuals in the healthy sample library, and J represents the total number of different types of attributes in the attribute library). i The species, age group, gender and weight attributes of each individual in the diseased sample library are marked, and then the attribute library S of each sample individual in the diseased sample library is obtained. i P j (i=1, 2, ..., M; j=1, 2, ..., J; M represents the total number of individuals in the diseased sample library, and J represents the total number of different types of attributes in the attribute library). The construction of the standard sample attribute library is achieved through the above steps.

[0044] S102, obtaining the breathing signal of the sample in the standard sample library, and extracting the features of the breathing signal; the extracted features include: time feature parameters, sound intensity feature parameters and frequency feature parameters;

[0045] S102 specifically includes:

[0046] The sample respiratory signal is obtained by using a sample respiratory signal recording and storage module; the sample respiratory signal recording and storage module includes: a data acquisition device, a signal processor, a data recorder, a storage medium and a data management system;

[0047] The data acquisition device is a device used to collect breathing signals of cetaceans, such as a microphone sensor.

[0048] The signal processor is used to process the acquired breathing signal; the processing includes: signal filtering, amplification, and conversion between analog signals and digital signals; the sound signal collected by the microphone sensor is filtered, amplified, and converted between analog signals and digital signals through the signal processor in turn, and other signal modulations are performed.

[0049] The data recorder is used to record the processed respiratory signal; the data recorder includes a data acquisition card and a storage device, and a software application, such as data acquisition software.

[0050] Storage media is used to store the recorded respiratory signals. For example, hard drive, solid state drive, cloud storage, etc. This ensures that the data can be preserved for a long time for subsequent analysis and review.

[0051] Data management system: A system that provides management and access to respiratory signal data. It includes a database system, a file system, and specialized data management software for organizing, querying, and retrieving stored respiratory signal data. The database system is archived in different file systems, and different file systems are read and called through the data management system.

[0052] User interface: A user interface for interacting with the respiratory signal recording and storage module. This can be a desktop application, web interface, mobile application, etc., providing data visualization, query, export and other functions.

[0053] Through the health sample library H i The respiratory signals of each individual in the sample are recorded and stored to obtain the healthy sample library H i Recording and storing data of breathing signals of each individual in A u H i (i=1, 2, ..., N; N represents the total number of individuals in the healthy sample library, u=1, 2, ..., U; U represents the number of individuals H in the healthy sample library i The total number of recorded respiratory signals).

[0054] By using the disease sample library S i The respiratory signals of each individual in the sample are recorded and stored to obtain the diseased sample library S i Recording and storing data of breathing signals of each individual in A v S i (i=1,2,…,M;M represents the total number of individuals in the diseased sample library, v=1,2,…,V;V represents the number of individuals in the diseased sample library i The total number of recorded respiratory signals).

[0055] S103, recording and extracting features of the respiratory signals of the samples to be evaluated, that is, constructing a standard sample respiratory signal database through a standard sample signal feature parameter extraction module.

[0056] By recording and storing the respiratory signals A of each individual in the healthy sample library u H i The time domain signal A u H i (t) Extract time characteristic parameters and obtain T j A u H i (i=1, 2, ..., N; N represents the total number of individuals in the healthy sample library, u=1, 2, ..., U; U represents the number of individuals H in the healthy sample library i The total number of recorded respiratory signals, j = 1, 2, ..., J; J represents the total number of time characteristic parameters).

[0057] Time characteristic parameters include signal duration, sound interval and signal rhythm. Signal duration refers to the length of time a signal persists or occurs. Signal duration is one of the important parameters that describe the duration of a signal.

[0058] The calculation formula for signal duration is: (i=1, 2, ..., N; N represents the total number of sound signals).

[0059] represents the signal duration of the i-th signal, represents the starting time of the i-th signal, Represents the end time of the i-th signal.

[0060] The inter-sound interval refers to the time interval between two consecutive sounds in a series of sounds or signals.

[0061] The calculation formula of signal interval is: (i=1, 2, ..., N-1; N represents the total number of sound signals).

[0062] represents the time interval between the i-th signal and the i+1-th signal, represents the starting time of the i+1th signal, Represents the starting time of the i-th signal.

[0063] Signal rhythm refers to the repetitive pattern or rhythm of a signal over time.

[0064] The calculation formula of signal rhythm is:

[0065] (k=1, 2, ..., K; K represents the total number of individuals, i=1, 2, ..., N-1; N represents the total number of all sound signals recorded by the Kth individual).

[0066] represents the signal rhythm of the kth individual, represents the starting time of the i+1th signal of the kth individual, Represents the starting time point of the i-th signal of the k-th individual.

[0067] By recording and storing the respiratory signals A of each individual in the healthy sample library u H i The time domain signal A u H i (t) Extract the sound intensity characteristic parameters and obtain P q A u H i(i=1, 2, ..., N; N represents the total number of individuals in the healthy sample library, u=1, 2, ..., U; U represents the number of individuals H in the healthy sample library i The total number of recorded breathing signals, q=1, 2, ..., Q; Q represents the total number of sound intensity characteristic parameters).

[0068] Sound intensity characteristic parameters include the loudness of the signal. Loudness is the strength or volume of sound, which is another basic attribute of sound. Loudness is related to the amplitude of the sound wave. The larger the amplitude, the louder the sound. The loudness of the signal mainly includes the peak sound pressure level and the root mean square sound pressure level of the signal.

[0069] The peak sound pressure level of the signal is calculated as:

[0070]

[0071] SPLp represents the peak sound pressure level of the signal, max(P) represents the peak sound pressure of the signal, in Pascal (Pa), and P0 is the reference sound pressure, which is usually taken as 20×10 -6 Pa.

[0072] The formula for calculating the root mean square sound pressure level of the signal is:

[0073]

[0074] SPL rms Represents the root mean square sound pressure level of the signal, p i Represents the sound pressure of the signal in Pascal (Pa), and N represents the total number of samples contained in the signal.

[0075] By recording and storing the respiratory signals A of each individual in the healthy sample library u H i The time domain signal A u H i (t) Perform Fourier transform to obtain the Fourier transformed signal FA u H i (ω), where the mathematical expression of Fourier transform is as follows:

[0076]

[0077] Among them, FA u H i (ω) is the signal A u H i (t) is represented in the frequency domain, ω is the frequency, i is the imaginary unit, and e is the base of the natural logarithm. u H i (ω) extract frequency characteristic parameters and obtain F k Au H i (i=1, 2, ..., N; N represents the total number of individuals in the healthy sample library, u=1, 2, ..., U; U represents the number of individuals H in the healthy sample library i The total number of recorded respiratory signals, k = 1, 2, ..., K; K represents the total number of frequency characteristic parameters).

[0078] Frequency characteristic parameters include the timbre and pitch of the signal. The timbre is determined by the spectral components and other acoustic characteristics of the sound. The timbre determines whether a sound is clear, soft, sharp or dull, and whether its sound quality is warm or cold, etc. Pitch is the high and low frequencies of the sound, which is one of the basic properties of the sound. Pitch refers to the pitch of the sound, that is, the high or low pitch of the sound. The pitch is determined by the frequency of the sound wave vibration. The higher the frequency, the higher the pitch, and the lower the frequency, the lower the pitch.

[0079] By recording and storing the respiratory signals A of each individual in the diseased sample library v S i The time domain signal A v S i (t) Extract time characteristic parameters and obtain T j A v S i (i=1,2,…,M;M represents the total number of individuals in the diseased sample library, v=1,2,…,V;V represents the number of individuals in the diseased sample library i The total number of recorded respiratory signals, j = 1, 2, ..., J; J represents the total number of time characteristic parameters).

[0080] Time characteristic parameters include signal duration, sound interval and signal rhythm. Signal duration refers to the length of time a signal persists or occurs. Signal duration is one of the important parameters to describe the duration of a signal. Sound interval refers to the time interval between two consecutive sounds in a series of sounds or signals. Signal rhythm refers to the repetitive pattern or rhythm of a signal in time.

[0081] By recording and storing the respiratory signals A of each individual in the diseased sample library v S i The time domain signal A v S i (t) Extract the sound intensity characteristic parameters and obtain P q A v S i (i=1,2,…,M;M represents the total number of individuals in the diseased sample library, v=1,2,…,V;V represents the number of individuals in the diseased sample library i The total number of recorded breathing signals, q=1, 2, ..., Q; Q represents the total number of sound intensity characteristic parameters).

[0082] Sound intensity characteristic parameters include the loudness of the signal. Loudness is the strength or volume of sound, which is another basic attribute of sound. Loudness is related to the amplitude of the sound wave. The larger the amplitude, the louder the sound. The loudness of the signal mainly includes the peak sound pressure level and the root mean square sound pressure level of the signal.

[0083] The peak sound pressure level of the signal is calculated as:

[0084]

[0085] SPL p Represents the peak sound pressure level of the signal, max(P) represents the peak sound pressure of the signal, in Pascal (Pa), and P0 is the reference sound pressure, usually taken as 20×10 -6 Pa.

[0086] The formula for calculating the root mean square sound pressure level of the signal is:

[0087]

[0088] SPL rms Represents the root mean square sound pressure level of the signal, p i Represents the sound pressure of the signal in Pascal (Pa), and N represents the total number of samples contained in the signal.

[0089] By recording and storing the respiratory signals A of each individual in the disease sample library v S i The time domain signal A v S i (t) Perform Fourier transform to obtain the Fourier transformed signal FA v S i (ω), where the mathematical expression of Fourier transform is as follows:

[0090]

[0091] Among them, FA v S i (ω) is the signal A v S i (t) is represented in the frequency domain, ω is the frequency, i is the imaginary unit, and e is the base of the natural logarithm. v S i (ω) extract frequency characteristic parameters and obtain F k A v S i (i=1,2,…,M;M represents the total number of individuals in the diseased sample library, v=1,2,…,V;V represents the number of individuals in the diseased sample library iThe total number of recorded respiratory signals, k = 1, 2, ..., K; K represents the total number of frequency characteristic parameters).

[0092] Frequency characteristic parameters include the timbre and pitch of the signal. The timbre is determined by the spectral components and other acoustic characteristics of the sound. The timbre determines whether a sound is clear, soft, sharp or dull, and whether its sound quality is warm or cold, etc. Pitch is the high and low frequencies of the sound, which is one of the basic properties of the sound. Pitch refers to the pitch of the sound, that is, the high or low pitch of the sound. The pitch is determined by the frequency of the sound wave vibration. The higher the frequency, the higher the pitch, and the lower the frequency, the lower the pitch.

[0093] A standard sample respiratory signal database is constructed according to the extracted features, and the standard sample respiratory signal database includes a healthy sample respiratory signal database and a diseased sample respiratory signal database.

[0094] The healthy sample respiratory signal database consists of the attribute library H of each sample individual in the healthy sample library i P j (i = 1, 2, ..., N; j = 1, 2, ..., J; N represents the total number of individuals in the healthy sample library, Q represents the total number of different types of attributes in the attribute library), time characteristic parameter T j A u H i (i=1, 2, ..., N; N represents the total number of individuals in the healthy sample library, u=1, 2, ..., U; U represents the number of individuals H in the healthy sample library i The total number of recorded breathing signals, j = 1, 2, ..., J; J represents the total number of time characteristic parameters), sound intensity characteristic parameter P q A u H i (i=1, 2, ..., N; N represents the total number of individuals in the healthy sample library, u=1, 2, ..., U; U represents the number of individuals H in the healthy sample library i The total number of recorded breathing signals, q = 1, 2, ..., Q; Q represents the total number of sound intensity characteristic parameters) and frequency characteristic parameter F k A u H i (i=1, 2, ..., N; N represents the total number of individuals in the healthy sample library, u=1, 2, ..., U; U represents the number of individuals H in the healthy sample library i The total number of recorded respiratory signals, k = 1, 2, ..., K; K represents the total number of frequency characteristic parameters).

[0095] The diseased sample respiratory signal database consists of the attribute library S of each sample individual in the diseased sample library i P j(i = 1, 2, ..., M; j = 1, 2, ..., Q; M represents the total number of individuals in the diseased sample library, Q represents the total number of different types of attributes in the attribute library), time characteristic parameter T j A v S i (i=1,2,…,M;M represents the total number of individuals in the diseased sample library, v=1,2,…,V;V represents the number of individuals in the diseased sample library i The total number of recorded breathing signals, j = 1, 2, ..., J; J represents the total number of time characteristic parameters), sound intensity characteristic parameter P q A v S i (i=1,2,…,M;M represents the total number of individuals in the diseased sample library, v=1,2,…,V;V represents the number of individuals in the diseased sample library i The total number of recorded breathing signals, q = 1, 2, ..., Q; Q represents the total number of sound intensity characteristic parameters) and frequency characteristic parameter F k A v S i (i=1,2,…,M;M represents the total number of individuals in the diseased sample library, v=1,2,…,V;V represents the number of individuals in the diseased sample library i The total number of recorded respiratory signals, k = 1, 2, ..., K; K represents the total number of frequency characteristic parameters).

[0096] S104, comparing the features extracted from the respiratory signal of the sample to be evaluated with the features extracted from the respiratory signals of healthy samples and diseased samples in the standard sample library;

[0097] First, build a sample library to be evaluated. The sample library to be evaluated consists of the sample E i (i = 1, 2, ..., O; O represents the total number of individuals in the sample library to be evaluated), and the cetacean individuals to be evaluated are selected under the conditions of natural populations in the wild and artificial breeding environments to form the sample library E to be evaluated. i (i=1, 2, ..., O; O represents the total number of individuals in the sample library to be evaluated) to carry out the research.

[0098] Second, construct a sample attribute library to be evaluated;

[0099] The attributes of the samples to be evaluated include the species, age group, gender and weight attributes of each sample in the sample library to be evaluated. i The species, age group, gender and weight attributes of each individual in the sample library to be evaluated are marked, and then the attribute library E of each sample individual in the sample library to be evaluated is obtained. i P j(i=1, 2, ..., O; j=1, 2, ..., J; O represents the total number of individuals in the healthy sample library, and J represents the total number of different types of attributes in the attribute library).

[0100] Third, construct a respiratory signal recording and storage module for the samples to be evaluated;

[0101] The respiratory signal recording and storage module of the sample to be evaluated includes: a data acquisition device, a signal processor, a data recorder, a storage medium and a data management system.

[0102] By treating the evaluation sample library E i The respiratory signals of each individual in the sample library E are recorded and stored to obtain the sample library E to be evaluated. i Recording and storing data of breathing signals of each individual in A w E i (i=1,2,…,O;O represents the total number of individuals in the sample library to be evaluated, w=1,2,…,W;W represents the number of individuals in the sample library to be evaluated E i The total number of recorded respiratory signals).

[0103] Fourth, a module for extracting characteristic parameters of the sample signal to be evaluated;

[0104] The respiratory signal A recorded and stored by each individual in the sample library to be evaluated w E i The time domain signal A w E i (t) Extract time characteristic parameters and obtain T j A w E i (i=1,2,…,O;O represents the total number of individuals in the sample library to be evaluated, w=1,2,…,W;W represents the number of individuals in the sample library to be evaluated E i The total number of recorded respiratory signals, j = 1, 2, ..., J; J represents the total number of time characteristic parameters).

[0105] By recording and storing the respiratory signals A of each individual in the sample library to be evaluated w E i The time domain signal A w E i (t) Extract the characteristic parameters of sound intensity and obtain P q A w E i (i=1,2,…,O;O represents the total number of individuals in the sample library to be evaluated, w=1,2,…,W;W represents the number of individuals in the sample library to be evaluated E i The total number of recorded breathing signals, q=1, 2, ..., Q; Q represents the total number of sound intensity characteristic parameters).

[0106] The peak sound pressure level of the signal is calculated as:

[0107]

[0108] SPL p Represents the peak sound pressure level of the signal, max(P) represents the peak sound pressure of the signal, in Pascal (Pa), and P0 is the reference sound pressure, usually taken as 20×10 -6 Pa.

[0109] The formula for calculating the root mean square sound pressure level of the signal is:

[0110]

[0111] SPL rms Represents the root mean square sound pressure level of the signal, p i Represents the sound pressure of the signal in Pascal (Pa), and N represents the total number of samples contained in the signal.

[0112] By recording and storing the respiratory signals A of each individual in the sample library to be evaluated w E i The time domain signal A w E i (t) Perform Fourier transform to obtain the Fourier transformed signal FA w E i (ω), where the mathematical expression of Fourier transform is as follows:

[0113]

[0114] Among them, FA w E i (ω) is the signal A w E i (t) is represented in the frequency domain, ω is the frequency, i is the imaginary unit, and e is the base of the natural logarithm. w E i (ω) extract frequency characteristic parameters and obtain F k A w E i (i=1,2,…,O;O represents the total number of individuals in the sample library to be evaluated, w=1,2,…,W;W represents the number of individuals in the sample library to be evaluated E i The total number of recorded respiratory signals, k = 1, 2, ..., K; K represents the total number of frequency characteristic parameters).

[0115] Fifth, construct a database of respiratory signals of samples to be evaluated;

[0116] The respiratory signal database of the sample to be evaluated consists of the attribute library E of each sample individual in the sample library to be evaluated. i P j (i = 1, 2, ..., O; j = 1, 2, ..., J; O represents the total number of individuals in the sample library to be evaluated, and J represents the total number of different types of attributes in the attribute library), time characteristic parameter T j A w E i (i=1,2,…,O;O represents the total number of individuals in the sample library to be evaluated, w=1,2,…,W;W represents the number of individuals in the sample library to be evaluated E i The total number of recorded breathing signals, j = 1, 2, ..., J; J represents the total number of time characteristic parameters), sound intensity characteristic parameter P q A w E i (i=1,2,…,O;O represents the total number of individuals in the sample library to be evaluated, w=1,2,…,W;W represents the number of individuals in the sample library to be evaluated E i The total number of recorded breathing signals, q = 1, 2, ..., Q; Q represents the total number of sound intensity characteristic parameters) and frequency characteristic parameter F k A w E i (i=1,2,…,O,O represents the total number of individuals in the sample library to be evaluated,w=1,2,…,W;W represents the number of individuals in the sample library to be evaluated E i The total number of recorded respiratory signals, k = 1, 2, ..., K; K represents the total number of frequency characteristic parameters).

[0117] By transforming the attribute library parameter H in the healthy sample breathing signal database i P j , time characteristic parameter T j A u H i , sound intensity characteristic parameter P q A u H i and frequency characteristic parameter F k A u H i Data and attribute library parameters E in the respiratory signal database of the sample to be evaluated i P j , time characteristic parameter T j A w E i , sound intensity characteristic parameter P q A w E i and frequency characteristic parameter F k A w E iStatistical analysis and comparison studies were performed to obtain the correlation coefficient (a) between the healthy sample respiratory signal database and the sample respiratory signal database to be evaluated.

[0118] The formula for calculating the correlation coefficient is:

[0119]

[0120] Where X i and Y i are the observed values ​​of two variables, and is their mean.

[0121] By calculating the time characteristic parameter T in the healthy sample respiratory signal database j A u H i and the time characteristic parameter T in the sample respiratory signal database to be evaluated j A w E i Perform correlation analysis to obtain the correlation coefficient RHE of the time characteristic parameters in the healthy sample respiratory signal database and the time characteristic parameters in the sample respiratory signal database to be evaluated t .

[0122] By calculating the sound intensity characteristic parameter P in the healthy sample breathing signal database q A u H i and the sound intensity characteristic parameter P in the sample breathing signal database to be evaluated q A w E i Perform correlation analysis to obtain the correlation coefficient RHE of the sound intensity characteristic parameters in the healthy sample respiratory signal database and the sound intensity characteristic parameters in the sample respiratory signal database to be evaluated p .

[0123] By calculating the frequency characteristic parameter F in the healthy sample breathing signal database k A u H i and the frequency characteristic parameter F in the sample respiratory signal database to be evaluated k A w E i Perform correlation analysis to obtain the correlation coefficient RHE of the frequency characteristic parameters in the healthy sample respiratory signal database and the frequency characteristic parameters in the sample respiratory signal database to be evaluated f .

[0124] The calculation formula of the first correlation coefficient a between the healthy sample respiratory signal database and the sample respiratory signal to be evaluated is:

[0125] a=RHEt +RHE p +RHE f ;

[0126] By converting the attribute library parameter S in the respiratory signal database of diseased samples i P j , time characteristic parameter T j A v S i , sound intensity characteristic parameter P q A v S i and frequency characteristic parameter F k A v S i Data and attribute library parameters E in the respiratory signal database of the sample to be evaluated i P j , time characteristic parameter T j A w E i , sound intensity characteristic parameter P q A w E i and frequency characteristic parameter F k A w E i Statistical analysis and comparison research are performed to obtain the first correlation coefficient a between the healthy sample respiratory signal database and the sample respiratory signal database to be evaluated.

[0127] By calculating the time characteristic parameter T in the respiratory signal database of diseased samples j A v S i and the time characteristic parameter T in the sample respiratory signal database to be evaluated j A w E i Perform correlation analysis to obtain the correlation coefficient RSE of the time characteristic parameters in the respiratory signal database of the diseased sample and the time characteristic parameters in the respiratory signal database of the sample to be evaluated t .

[0128] By calculating the sound intensity characteristic parameter P in the respiratory signal database of diseased samples q A v S i and the sound intensity characteristic parameter P in the sample breathing signal database to be evaluated q A w E i Perform correlation analysis to obtain the correlation coefficient RSE of the sound intensity characteristic parameters in the respiratory signal database of the diseased sample and the sound intensity characteristic parameters in the respiratory signal database of the sample to be evaluated p .

[0129] By calculating the frequency characteristic parameter F in the respiratory signal database of diseased samples k A v S i and the frequency characteristic parameter F in the sample respiratory signal database to be evaluated k A w E i Perform correlation analysis to obtain the correlation coefficient RSE of the frequency characteristic parameters in the respiratory signal database of the diseased sample and the frequency characteristic parameters in the respiratory signal database of the sample to be evaluated f .

[0130] The calculation formula of the second correlation coefficient b between the respiratory signal database of the diseased sample and the respiratory signal of the sample to be evaluated is:

[0131] b=RSE t +RSE p +RSE f .

[0132] The comparison module of the standard sample respiratory signal database and the sample respiratory signal database to be evaluated includes a clustering and discriminant analysis module and a machine learning module; the machine learning module includes a support vector machine (SVM), a k-nearest neighbor algorithm (KNN) and a neural network pattern recognition model.

[0133] Cluster analysis is an unsupervised learning method that aims to divide samples in a dataset into groups with similar characteristics, called clusters. In cluster analysis, the similarity between samples is usually measured by distance or similarity metrics, such as Euclidean distance, Manhattan distance, etc. Common clustering algorithms include K-means clustering, hierarchical clustering, and DBSCAN.

[0134] Discriminant analysis is a supervised learning method whose goal is to build a classifier by learning samples of known categories to predict the category of new samples. Discriminant analysis focuses on finding the features or decision boundaries that can best distinguish different categories. Common discriminant analysis methods include linear discriminant analysis (LDA) and logistic regression.

[0135] Support Vector Machine (SVM): Support vector machine is a supervised learning algorithm, commonly used for classification and regression analysis. The goal of SVM is to find an optimal hyperplane that can separate samples of different categories and maximize the interval between the boundary and the nearest sample point.

[0136] K-nearest neighbor algorithm (KNN): K-nearest neighbor algorithm is a simple and intuitive supervised learning algorithm for classification and regression analysis. For classification tasks, given a new sample, the KNN algorithm finds the closest k known samples and classifies the new sample into the most common category among these k samples.

[0137] Neural Network Pattern Recognition: Neural network pattern recognition is a pattern recognition technology based on neural networks. A neural network consists of a hierarchical structure of multiple neurons. During the training process, the mapping relationship between the input and output is learned by adjusting the connection weights between neurons. Neural network pattern recognition can learn complex non-linear relationships and can achieve good classification results when trained sufficiently.

[0138] S105. Determine the health status of the sample to be evaluated according to the comparison result.

[0139] The module for judging the health status of the sample to be evaluated includes dividing the sample to be evaluated into healthy individuals and diseased individuals. By comparing the magnitudes of the first correlation coefficient a between the healthy sample respiratory signal database and the sample to be evaluated respiratory signal database and the second correlation coefficient b between the diseased sample respiratory signal database and the sample to be evaluated respiratory signal database, a judgment of diseased individuals (correlation coefficient: a > b) and healthy individuals (correlation coefficient: a < b) is made on the health status of the sample to be evaluated.

[0140] Embodiment 2

[0141] A computer device includes: a memory, a processor, and a computer program stored on the memory and executable on the processor. The processor executes the computer program to implement a method for monitoring the health status of cetaceans based on respiratory sound characteristics in Embodiment 1.

[0142] Among them, the memory is a computer-readable storage medium.

[0143] Embodiment 3

[0144] A computer program product includes a computer program which, when executed by a processor, implements a method for monitoring the health status of cetaceans based on respiratory sound characteristics in Embodiment 1.

[0145] It should be noted that the object information (including but not limited to object device information, object personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present invention are all information and data authorized by the object or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with the relevant laws, regulations, and standards of relevant countries and regions.

[0146] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided by the present invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided by the present invention may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited thereto. The processor involved in each embodiment provided by the present invention may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited thereto.

[0147] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0148] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for monitoring the health status of cetaceans based on breathing sound characteristics, characterized in that: The method comprises: Obtain a standard sample library and determine the attributes of the samples; the standard sample library includes: a set number of healthy samples and diseased samples; the attributes of the samples include: species, age group, gender and weight of the samples; Obtain the breathing signal of the sample in the standard sample library, and extract the features of the breathing signal; the extracted features include: time feature parameters, sound intensity feature parameters and frequency feature parameters; Record the respiratory signals and extract the features of the samples to be evaluated; Compare the features extracted from the respiratory signals of the sample to be evaluated with the features extracted from the respiratory signals of healthy samples and diseased samples in the standard sample library; Determine the health status of the sample to be evaluated based on the comparison result; The features extracted from the respiratory signals of the sample to be evaluated are compared with the features extracted from the respiratory signals of healthy samples and diseased samples in the standard sample library, specifically including: Determine the first correlation coefficient a between the features extracted from the respiratory signal of each healthy sample and the features extracted from the respiratory signal of the sample to be evaluated; a=RHE t +RHE p +RHE f , RHE t is the correlation coefficient between the time characteristic parameters in the healthy sample respiratory signal database and the time characteristic parameters in the sample respiratory signal database to be evaluated, RHE p is the correlation coefficient between the sound intensity characteristic parameters in the healthy sample respiratory signal database and the sound intensity characteristic parameters in the sample respiratory signal database to be evaluated, RHE f is the correlation coefficient between the frequency characteristic parameters in the healthy sample respiratory signal database and the frequency characteristic parameters in the sample respiratory signal database to be evaluated; Determine a second correlation coefficient b between the feature extracted from the respiratory signal of each diseased sample and the feature extracted from the respiratory signal of the sample to be evaluated; the first correlation coefficient and the second correlation coefficient both include: the correlation coefficient of the time characteristic parameter, the correlation coefficient of the sound intensity characteristic parameter and the correlation coefficient of the frequency characteristic parameter; b=RSE t +RSE p +RSE f , RSE t is the correlation coefficient between the time characteristic parameters in the respiratory signal database of diseased samples and the time characteristic parameters in the respiratory signal database of samples to be evaluated, RSE p is the correlation coefficient between the sound intensity characteristic parameters in the respiratory signal database of the diseased sample and the sound intensity characteristic parameters in the respiratory signal database of the sample to be evaluated, RSE f is the correlation coefficient between the frequency characteristic parameters in the respiratory signal database of diseased samples and the frequency characteristic parameters in the respiratory signal database of samples to be evaluated; Determine the magnitude of the first correlation coefficient and the second correlation coefficient.

2. The method for monitoring the health status of cetaceans based on breathing sound characteristics according to claim 1, characterized in that: Obtain the respiratory signals of samples in the standard sample library, including: The sample respiratory signal is obtained by using a sample respiratory signal recording and storage module; the sample respiratory signal recording and storage module includes: a data acquisition device, a signal processor, a data recorder, a storage medium and a data management system; Wherein, the data acquisition device is a device for collecting breathing signals of cetaceans; The signal processor is used to process the acquired respiratory signal; the processing includes: signal filtering, amplification, and conversion between analog signal and digital signal; The data recorder is used to record the processed respiratory signal; The storage medium is used to store the recorded respiratory signals.

3. The method for monitoring the health status of cetaceans based on breathing sound characteristics according to claim 1, characterized in that: Extract features of respiratory signals, including: Extracting time characteristic parameters according to the time domain signal in the respiratory signal; the time characteristic parameters include: signal duration, phonation interval and signal rhythm; Extracting sound intensity characteristic parameters according to the time domain signal in the breathing signal; the sound intensity characteristic parameters include the loudness of the signal; Fourier transform is performed on the time domain signal in the breathing signal, and frequency characteristic parameters are extracted; the frequency characteristic parameters include: timbre and pitch.

4. The method for monitoring the health status of cetaceans based on breathing sound characteristics according to claim 1, characterized in that: The method of comparing the features extracted from the respiratory signals of the sample to be evaluated with the features extracted from the respiratory signals of healthy samples and diseased samples in the standard sample library specifically includes: a machine learning module; the machine learning module includes: a support vector machine, a k-nearest neighbor algorithm and a neural network pattern recognition model.

5. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a method for monitoring the health status of cetaceans based on breathing sound characteristics as described in any one of claims 1 to 4.

6. A computer device according to claim 5, characterized in that: The memory is a computer-readable storage medium.

7. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for monitoring the health status of cetaceans based on breathing sound characteristics as described in any one of claims 1 to 4 is implemented.

Images