CN115191990A - Cough detection method and system based on wearable device - Google Patents
Cough detection method and system based on wearable device Download PDFInfo
- Publication number
- CN115191990A CN115191990A CN202210861062.6A CN202210861062A CN115191990A CN 115191990 A CN115191990 A CN 115191990A CN 202210861062 A CN202210861062 A CN 202210861062A CN 115191990 A CN115191990 A CN 115191990A
- Authority
- CN
- China
- Prior art keywords
- cough
- respiratory
- data
- characteristic
- data set
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/08—Detecting, measuring or recording devices for evaluating the respiratory organs
- A61B5/0823—Detecting or evaluating cough events
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
- A61B5/6802—Sensor mounted on worn items
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Biophysics (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Pathology (AREA)
- Pulmonology (AREA)
- Physiology (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention provides a cough detection method and a system based on wearable equipment, which belong to the field of artificial intelligence design and comprise the following steps: acquiring basic information of a user to be monitored, including medical record data and sign record data, inputting the basic information into a respiratory cough characteristic evaluation model, and generating a respiratory cough characteristic reference data set; uploading a breathing cough monitoring data sequence through a pressure sensor arranged at a preset position on a user to be monitored, and extracting a breathing cough characteristic comparison data set; judging whether the breathing cough characteristic contrast data set meets a breathing cough characteristic reference data set or not; if the breathing cough characteristic contrast data set does not meet the breathing cough characteristic reference data set, generating a breathing cough abnormal signal; and after identifying the respiratory cough monitoring data sequence according to the respiratory cough abnormal signal, sending the respiratory cough monitoring data sequence to a display terminal for displaying. The technical problem that in the prior art, due to the fact that cough data are not processed and screened, reference values are too low is solved.
Description
Technical Field
The invention relates to the technical field of artificial intelligence correlation, in particular to a cough detection method and system based on wearable equipment.
Background
Respiratory diseases have a high incidence rate, so the respiratory diseases have a great threat to human health, the multiple symptoms of the respiratory diseases can cause the patient to cough, the reasons for causing the patient to cough are various, different patient states correspond to different cough characteristics, but since nonprofessionals are difficult to capture the cough characteristics to judge the state of the patient, and medical staff cannot observe the patient for a long time, a scheme for assisting the medical staff to monitor the cough of the patient is urgently needed.
Only through the cough data to the patient in the present technique monitor and show, send medical personnel to handle, but this type of data is comparatively tedious, does not handle the screening to cough data, and the reference value is lower when leading to medical personnel to look over.
In the prior art, cough data is not processed and screened, so that the technical problem of low reference value exists.
Disclosure of Invention
The application provides a cough detection method and system based on wearable equipment, and solves the technical problem that reference value is too low due to the fact that cough data are not processed and screened in the prior art.
In view of the foregoing problems, embodiments of the present application provide a wearable device-based cough detection method and system.
In a first aspect, the present application provides a wearable device-based cough detection method, where the method applies a wearable device-based cough detection system, the system includes a display terminal, the system is in communication connection with a pressure sensor, and the method includes: acquiring basic information of a user to be monitored, wherein the basic information of the user to be monitored comprises medical record data and physical sign record data; inputting the medical record data and the physical sign record data into a respiratory cough characteristic evaluation model to generate a respiratory cough characteristic reference data set; uploading a breathing cough monitoring data sequence through a pressure sensor deployed at a preset position on the user to be monitored; extracting a breathing cough characteristic comparison data set according to the breathing cough monitoring data sequence; determining whether the respiratory cough feature contrast data set satisfies the respiratory cough feature baseline data set; if the breathing cough characteristic contrast data set does not meet the breathing cough characteristic reference data set, generating a breathing cough abnormal signal; and identifying the respiratory cough monitoring data sequence according to the respiratory cough abnormal signal, and then sending the respiratory cough monitoring data sequence to a display terminal for displaying.
In another aspect, the present application provides a cough detection system based on wearable device, wherein, the system includes a display terminal, the system and pressure sensor communication connection, the system includes: the system comprises a user information loading module, a monitoring module and a monitoring module, wherein the user information loading module is used for acquiring basic information of a user to be monitored, and the basic information of the user to be monitored comprises medical record data and sign record data; the characteristic reference data generation module is used for inputting the medical record data and the physical sign record data into a respiratory cough characteristic evaluation model to generate a respiratory cough characteristic reference data set; the monitoring data uploading module is used for uploading a breathing cough monitoring data sequence through a pressure sensor arranged at a preset position on the body of the user to be monitored; the characteristic comparison data extraction module is used for extracting a respiratory cough characteristic comparison data set according to the respiratory cough monitoring data sequence; the characteristic data comparison module is used for judging whether the breathing cough characteristic comparison data set meets the breathing cough characteristic reference data set or not; an abnormal signal generating module, configured to generate a breathing cough abnormal signal if the breathing cough feature contrast data set does not satisfy the breathing cough feature reference data set; and the abnormity identification module is used for identifying the respiratory cough monitoring data sequence according to the respiratory cough abnormity signal and then sending the respiratory cough monitoring data sequence to a display terminal for displaying.
One or more technical solutions provided in the present application have at least the following technical effects or advantages:
because the basic information uploaded to the user to be monitored, namely the patient, is adopted: medical record and physical sign record; using an intelligent model: the breathing cough characteristic evaluation model determines a breathing cough characteristic reference data set in a normal state, which has a higher degree of engagement with a user to be monitored, according to the medical record and the physical record; monitoring a breath cough characteristic contrast data set through a pressure sensor; the reference data set and the comparison data set are compared, after abnormal identification is carried out on the breathing cough monitoring data sequence of which the breathing cough characteristic comparison data set does not meet the breathing cough characteristic reference data set, the whole breathing cough monitoring data sequence is sent to a display terminal for a professional to check, the breathing cough characteristic reference data set which is in agreement with a user to be monitored is evaluated through an intelligent model and then compared with monitored real-time data, abnormal data are identified, the display terminal is highlighted later, the professional can check the breathing cough monitoring data in a key mode conveniently, and the technical effect of improving the reference of the breathing cough monitoring data is achieved.
The foregoing description is only an overview of the technical solutions of the present application, and the present application can be implemented according to the content of the description in order to make the technical means of the present application more clearly understood, and the following detailed description of the present application is given in order to make the above and other objects, features, and advantages of the present application more clearly understandable.
Drawings
Fig. 1 is a schematic flow chart of a cough detection method based on a wearable device according to an embodiment of the present application;
fig. 2 is a schematic flow chart of determination of a reference data set of a cough due to breathing in a wearable device-based cough detection method according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of a wearable device-based cough detection system according to an embodiment of the present application;
fig. 4 is a diagram of possible styles of a wearable device in a wearable device-based cough detection method according to an embodiment of the present application;
description of the reference numerals: the system comprises a display terminal 001, a pressure sensor 002, a user information loading module 11, a characteristic reference data generating module 12, a monitoring data uploading module 13, a characteristic comparison data extracting module 14, a characteristic data comparing module 15, an abnormal signal generating module 16 and an abnormal identification module 17.
Detailed Description
The embodiment of the application provides a cough detection method and system based on wearable equipment, and solves the technical problem that in the prior art, reference value is too low due to the fact that cough data are not processed and screened. The breathing cough characteristic standard data set which is fit for the user to be monitored is evaluated through the intelligent model and is compared with the monitored real-time data, abnormal data are identified, the display terminal is highlighted in the next step of the process, professional staff can check the abnormal data in a key mode, and the technical effect of improving the reference of the breathing cough monitoring data is achieved.
Summary of the application
At present, the mode of monitoring the respiratory cough is mainly to transmit real-time monitored data to medical staff for judgment, and due to the lack of pretreatment on the monitored data, the uploaded data have high redundancy and difficulty in realizing the improvement of the reference value of the monitored data, the difficulty lies in that the physical states of different users correspond to the respiratory cough characteristics in different normal states, so that the monitored respiratory cough characteristics cannot be differentially processed, and the technical problem of low reference value is caused.
In view of the above technical problems, the technical solution provided by the present application has the following general idea:
the embodiment of the application provides a cough detection method and a cough detection system based on wearable equipment. Due to the adoption of uploading the basic information of the user to be monitored, namely the patient: medical record and physical record; using an intelligent model: the respiratory cough characteristic evaluation model determines a respiratory cough characteristic reference data set in a normal state, which has higher fitness with a user to be monitored, according to the medical record and the physical record; monitoring a breath cough characteristic contrast data set through a pressure sensor; the reference data set and the comparison data set are compared, after abnormal identification is carried out on the breathing cough monitoring data sequence of which the breathing cough characteristic comparison data set does not meet the breathing cough characteristic reference data set, the whole breathing cough monitoring data sequence is sent to a display terminal for a professional to check, the breathing cough characteristic reference data set which is in agreement with a user to be monitored is evaluated through an intelligent model and then compared with monitored real-time data, abnormal data are identified, the display terminal is highlighted later, the professional can check the breathing cough monitoring data in a key mode conveniently, and the technical effect of improving the reference of the breathing cough monitoring data is achieved.
Having described the principles of the present application, various non-limiting embodiments thereof will now be described in detail with reference to the accompanying drawings.
Example one
As shown in fig. 1, an embodiment of the present application provides a wearable device-based cough detection method, where the method employs a wearable device-based cough detection system, the system includes a display terminal, the system is in communication connection with a pressure sensor, and the method includes the steps of:
s100: acquiring basic information of a user to be monitored, wherein the basic information of the user to be monitored comprises medical record data and physical sign record data;
in particular, the user to be monitored is a user who needs to perform respiratory cough monitoring, including but not limited to: users in the form of patients, health personnel, etc.; the basic information of the user to be monitored is a data set representing the physical state of the user to be monitored, and the data set comprises the following components: medical record data and physical sign record data, wherein the medical record data includes but is not limited to: sex, age, respiratory disease recording data: type of disease, length of illness, etc.; vital signs recording data include, but are not limited to: height, weight, etc. The medical record data and the physical sign record data are collected, so that the breathing cough datum data of the user to be monitored can be evaluated conveniently by using an intelligent model, the collected medical record data and the collected physical sign record data are set to be in a state to be responded, and the user waits for the subsequent quick call.
S200: inputting the medical record data and the physical sign record data into a respiratory cough characteristic evaluation model to generate a respiratory cough characteristic reference data set;
further, as shown in fig. 2, based on the inputting of the medical record data and the physical sign record data into the cough respiratory characteristics evaluation model, a cough respiratory characteristics reference data set is generated, and step S200 includes the steps of:
s210: acquiring a respiratory cough waveform threshold value evaluation module, a respiratory cough single-time duration threshold value evaluation module and a respiratory gas time ratio threshold value evaluation module according to the respiratory cough characteristic evaluation model;
s220: inputting the medical record data and the physical sign record data into the respiratory cough waveform threshold evaluation module to generate a respiratory cough waveform characteristic threshold;
s230: inputting the medical record data and the physical sign record data into the respiratory cough single-time duration threshold evaluation module to generate a respiratory cough single-time duration threshold;
s240: inputting the medical record data and the physical sign record data into a time ratio threshold evaluation module of the respiratory gas to generate a time ratio threshold of the respiratory gas;
s250: adding the breath cough waveform characteristic threshold, the breath cough single-time duration threshold and the time ratio of breath gas threshold into the breath cough characteristic baseline dataset.
Specifically, the respiratory cough feature evaluation model refers to an intelligent model trained based on a deep artificial neural network, since the respiratory cough feature can be characterized by using data quantified by a respiratory cough waveform, a respiratory air time ratio, a respiratory cough intensity, a respiratory cough frequency and the like, and different types of users to be monitored, such as: the breathing cough characteristics under the normal state corresponding to users with different medical record data and physical sign record data are different. The user states are quantized by using the medical record data and the physical sign record data, the cough states are quantized by using the breathing cough characteristics, multiple groups of quantized data can be collected, and a breathing cough characteristic evaluation model is constructed on the basis of the deep artificial neural network, so that the breathing cough characteristics in normal states of different user states can be determined, and the breathing cough characteristic evaluation model can be recorded as a breathing cough characteristic reference data set.
The preferred use of the embodiments of the present application: the respiratory cough waveform threshold evaluation module, the respiratory cough single-time duration threshold evaluation module and the respiratory gas time ratio threshold evaluation module are respectively used for extracting respiratory cough waveform characteristics in a normal state, recording the respiratory cough waveform characteristics as a respiratory cough waveform characteristic threshold, the respiratory cough single-time duration characteristics in the normal state, recording the respiratory cough single-time duration threshold and the respiratory gas time ratio characteristics in the normal state as a respiratory gas time ratio threshold. The cough state is represented by using the waveform characteristics of the breathing cough, the single time length of the breathing cough and the time bit characteristics of the breathing gas, and the patient state is represented by using the medical record data and the physical record data.
The respiratory cough waveform threshold evaluation module, the respiratory cough single-time duration threshold evaluation module and the respiratory gas time ratio threshold evaluation module belong to submodels of a respiratory cough characteristic evaluation model, are respectively constructed on the basis of a deep artificial neural network, and are further combined as parallel node submodels of the internal structure of the respiratory cough characteristic evaluation model to construct the respiratory cough characteristic evaluation model.
The method comprises the steps of firstly, carrying out quantitative characterization on data selected for a user state and a cough state, then, acquiring a training data set based on a quantitative characterization result, and constructing a respiratory cough characteristic evaluation model based on an artificial neural network, so that each cough characteristic threshold value in a normal state with higher fitness with a user to be monitored is determined according to the user state, and reliable reference data are provided for identification of monitoring data in the next step.
Further, based on the respiratory cough characteristic evaluation model, a respiratory cough waveform threshold evaluation module, a respiratory cough single-time duration threshold evaluation module, and a respiratory gas time ratio threshold evaluation module are obtained, and step S210 includes the steps of:
s211: extracting a data set constructed by a respiratory cough waveform threshold value evaluation module, a data set constructed by a respiratory cough single-time duration threshold value evaluation module and a data set constructed by a respiratory gas time ratio threshold value evaluation module based on big data;
s212: constructing a data set through the respiratory cough waveform threshold evaluation module, and training the respiratory cough waveform threshold evaluation module;
s213: constructing a data set through the breath cough single time length threshold value evaluation module, and training the breath cough single time length threshold value evaluation module;
s214: and constructing a data set through the time ratio threshold evaluation module of the respiratory gas, and training the time ratio threshold evaluation module of the respiratory gas.
Specifically, the construction data set of the respiratory cough waveform threshold evaluation module refers to a data set for constructing the respiratory cough waveform threshold evaluation module based on big data acquisition, and comprises a plurality of groups: the medical record data, the physical sign record data and the breathing cough waveform characteristic information in the normal state are recorded in the medical record data, the physical sign record data and the breathing cough waveform characteristic information in the normal state in the embodiment of the application, the breathing cough waveform characteristic information in the normal state refers to the breathing cough waveform characteristic information corresponding to the user state which does not threaten the health of the user, the medical record data and the physical sign record data are used as input training data in the later step, the breathing cough waveform characteristic information in the normal state is used as output identification information, supervised learning is carried out based on an artificial neural network, and a breathing cough waveform threshold value evaluation module is trained, wherein the breathing cough waveform characteristic refers to a digital signal converted from monitoring breathing cough time sequence data through a pressure sensor.
The construction data set of the breath cough single time length threshold value evaluation module refers to a data set used for constructing the breath cough single time length threshold value evaluation module based on big data acquisition, and comprises a plurality of groups: medical record data, sign record data and the breathing cough single time duration characteristic information of normal state, refer to the breathing cough single time duration characteristic information that the user state that can not cause threat to user health corresponds, the step is with medical record data, sign record data as input training data, the breathing cough single time duration characteristic information of normal state is as output identification information, based on artificial neural network, supervised learning is carried out, the breathing cough single time duration threshold value evaluation module is trained, wherein, the breathing cough single time duration characteristic refers to the breathing cough duration information of each time monitored through pressure sensor.
Time ratio threshold evaluation module of respiratory gas construction data set refers to a data set for constructing a time ratio threshold evaluation module of respiratory gas based on big data acquisition, comprising a plurality of sets: medical record data, sign record data and the time ratio characteristic information of the respiratory gas of normal condition, this application embodiment, refer to the time ratio characteristic information of the respiratory gas that the user state that can not cause the threat to user health corresponds, the step will be medical record data, sign record data as input training data, the time bit characteristic information of the respiratory gas of normal condition is as output identification information, based on artificial neural network, there is supervised learning, train the time ratio threshold value evaluation module of respiratory gas, wherein, the time ratio characteristic of respiratory gas refers to the time ratio of respiratory gas each time when coughing each time through pressure sensor monitoring, it is preferred: expiration/inspiration duration characterization.
Further, the method is applied to a wearable device-based cough detection system, the system belongs to a data provider, the method step S200 further includes step S260, and step S260 further includes the steps of:
s261: acquiring a first data provider, wherein the second data provider reaches an Nth data provider;
s262: the activation model building cloud cooperation system receives the first data provider, the second data provider reaches the Nth data provider, and uploads first encrypted data and second encrypted data reaches the Nth encrypted data;
s263: acquiring a respiratory cough waveform threshold value evaluation module construction data set, a respiratory cough single-time duration threshold value evaluation module construction data set and a respiratory gas time ratio threshold value evaluation module construction data set by decrypting the first encrypted data, the second encrypted data and the Nth encrypted data;
s264: constructing a respiratory cough characteristic evaluation model according to the respiratory cough waveform threshold value evaluation module construction data set, the respiratory cough single-time duration threshold value evaluation module construction data set and the respiratory gas time ratio threshold value evaluation module construction data set;
s265: distributing the respiratory cough signature assessment model to the first data provider, the second data provider up to the Nth data provider.
Specifically, since the user status information of the user to be monitored: medical record data, physical sign record data and the like are recorded in medical detection institutions or various hospitals, and in order to protect privacy information of users, the information is usually stored in an encrypted manner, so that the information is difficult to directly call, and then a model is constructed by adopting the idea of federal learning, wherein the process is as follows:
the first data provider, the second data provider and the Nth data provider refer to different medical detection institutions or hospitals, N represents the number of the data providers and is defined and set by staff, and various types of information such as medical record data, physical sign record data and cough characteristic record data of a user to be monitored are stored in the data providers. In order to break a data island, activating a model to construct a cloud-side collaboration system, wherein the cloud-side collaboration system is a trusted collaboration system of a third party for model construction; through a first data provider, a second data provider and an Nth data provider, a data set is constructed by a breathing cough waveform threshold value evaluation module, a data set is constructed by a breathing cough single-time duration threshold value evaluation module and a data set is constructed by a breathing gas time ratio threshold value evaluation module, which are locally stored and used for constructing a breathing cough characteristic evaluation model, and the data set is encrypted in an encryption mode issued by a model construction cloud cooperation system and then uploaded to the model construction cloud cooperation system; the model building cloud cooperation system integrates a plurality of participants to provide encrypted data, and due to the fact that the encryption mode is issued by the cooperation system, the model building cloud cooperation system can decrypt a data set, model building training is further conducted at the cloud, and after the model training is completed, the trained model can be distributed to the model building cloud cooperation system for use.
Through the idea based on federal learning, the respiratory cough characteristic evaluation model is constructed, a data isolated island is broken under the premise of guaranteeing the privacy data of a user, and stable training of the respiratory cough characteristic evaluation model is achieved.
S300: uploading a breath cough monitoring data sequence through a pressure sensor deployed at a preset position on the body of the user to be monitored;
further, as shown in fig. 3, based on the uploading of the breath cough monitoring data sequence by the pressure sensor disposed at the preset position on the user to be monitored, step S300 includes the steps of:
s310: acquiring a voltage signal sequence of respiratory cough monitoring data according to the pressure sensor;
s320: inputting the voltage signal sequence of the respiratory cough monitoring data into a signal amplifying circuit, and outputting the amplified voltage signal sequence of the respiratory cough monitoring data;
s330: carrying out digital signal conversion on the amplified voltage signal sequence of the respiratory cough monitoring data to generate a digital signal sequence of the respiratory cough monitoring data;
s340: and setting the digital signal sequence of the respiratory cough monitoring data as the respiratory cough monitoring data sequence.
Specifically, the preset position refers to an area where a pressure sensor is deployed on the body of a user to be monitored, can be set by a worker in a self-defined mode, and is set as the abdomen or the chest by default; the pressure sensor is a sensor for monitoring the real-time breathing cough state of the user to be monitored, preferably a wearable sensor, and is stably deployed at a preset position in a wearing manner, the form is preferably as shown in fig. 4, but no limitation is imposed on the specific wearing form; the breathing cough monitoring data sequence refers to a result obtained by monitoring the breathing cough data of a user to be monitored in real time through a pressure sensor, and the acquisition process is preferably as follows:
the voltage signal sequence of the respiratory cough monitoring data refers to a real-time voltage signal monitored by a pressure sensor, the pressure sensor can convert the pressure signal into a voltage signal, the pressure signal is formed by different respiratory cough states of a user to be monitored, and the respiratory cough state of the user to be monitored can be represented by the voltage signal sequence of the respiratory cough monitoring data; the signal amplifying circuit is a circuit for amplifying a voltage signal, and is also commonly used at present, and is not described herein; the amplified voltage signal sequence of the respiratory cough monitoring data refers to a signal set obtained by amplifying the voltage signal sequence of the respiratory cough monitoring data; the digital signal sequence of the respiratory cough monitoring data refers to a signal set obtained by converting an amplified voltage signal sequence of the respiratory cough monitoring data into a digital signal, and the conversion of the voltage signal into the digital signal is relatively wide in application and is not described in detail herein. And setting the digital signal sequence of the respiratory cough monitoring data as a respiratory cough monitoring data sequence, setting the digital signal sequence as a state to be responded, and waiting for calling in the next step.
The voltage signal is amplified through the signal amplification circuit, the amplification factor is defined by a worker, information loss can be avoided when the digital signal is converted in the next step, and the accuracy of the respiratory cough monitoring data sequence is improved.
S400: extracting a breathing cough characteristic comparison data set according to the breathing cough monitoring data sequence;
further, based on the extracting of the breath cough feature contrast data set according to the breath cough monitoring data sequence, step S400 includes the steps of:
s410: acquiring a respiratory cough feature extraction dimension according to the respiratory cough feature reference data set;
s420: and traversing the breathing cough characteristic extraction dimension to perform characteristic extraction on the breathing cough monitoring data sequence to generate the breathing cough characteristic comparison data set.
Specifically, the respiratory cough feature contrast data set refers to a monitoring feature set extracted from the respiratory cough monitoring data sequence and corresponding to features in the respiratory cough feature reference data set, and the extraction process is preferably as follows: according to the digital signal of the breathing cough monitoring data sequence, a first coordinate axis is constructed by taking time as a reference, a second coordinate axis is constructed by taking the monitored digital specific value as a reference, and then the breathing cough monitoring data sequence is input, so that a breathing cough oscillogram can be obtained, the breathing cough intensity can be determined according to the fluctuation range, the breathing cough frequency can be determined according to the cycle number, the expiration time and the inspiration time of the breathing cough can be determined according to the cycle length, and further the cough duration and the breathing time ratio of each time can be determined. And setting the extracted respiratory cough characteristic contrast data set as a state to be responded, and waiting for quick calling in the next step.
S500: determining whether the respiratory cough feature contrast data set satisfies the respiratory cough feature baseline data set;
s600: if the breathing cough characteristic contrast data set does not meet the breathing cough characteristic reference data set, generating a breathing cough abnormal signal;
further, based on the if the respiratory cough feature contrast data set does not satisfy the respiratory cough feature reference data set, the step S600 generates a respiratory cough abnormal signal, and includes the steps of:
s610: when the respiratory cough feature comparison data set does not meet the respiratory cough feature reference data set, acquiring a respiratory cough deviation feature type and a respiratory cough feature comparison deviation value, wherein the respiratory cough deviation feature type corresponds to the respiratory cough feature comparison deviation value in a one-to-one correspondence manner;
s620: and adding the respiratory cough deviation characteristic type and the respiratory cough characteristic ratio into the respiratory cough abnormal signal.
Specifically, the breathing cough abnormal signal refers to comparing feature values corresponding to the breathing cough feature comparison data set and the breathing cough feature reference data set one by one, and if the feature values of the breathing cough feature comparison data set do not belong to the feature values recorded in the breathing cough feature reference data set, it indicates that the probability that the breathing cough state of the user to be monitored is abnormal is high, and then the breathing cough abnormal signal is generated.
The abnormal breathing cough signal is determined as follows: the respiratory cough deviation characteristic type refers to a characteristic type that after comparing the respiratory cough characteristic comparison data set with characteristic values corresponding to the respiratory cough characteristic reference data set one by one, the obtained respiratory cough characteristic comparison data set does not belong to the characteristic values recorded in the respiratory cough characteristic reference data set, and includes but is not limited to: waveform, period, amplitude, number of periods, etc.; the cough from breath characteristic ratio deviation value refers to the deviation between an abnormal characteristic value and a normal characteristic value, including but not limited to: the deviation of the waveform, the deviation of the period, the deviation of the amplitude, the deviation of the number of periods and the like are stored in a manner of correlating the respiratory cough deviation characteristic types and the respiratory cough characteristic comparison deviation values which correspond one to one, and the respiratory cough abnormal signals are added to facilitate the follow-up checking of medical staff.
S700: and identifying the respiratory cough monitoring data sequence according to the respiratory cough abnormal signal, and then sending the respiratory cough monitoring data sequence to a display terminal for displaying.
Specifically, the anomaly monitoring information in the respiratory cough monitoring data sequence is mainly identified according to respiratory cough deviation characteristic types and respiratory cough characteristic comparison deviation values which correspond to one another in the respiratory cough anomaly signals one by one, the display terminal refers to a cough characteristic display terminal for monitoring the respiratory cough state of a user to be monitored by medical staff, the characteristics with high anomaly probability are identified through the respiratory cough monitoring data sequence, differential display is preferably performed on the display terminal, for example, special color area identification is used, and the technical effect of improving the referential performance of the respiratory cough monitoring data sequence is achieved.
To sum up, the cough detection method and system based on the wearable device provided by the embodiment of the application have the following technical effects:
1. because the basic information uploaded to the user to be monitored, namely the patient, is adopted: medical record and physical record; using an intelligent model: the respiratory cough characteristic evaluation model determines a respiratory cough characteristic reference data set in a normal state, which has higher fitness with a user to be monitored, according to the medical record and the physical record; monitoring a breath cough characteristic contrast data set through a pressure sensor; the reference data set and the comparison data set are compared, after abnormal identification is carried out on a breathing cough monitoring data sequence of the breathing cough characteristic comparison data set which does not meet the breathing cough characteristic reference data set, the whole breathing cough monitoring data sequence is sent to a display terminal for a professional to check, the breathing cough characteristic reference data set which is in accordance with a user to be monitored is evaluated through an intelligent model and is compared with monitored real-time data, the abnormal data are identified, the later display terminal is convenient to highlight, the professional is convenient to check in a key mode, and the technical effect of improving the reference of the breathing cough monitoring data is achieved.
Example two
Based on the same inventive concept as the cough detection method based on the wearable device in the foregoing embodiment, as shown in fig. 3, an embodiment of the present application provides a cough detection system based on the wearable device, where the system includes a display terminal 001, the system is in communication connection with a pressure sensor 002, and the system includes:
the system comprises a user information loading module 11, a monitoring module and a monitoring module, wherein the user information loading module is used for acquiring basic information of a user to be monitored, and the basic information of the user to be monitored comprises medical record data and physical sign record data;
the characteristic reference data generation module 12 is configured to input the medical record data and the physical sign record data into a respiratory cough characteristic evaluation model, and generate a respiratory cough characteristic reference data set;
the monitoring data uploading module 13 is used for uploading a breathing cough monitoring data sequence through a pressure sensor 002 which is deployed at a preset position on the body of the user to be monitored;
a feature comparison data extraction module 14, configured to extract a breathing cough feature comparison data set according to the breathing cough monitoring data sequence;
a feature data comparison module 15, configured to determine whether the comparison data set of respiratory cough features satisfies the reference data set of respiratory cough features;
an abnormal signal generating module 16, configured to generate a respiratory cough abnormal signal if the respiratory cough feature comparison data set does not satisfy the respiratory cough feature reference data set;
and the abnormity identification module 17 is used for identifying the respiratory cough monitoring data sequence according to the respiratory cough abnormity signal and then sending the respiratory cough monitoring data sequence to the display terminal 001 for displaying.
Further, the feature reference data generating module 12 performs steps including:
according to the respiratory cough characteristic evaluation model, a respiratory cough waveform threshold evaluation module, a respiratory cough single-time duration threshold evaluation module and a respiratory gas time ratio threshold evaluation module are obtained;
inputting the medical record data and the physical sign record data into the respiratory cough waveform threshold evaluation module to generate a respiratory cough waveform characteristic threshold;
inputting the medical record data and the physical sign record data into the respiratory cough single-time duration threshold evaluation module to generate a respiratory cough single-time duration threshold;
inputting the medical record data and the physical sign record data into a time ratio threshold evaluation module of the respiratory gas to generate a time ratio threshold of the respiratory gas;
adding the breath cough waveform characteristic threshold, the breath cough single-time duration threshold and the time ratio of breath gas threshold into the breath cough characteristic baseline dataset.
Further, the feature reference data generating module 12 executes the steps further including:
extracting a data set constructed by a respiratory cough waveform threshold value evaluation module, a data set constructed by a respiratory cough single-time duration threshold value evaluation module and a data set constructed by a respiratory gas time ratio threshold value evaluation module based on big data;
constructing a data set through the respiratory cough waveform threshold value evaluation module, and training the respiratory cough waveform threshold value evaluation module;
constructing a data set through the respiratory cough single-time duration threshold evaluation module, and training the respiratory cough single-time duration threshold evaluation module;
and constructing a data set through the time ratio threshold evaluation module of the respiratory gas, and training the time ratio threshold evaluation module of the respiratory gas.
Further, the characteristic reference data generating module 12 performs the steps further including:
acquiring a first data provider, and acquiring a second data provider till an Nth data provider;
activating a model building cloud cooperation system, receiving the first data provider, uploading first encrypted data by the second data provider till the Nth data provider, and uploading second encrypted data till the Nth encrypted data;
acquiring a respiratory cough waveform threshold value evaluation module construction data set, a respiratory cough single-time duration threshold value evaluation module construction data set and a respiratory gas time ratio threshold value evaluation module construction data set by decrypting the first encrypted data, the second encrypted data and the Nth encrypted data;
constructing a data set according to the breathing cough waveform threshold value evaluation module construction data set, the breathing cough single-time duration threshold value evaluation module construction data set and the breathing gas time ratio threshold value evaluation module construction data set, and constructing the breathing cough characteristic evaluation model;
distributing the breath cough signature evaluation model to the first data provider, the second data provider up to the Nth data provider.
Further, the monitoring data uploading module 13 executes steps including:
acquiring a voltage signal sequence of respiratory cough monitoring data according to the pressure sensor;
inputting the voltage signal sequence of the respiratory cough monitoring data into a signal amplifying circuit, and outputting the amplified voltage signal sequence of the respiratory cough monitoring data;
carrying out digital signal conversion on the amplified voltage signal sequence of the respiratory cough monitoring data to generate a digital signal sequence of the respiratory cough monitoring data;
setting the digital signal sequence of the respiratory cough monitoring data as the respiratory cough monitoring data sequence.
Further, the feature comparison data extraction module 14 performs steps including:
acquiring a respiratory cough feature extraction dimension according to the respiratory cough feature reference data set;
and traversing the breathing cough feature extraction dimensionality to perform feature extraction on the breathing cough monitoring data sequence to generate the breathing cough feature comparison data set.
Further, the abnormal signal generating module 16 performs steps including:
when the respiratory cough feature comparison data set does not meet the respiratory cough feature reference data set, acquiring a respiratory cough deviation feature type and a respiratory cough feature comparison deviation value, wherein the respiratory cough deviation feature type corresponds to the respiratory cough feature comparison deviation value in a one-to-one correspondence manner;
and adding the respiratory cough deviation characteristic type and the respiratory cough characteristic ratio into the respiratory cough abnormal signal.
Any steps of the methods described above may be stored as computer instructions or programs in a non-limiting computer memory and may be identified by a non-limiting computer processor call to implement any of the methods in the embodiments of the present application without unnecessary limitation.
Furthermore, the first and second elements may represent more than an order, may represent a specific concept, and/or may be selected individually or collectively from a plurality of elements. It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the present application and its equivalent technology, it is intended that the present application include such modifications and variations.
Claims (8)
1. A wearable device based cough detection method, wherein the method employs a wearable device based cough detection system, the system includes a display terminal, the system is in communication connection with a pressure sensor, and the method includes:
acquiring basic information of a user to be monitored, wherein the basic information of the user to be monitored comprises medical record data and physical sign record data;
inputting the medical record data and the physical sign record data into a respiratory cough characteristic evaluation model to generate a respiratory cough characteristic reference data set;
uploading a breath cough monitoring data sequence through a pressure sensor deployed at a preset position on the body of the user to be monitored;
extracting a respiratory cough characteristic comparison data set according to the respiratory cough monitoring data sequence;
determining whether the respiratory cough feature contrast data set satisfies the respiratory cough feature baseline data set;
if the breathing cough characteristic contrast data set does not meet the breathing cough characteristic reference data set, generating a breathing cough abnormal signal;
and after the respiratory cough monitoring data sequence is identified according to the respiratory cough abnormal signal, the respiratory cough monitoring data sequence is sent to a display terminal for displaying.
2. The method of claim 1, wherein the inputting the medical record data and the vital sign record data into a cough breathing characteristic assessment model to generate a cough breathing characteristic baseline data set comprises:
acquiring a respiratory cough waveform threshold value evaluation module, a respiratory cough single-time duration threshold value evaluation module and a respiratory gas time ratio threshold value evaluation module according to the respiratory cough characteristic evaluation model;
inputting the medical record data and the physical sign record data into the respiratory cough waveform threshold evaluation module to generate a respiratory cough waveform characteristic threshold;
inputting the medical record data and the physical sign record data into the respiratory cough single-time duration threshold evaluation module to generate a respiratory cough single-time duration threshold;
inputting the medical record data and the physical sign record data into a time ratio threshold evaluation module of the respiratory gas to generate a time ratio threshold of the respiratory gas;
adding the breath cough waveform characteristic threshold, the breath cough single-time duration threshold and the time ratio of breath gas threshold into the breath cough characteristic baseline dataset.
3. The method of claim 2, wherein said obtaining a breath cough waveform threshold evaluation module, a breath cough single time duration threshold evaluation module, and a breath gas time ratio threshold evaluation module according to the breath cough characteristic evaluation model comprises:
extracting a data set constructed by a respiratory cough waveform threshold value evaluation module, a data set constructed by a respiratory cough single-time duration threshold value evaluation module and a data set constructed by a respiratory gas time ratio threshold value evaluation module based on big data;
constructing a data set through the respiratory cough waveform threshold value evaluation module, and training the respiratory cough waveform threshold value evaluation module;
constructing a data set through the respiratory cough single-time duration threshold evaluation module, and training the respiratory cough single-time duration threshold evaluation module;
and constructing a data set through the time ratio threshold evaluation module of the respiratory gas, and training the time ratio threshold evaluation module of the respiratory gas.
4. The method of claim 3, wherein the method is applied to a wearable device-based cough detection system, the system belonging to a data provider, the method further comprising:
acquiring a first data provider, and acquiring a second data provider till an Nth data provider;
activating a model building cloud cooperation system, receiving the first data provider, uploading first encrypted data by the second data provider till the Nth data provider, and uploading second encrypted data till the Nth encrypted data;
acquiring a respiratory cough waveform threshold value evaluation module construction data set, a respiratory cough single-time duration threshold value evaluation module construction data set and a respiratory gas time ratio threshold value evaluation module construction data set by decrypting the first encrypted data, the second encrypted data and the Nth encrypted data;
constructing a respiratory cough characteristic evaluation model according to the respiratory cough waveform threshold value evaluation module construction data set, the respiratory cough single-time duration threshold value evaluation module construction data set and the respiratory gas time ratio threshold value evaluation module construction data set;
distributing the respiratory cough signature assessment model to the first data provider, the second data provider up to the Nth data provider.
5. The method of claim 1, wherein uploading a sequence of breath cough monitoring data via a pressure sensor deployed at a predetermined location on the user to be monitored comprises:
acquiring a voltage signal sequence of respiratory cough monitoring data according to the pressure sensor;
inputting the voltage signal sequence of the respiratory cough monitoring data into a signal amplifying circuit, and outputting the amplified voltage signal sequence of the respiratory cough monitoring data;
carrying out digital signal conversion on the amplified voltage signal sequence of the respiratory cough monitoring data to generate a digital signal sequence of the respiratory cough monitoring data;
setting the digital signal sequence of the respiratory cough monitoring data as the respiratory cough monitoring data sequence.
6. The method of claim 1, wherein said extracting a respiratory cough signature contrast data set from said sequence of respiratory cough monitoring data comprises:
acquiring a respiratory cough feature extraction dimension according to the respiratory cough feature reference data set;
and traversing the breathing cough characteristic extraction dimension to perform characteristic extraction on the breathing cough monitoring data sequence to generate the breathing cough characteristic comparison data set.
7. The method of claim 1, wherein generating a respiratory cough anomaly signal if the respiratory cough signature comparison dataset does not satisfy the respiratory cough signature reference dataset comprises:
when the respiratory cough feature comparison data set does not meet the respiratory cough feature reference data set, acquiring a respiratory cough deviation feature type and a respiratory cough feature comparison deviation value, wherein the respiratory cough deviation feature type corresponds to the respiratory cough feature comparison deviation value in a one-to-one correspondence manner;
and adding the respiratory cough deviation characteristic type and the respiratory cough characteristic ratio into the respiratory cough abnormal signal.
8. A cough detection system based on wearable equipment, characterized in that, the system includes display terminal, system and pressure sensor communication connection, the system includes:
the system comprises a user information loading module, a monitoring module and a monitoring module, wherein the user information loading module is used for acquiring basic information of a user to be monitored, and the basic information of the user to be monitored comprises medical record data and sign record data;
the characteristic reference data generation module is used for inputting the medical record data and the physical sign record data into a respiratory cough characteristic evaluation model to generate a respiratory cough characteristic reference data set;
the monitoring data uploading module is used for uploading a breathing cough monitoring data sequence through a pressure sensor arranged at a preset position on the body of the user to be monitored;
the characteristic comparison data extraction module is used for extracting a respiratory cough characteristic comparison data set according to the respiratory cough monitoring data sequence;
the characteristic data comparison module is used for judging whether the breath cough characteristic comparison data set meets the breath cough characteristic reference data set or not;
the abnormal signal generating module is used for generating a breathing cough abnormal signal if the breathing cough characteristic contrast data set does not meet the breathing cough characteristic reference data set;
and the abnormity identification module is used for identifying the respiratory cough monitoring data sequence according to the respiratory cough abnormity signal and then sending the respiratory cough monitoring data sequence to a display terminal for displaying.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210861062.6A CN115191990A (en) | 2022-07-22 | 2022-07-22 | Cough detection method and system based on wearable device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210861062.6A CN115191990A (en) | 2022-07-22 | 2022-07-22 | Cough detection method and system based on wearable device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115191990A true CN115191990A (en) | 2022-10-18 |
Family
ID=83584883
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210861062.6A Pending CN115191990A (en) | 2022-07-22 | 2022-07-22 | Cough detection method and system based on wearable device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115191990A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117145752A (en) * | 2023-10-26 | 2023-12-01 | 意朗智能科技(南通)有限公司 | Filtering fault identification method and system for air compressor |
-
2022
- 2022-07-22 CN CN202210861062.6A patent/CN115191990A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117145752A (en) * | 2023-10-26 | 2023-12-01 | 意朗智能科技(南通)有限公司 | Filtering fault identification method and system for air compressor |
| CN117145752B (en) * | 2023-10-26 | 2024-01-30 | 意朗智能科技(南通)有限公司 | Filtering fault identification method and system for air compressor |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107798350B (en) | Cardiopulmonary sound signal identification method and system | |
| US20030177038A1 (en) | Early detection of disease outbreak using electronic patient data to reduce public health threat from bio-terrorism | |
| WO2005092176A1 (en) | Real-time remote diagnosis of in vivo images | |
| CN104523252B (en) | A kind of mobile health service method and system, mobile terminal | |
| AbdElnapi et al. | A survey of internet of things technologies and projects for healthcare services | |
| US9633168B2 (en) | Biometric identity validation for use with unattended tests for medical conditions | |
| KR20140063100A (en) | Apparatus and methods for remote cardiac disease management | |
| CN108492890B (en) | Human health state monitoring system and method | |
| CN106777929A (en) | Eeg monitoring system based on smart cloud | |
| CN115769302A (en) | Epidemic disease monitoring system | |
| CN115191990A (en) | Cough detection method and system based on wearable device | |
| KR20130118512A (en) | System and method for monitoring the health of a patient using face recognition technology | |
| CN209232420U (en) | A kind of intelligence delirium assessment device | |
| JP5544365B2 (en) | Improvements in multi-parameter monitoring or improvements related to multi-parameter monitoring | |
| CN115862868A (en) | Psychological assessment system, psychological assessment platform, electronic device and storage medium | |
| CN116098595B (en) | System and method for monitoring and preventing sudden cardiac death and sudden cerebral death | |
| Kanagasabai et al. | Brain-computer interface learning system for Quadriplegics | |
| CN111863277A (en) | Crowd body temperature monitoring system based on big data | |
| KR101788034B1 (en) | System and method for patient monitoring | |
| US20210186344A1 (en) | Clinical artificial intelligence (ai) software and terminal gateway hardware method for monitoring a subject to detect a possible respiratory disease | |
| Almazroa et al. | An Internet of Things (IoT) management system for improving homecare-A case study | |
| CN110428894B (en) | Distributed monitoring system and method | |
| CN112037916A (en) | Shared multifunctional sudden death prevention physiological information detection system and method thereof | |
| Eraslan et al. | Novel Biometric Approach Based on Diaphragmatic Respiratory Movements Using Single-Lead EMG Signals | |
| JP2021133005A (en) | Auscultation system and auscultation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |