A portable IoT heart and lung sound auscultation system

Through a mobile portable IoT cardiopulmonary sound auscultation system, combined with single-part and multi-part intelligent analysis reports, and using multiple experts to remote consultation, the problem of low diagnostic accuracy in the existing technology is solved, and the diagnosis accuracy and reliability of cardiopulmonary diseases is improved. .

CN114601492BActive Publication Date: 2024-11-19CHONGQING UNIV +1
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202210230567.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-03-10
Publication Date
2024-11-19
Estimated Expiration
2042-03-10

AI Technical Summary

Technical Problem

The existing IoT cardiopulmonary auscultation system ignores the complexity of cardiopulmonary diseases and the heterogeneity of auscultation sounds in different parts during diagnosis, resulting in low diagnostic accuracy.

Method used

Using a mobile portable IoT cardiopulmonary sound auscultation system, auscultivation recording of the user's heart and lungs is obtained, and a single-part intelligent analysis report is combined with multi-part intelligent analysis report and algorithm model analysis report, and remote consultation of multiple experts is conducted to comprehensively consider the heart and lungs. complexity of the disease and heterogeneity of the auscultation sound.

Benefits of technology

It improves the accuracy of cardiopulmonary disease diagnosis, reduces the dependence on the diagnostic environment and expert consultants, enhances the convenience of self-screening and the reliability of diagnosis, and reduces the possibility of misdiagnosis and missed diagnosis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114601492B_ABST
    Figure CN114601492B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of electronic auscultation and telemedicine technology, and provides a mobile portable Internet of Things cardiopulmonary sound auscultation system, including: recording the user's heart and lungs for auscultation to obtain a first user auscultation report; performing intelligent analysis on the first user auscultation report to obtain each single-site intelligent analysis report; obtaining a multi-site intelligent analysis report based on each single-site intelligent analysis report; obtaining a second user auscultation report based on the first user auscultation report, each single-site intelligent analysis report, and the multi-site intelligent analysis report; analyzing the second user auscultation report using a number of algorithm models to obtain an algorithm model analysis report; conducting a multi-expert remote consultation on the algorithm model analysis report and the second user auscultation report to obtain an expert analysis report. Since the multi-intelligent analysis report comprehensively analyzes the pathological auscultation sound features of multiple sites, the second user auscultation report is more accurate, and can further assist in diagnosis, thereby improving the accuracy of diagnosis of cardiopulmonary diseases.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of electronic auscultation and telemedicine, and in particular to a mobile portable Internet of Things cardiopulmonary sound auscultation system. Background Art

[0002] Auscultation is an important medical diagnostic method, mainly targeting heart and lung sounds. Medical staff can make a preliminary judgment on the possible chest and lung diseases of patients by identifying abnormalities in heart and lung sounds, whether there are additional sounds, and the differences in the location, amplitude, frequency, and timing of occurrence. With the continuous development of information technology and integrated circuit technology, the society's demand for more objective, more transparent, more controllable, and more portable telemedicine is increasing day by day, and the medical Internet of Things model based on electronic auscultation technology has also become a popular development direction in the current medical industry. In the existing technology, the Internet of Things heart and lung auscultation system first obtains the user's heart or lung audio conditions, and then obtains a single intelligent analysis report of the user's heart or lung part, and then directly makes a preliminary diagnosis of the disease based on the single intelligent analysis report of any part of the heart or lung. This diagnostic method ignores the complexity of cardiopulmonary diseases and the heterogeneity of auscultation sounds in different parts, resulting in low diagnostic accuracy. Summary of the invention

[0003] In view of the shortcomings of the prior art, the present invention proposes a mobile portable Internet of Things heart and lung sound auscultation system to improve the accuracy of diagnosis.

[0004] The technical solution adopted by the present invention is a mobile and portable Internet of Things heart and lung sound auscultation system.

[0005] In the first feasible method, the auscultation recording of the user's heart and lungs is performed to obtain the first user's auscultation report; the first user's auscultation report is intelligently analyzed to obtain each single-part intelligent analysis report; a multi-part intelligent analysis report is obtained based on each single-part intelligent analysis report; a second user's auscultation report is obtained based on the first user's auscultation report, each single-part intelligent analysis report and the multi-part intelligent analysis report; the second user's auscultation report is analyzed using several algorithm models to obtain an algorithm model analysis report; a multi-expert remote consultation is performed on the algorithm model analysis report and the second user's auscultation report to obtain an expert analysis report.

[0006] In combination with the first feasible method, in a second feasible method, auscultation recording is performed on the user's heart and lungs to obtain a first user auscultation report: the vibration mechanical energy signals of the user's heart and lungs are converted into corresponding current signals; each current signal is pre-processed to denoise and obtain each analog electrical signal; each analog electrical signal is converted into a digital-to-analog format to generate a user heart audio signal and a user lung audio signal; the user heart audio signal and the user lung audio signal are determined as the first user auscultation report.

[0007] In combination with the second feasible method, in the third feasible method, the auscultation of the user's heart and lungs is recorded to obtain the first user's auscultation report, which also includes: when the auscultation of the user's heart and lungs is recorded, the auscultation response and condition description of any part of the user's body are recorded to obtain the user's real-time evaluation.

[0008] In combination with the second feasible method, in a fourth feasible method, the user's heart audio signal and the user's lung audio signal are intelligently analyzed to obtain intelligent analysis reports for each single part, including: bandpass filtering each audio signal and performing noise reduction processing on each filtered audio signal; extracting features from each noise-reduced audio signal to obtain features of each part; classifying the features of each part using a preset audio signal processing model to obtain abnormal situation identification results of each audio signal; and obtaining intelligent analysis reports for each single part based on the abnormal situation identification results of each audio signal.

[0009] In combination with the first feasible method, in the fifth feasible method, a multi-part intelligent analysis report is obtained based on each single-part intelligent analysis report, including: processing each single-part intelligent analysis report according to a preset auscultation order and a preset vector matrix form to obtain a vector matrix; inputting the vector matrix into a preset intelligent analysis report processing model to obtain a multi-part intelligent analysis report.

[0010] In combination with the first feasible method, in the sixth feasible method, several algorithm models are used to analyze the auscultation report of the second user, and before obtaining the algorithm model analysis report, it also includes: screening out several candidate algorithm models according to the auscultation report of the second user; each candidate algorithm model includes several candidate algorithm models of the same type of cardiopulmonary disease and candidate algorithm models of different types of cardiopulmonary disease; obtaining the correlation between each candidate algorithm model and each type of cardiopulmonary disease, as well as the accuracy of each candidate algorithm model; and selecting several algorithm models according to each correlation and each accuracy.

[0011] In combination with the sixth possible implementation method, in the seventh possible implementation method, several algorithm models are selected according to each relevance and each accuracy, including: determining the candidate algorithm model corresponding to the relevance greater than the first preset threshold as the selected algorithm model; and / or, determining the candidate algorithm model corresponding to the accuracy greater than the second preset threshold as the selected algorithm model.

[0012] In combination with the first feasible method, in the eighth feasible method, before conducting a multi-expert remote consultation on the algorithm model analysis report and the second user's auscultation report, it also includes: obtaining the disease field involved in the second user's auscultation report; screening out a number of candidate expert consultants in the disease field; obtaining the popularity of each candidate expert consultant; and selecting multiple expert consultants based on the popularity of each candidate expert consultant.

[0013] In combination with the first feasible method, in a ninth feasible method, a multi-expert remote consultation is conducted on the algorithm model analysis report and the second user auscultation report to obtain an expert analysis report, including: the user applies for a multi-expert remote consultation, and sends the algorithm model analysis report and the second user auscultation report to each expert consultant; after each expert consultant responds to the user's consultation application, the expert analysis report is sent to the user.

[0014] In combination with the ninth possible implementation method, in the tenth possible implementation method, after each expert consultant responds to the user's consultation request, the expert analysis report is sent to the user, including: each expert consultant responds to the user's consultation request, receives the algorithm model analysis report and the second user's auscultation report; analyzes the algorithm model analysis report and the second user's auscultation report, gives expert diagnosis opinions, and forms an expert analysis report; and sends the expert analysis report to the user.

[0015] It can be seen from the above technical solution that the beneficial technical effects of the present invention are as follows:

[0016] 1. Due to the heterogeneity of auscultation sounds in different parts, the user's heart and lung audio signals are intelligently analyzed separately to obtain intelligent analysis reports for each single part. Since the complexity of cardiopulmonary diseases involves each single part, multi-part intelligent analysis reports are obtained based on the intelligent analysis reports of each single part. In this way, both the heterogeneity of auscultation sounds in different parts and the complexity of cardiopulmonary diseases are taken into account, so that the multi-intelligent analysis report integrates the pathological auscultation sound characteristics of multiple parts, which is more accurate, and can assist in diagnosis, thereby improving the accuracy of cardiopulmonary disease diagnosis.

[0017] 2. Since intelligent analysis is first used to obtain a single intelligent analysis report, and then the algorithm model is used to analyze the second user's stethoscope report to obtain the algorithm model analysis report, the user's dependence on the diagnostic environment and expert consultants is reduced, making it easier for users to conduct self-screening anytime and anywhere, discover possible latent conditions, and not delay the best time to seek medical treatment; the algorithm model analysis report can also be used as an auxiliary diagnostic opinion to further improve the accuracy of cardiopulmonary disease diagnosis.

[0018] 3. Use a multi-algorithm model to analyze the second user's auscultation report to obtain an algorithm model analysis report, and then conduct a multi-expert remote consultation on the algorithm model analysis report and the second user's auscultation report to obtain an expert analysis report. In this way, the use of a many-to-one diagnosis mode reduces the possibility of misdiagnosis and missed diagnosis, and further improves the accuracy of cardiopulmonary disease diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the prior art description. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0020] Figure 1 A schematic diagram of a mobile portable IoT heart and lung sound auscultation system provided by the present invention;

[0021] Figure 2 A wireless connection flow chart of a common user terminal and a signal acquisition terminal of a mobile portable Internet of Things cardiopulmonary sound auscultation system provided by the present invention;

[0022] Figure 3 A schematic diagram of a common user terminal provided by the present invention;

[0023] Figure 4 A schematic diagram of the structure of a mobile portable Internet of Things heart and lung sound auscultation system provided by the present invention;

[0024] Figure 5 A payment and recharge flow chart of a mobile portable Internet of Things cardiopulmonary sound auscultation system provided by the present invention;

[0025] Figure 6 A comprehensive consultation application flow chart of a mobile portable Internet of Things cardiopulmonary sound auscultation system provided by the present invention;

[0026] Figure 7 A comprehensive consultation acceptance and processing flow chart of a mobile portable Internet of Things cardiopulmonary sound auscultation system provided by the present invention. DETAILED DESCRIPTION

[0027] The following embodiments of the technical solution of the present invention are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and are therefore only used as examples, and cannot be used to limit the protection scope of the present invention.

[0028] It should be noted that, unless otherwise specified, the technical terms or scientific terms used in this application should have the common meanings understood by those skilled in the art to which the present invention belongs.

[0029] Combination Figure 1 As shown, this embodiment provides a mobile portable IoT cardiopulmonary sound auscultation system, comprising the following steps:

[0030] Step S101, auscultating the user's heart and lungs to obtain a first user auscultation report;

[0031] Step S102, performing intelligent analysis on the first user's auscultation report to obtain intelligent analysis reports for each single part;

[0032] Step S103, obtaining a multi-part intelligent analysis report according to each single-part intelligent analysis report;

[0033] Step S104, obtaining a second user's auscultation report according to the first user's auscultation report, each single-part intelligent analysis report, and the multi-part intelligent analysis report;

[0034] Step S105: Analyze the auscultation report of the second user using a plurality of algorithm models to obtain an algorithm model analysis report;

[0035] Step S106: Conduct a multi-expert remote consultation on the algorithm model analysis report and the second user's auscultation report to obtain an expert analysis report.

[0036] Optionally, auscultation recording is performed on the user's heart and lungs to obtain a first user auscultation report: the vibration mechanical energy signals of the user's heart and lungs are converted into corresponding current signals; each current signal is pre-processed to denoise and obtain each analog electrical signal; each analog electrical signal is converted from digital to analog to generate a user heart audio signal and a user lung audio signal; the user heart audio signal and the user lung audio signal are determined as the first user auscultation report.

[0037] Optionally, the vibration mechanical energy signals of the user's heart and lungs are converted into corresponding current signals through sensors; the current signals are amplified by amplifiers, and noise outside the human cardiopulmonary audio band and industrial frequency signal interference generated by the equipment in a 220V AC power environment are removed to achieve denoising preprocessing and obtain analog electrical signals; each analog electrical signal is converted into digital-to-analog through a digital-to-analog conversion module to generate a user heart audio signal and a user lung audio signal.

[0038] Optionally, the mobile portable IoT cardiopulmonary sound auscultation system includes a signal acquisition terminal and an ordinary user terminal. The mobile portable IoT cardiopulmonary sound auscultation system obtains the user's heart audio signal and the user's lung audio signal through the signal acquisition terminal, picks up and stores the human heart and lung sound signals, and transmits audio data with the ordinary user terminal through wireless communication. In some embodiments, the signal acquisition terminal is an electronic stethoscope, an intelligent stethoscope, or a digital stethoscope device.

[0039] Optionally, the signal acquisition end includes: a sensor module, a signal preprocessing module, and a digital-to-analog conversion module. The sensor module includes a sensor and its interface matching circuit, which is used to convert the tiny vibration mechanical energy signal transmitted from the user's heart and lung sounds to the body surface into a current signal; the sensor includes but is not limited to a piezoelectric ceramic sensor, a piezoelectric film sensor, an electret microphone sensor, a capacitive sensor, a piezoresistive sensor, and a cantilever beam sensor. The signal preprocessing module is used to perform denoising preprocessing on each current signal to obtain each analog electrical signal; the signal preprocessing module includes an operational amplifier, a bandpass filter, and an industrial frequency notch filter of the signal. Among them, the operational amplifier amplifies the current signal from the sensor module; the bandpass filter removes noise outside the human heart and lung audio band, and the human heart and lung audio band is 20-2500Hz; the industrial frequency notch filter removes the industrial frequency signal interference generated by the equipment in a 220V mains environment, and is set to 50Hz. The digital-to-analog conversion module converts the analog electrical signal from the signal preprocessing module into a digital signal, or converts the digital signal into an analog electrical signal, that is, the user's heart audio signal and the user's lung audio signal, and transmits it to the real-time playback module.

[0040] Optionally, the signal acquisition end also includes: a device main control module, a device information display module, a data storage module, a power supply module, a device wireless communication module and a real-time playback module. The device main control module is used to control the device's power on and off, cardiopulmonary sound recording, wireless data transmission, data storage, and volume adjustment functions. The device information display module is a display screen of the signal acquisition end, which displays the device's volume, power, wireless transmission status, and data storage status. The data storage module is used to store the recorded cardiopulmonary audio in a TF (Trans Flash) card or SD (Secure Digital Memory) card. The power supply module provides power for the circuit operation of the signal acquisition end, including a lithium-ion rechargeable battery. The device wireless communication module is used to establish a wireless connection between the signal acquisition end and the ordinary user end, and to realize the wireless transmission of audio data between the two. Wireless connection and wireless data transmission methods include but are not limited to WIFI (Wireless Fidelity, wireless local area network), Bluetooth, ZigBEE (ZigBee, a synonym for IEEE 802.15.4 protocol), NFC (Near Field Communication, near-field wireless communication), and MTC (Machine Type Communication) communication transmission methods. The real-time playback module is used to play the current auscultation cardiopulmonary audio in real time, so that users or expert consultants can listen through wired headphones.

[0041] Optionally, the mobile portable Internet of Things cardiopulmonary sound auscultation system also includes a general user terminal.

[0042] Combination Figure 2As shown, in some embodiments, the wireless connection process between the common user terminal and the signal acquisition terminal includes the following steps:

[0043] Step S201, determine whether Bluetooth is turned on; if Bluetooth is not turned on, execute step S202, and if Bluetooth is turned on, execute step S204;

[0044] Step S202, entering the homepage without Bluetooth status;

[0045] Step S203, click the stethoscope connection button, and then return to step S201;

[0046] Step S204, entering the Bluetooth status page;

[0047] Step S205, click the stethoscope connection button, and then execute step S206;

[0048] Step S206, determining whether there is a connectable stethoscope nearby; if there is no connectable stethoscope nearby, executing step S207, and if there is a connectable stethoscope nearby, executing step S208;

[0049] Step S207, search again, and then return to step S206;

[0050] Step S208, select a stethoscope for connection;

[0051] Step S209: confirm that the stethoscope is connected.

[0052] Optionally, recording the user's heart and lungs by auscultation to obtain the first user's auscultation report also includes: when recording the user's heart and lungs by auscultation, recording the auscultation response and condition description of any part of the user's body to obtain the user's real-time evaluation. By recording the auscultation response and condition description of the current part in real time during the auscultation recording process, the auscultation report can be made more complete, thereby providing a more comprehensive auxiliary diagnosis reference opinion and further improving the accuracy of the diagnosis.

[0053] In some embodiments, when the user is recording auscultation, if he or she has any reaction to a certain part or needs to describe any feelings, he or she can promptly record them in writing in the real-time evaluation; after the first user auscultation report is formed, the pathological auscultation sound components of the heart and lung sounds of a single part or multiple parts can be analyzed, helping users to improve the efficiency of utilizing their heart and lung auscultation sounds from a more scientific, objective, and comprehensive perspective, and to more completely evaluate the health of their cardiopulmonary system.

[0054] Optionally, an intelligent analysis is performed on the user's heart audio signal and the user's lung audio signal to obtain an intelligent analysis report for each single part, including: bandpass filtering each audio signal and performing noise reduction processing on each filtered audio signal; extracting features from each noise-reduced audio signal to obtain features of each part; classifying the features of each part using a preset audio signal processing model to obtain abnormal situation identification results of each audio signal; and obtaining an intelligent analysis report for each single part based on the abnormal situation identification results of each audio signal.

[0055] Optionally, the user heart audio signal and the user lung audio signal are band-pass filtered by band-pass filters respectively, so that the filtered user heart audio signal is at 20-600 Hz, and the filtered user lung audio signal is at 50-2000 Hz.

[0056] Optionally, each filtered audio signal is subjected to noise reduction processing by wavelet packet decomposition noise reduction and clustered empirical mode decomposition and reconstruction.

[0057] Optionally, one or more of wavelet coefficient energy, Mel-frequency cepstral coefficients, empirical mode decomposition, the ratio of Shannon entropy maximum values ​​to the difference between the maximum values, LBP (Local Binary Pattern) frequency histogram sequence, spectral roll-off rate or spectral centroid are used to extract features from each audio signal after noise reduction processing.

[0058] Optionally, the audio signal processing model is obtained by training one or more of a random forest, a logistic regression, a convolutional neural network, a recurrent neural network, a Gaussian mixture model, or a BP (back propagation) genetic neural network.

[0059] Optionally, each single-part intelligent analysis report is obtained according to the abnormal situation identification result of each audio signal, including: processing the abnormal situation identification result of each audio signal into preset text and format to form each single-part intelligent analysis report.

[0060] In some embodiments, since heart sounds are mostly concentrated in 20-600Hz and lung sounds are mostly concentrated in 50-2000Hz, different bandpass filters are used to filter the heart sounds and lung sounds to obtain user heart audio signals at 20-600Hz and user lung audio signals at 50-2000Hz; the filtered user heart audio signals and user lung audio signals are denoised by wavelet packet decomposition denoising and clustering empirical mode decomposition and reconstruction for physiological signals; in feature extraction, since the preliminary diagnosis of the patient's condition in medicine is based on pathological heart and lung sounds, it depends to a large extent on the frequency, amplitude, duration and timing of occurrence of pathological heart and lung sounds. In order to reflect these physical properties, wavelet coefficient energy, Mel cepstral coefficient, empirical mode decomposition, Shannon entropy maximum value ratio and maximum value difference, LBP frequency, etc. are used. One or more of histogram sequence, spectral roll-off rate, and spectral centroid are used for feature extraction. The proposed features can better reflect the frequency, amplitude, duration, and timing of occurrence of pathological cardiopulmonary sounds, and thus can more practically and intuitively reflect the frequency domain, time domain, and amplitude characteristics of pathological cardiopulmonary sounds; one or more of random forest, logistic regression, convolutional neural network, recurrent neural network, Gaussian mixture model, and BP genetic neural network that process physiological timing signals and spectral features are used as machine learning classifiers, and the machine learning classifier is trained to obtain an audio signal processing model; then, the features of each part are classified and processed through the trained audio signal processing model to obtain the abnormal situation recognition results of each audio signal, and the abnormal situation recognition results of each audio signal are processed in preset text and format to form intelligent analysis reports for each single part.

[0061] Optionally, a multi-part intelligent analysis report is obtained based on each single-part intelligent analysis report, including: processing each single-part intelligent analysis report according to a preset auscultation order and a preset vector matrix format to obtain a vector matrix; inputting the vector matrix into a preset intelligent analysis report processing model to obtain a multi-part intelligent analysis report.

[0062] Optionally, the intelligent analysis report processing model is obtained by training a recurrent neural network, in which nodes between hidden layers are connected, and the input of each current hidden layer includes the output of the input layer and the output of the hidden layer at the previous moment.

[0063] In some embodiments, the results of the intelligent analysis of each single part are arranged in a unified order of auscultation and used as the input of the intelligent analysis report processing model in the form of a standardized vector matrix; the recurrent neural network is the classifier of the intelligent analysis report processing model. The recurrent neural network will memorize the previous information and apply it to the calculation of the current output, that is, the nodes between the hidden layers are no longer disconnected but connected, and the input of the hidden layer includes not only the output of the input layer but also the output of the hidden layer at the previous moment. In this way, when processing the sequence vector, both the input and the output are taken into account, which improves the classification effect of the sequence vector, and then improves the accuracy of the intelligent analysis report processing model, so that a more accurate multi-part intelligent analysis report can be obtained.

[0064] Optionally, the second user's auscultation report is analyzed using several algorithm models, and before obtaining the algorithm model analysis report, it also includes: screening out several candidate algorithm models based on the second user's auscultation report; each candidate algorithm model includes several candidate algorithm models of the same type of cardiopulmonary disease and candidate algorithm models of different types of cardiopulmonary disease; obtaining the correlation between each candidate algorithm model and each type of cardiopulmonary disease, and the accuracy of each candidate algorithm model; selecting several algorithm models based on the correlations and accuracies.

[0065] Optionally, each algorithm model to be selected corresponds to a cardiopulmonary disease type, and each cardiopulmonary disease type corresponds to a disease field. The second user's auscultation report contains disease fields or cardiopulmonary disease types predicted by intelligent analysis. The disease fields or cardiopulmonary disease types in the second user's auscultation report are matched with the disease fields or cardiopulmonary disease types corresponding to the algorithm model to be selected to obtain several algorithm models to be selected.

[0066] Optionally, the correlation between the algorithm model to be selected and each type of cardiopulmonary disease and the accuracy of the algorithm model to be selected are obtained, including: counting the evaluations of the algorithm model to be recommended by multiple users who have used the algorithm model to be recommended, and obtaining the correlation between the algorithm model to be selected and each type of cardiopulmonary disease and the accuracy of the algorithm model to be selected.

[0067] Optionally, the algorithm model is a whole that combines different auscultation sound features with different machine learning models; using several algorithm models to analyze the second user's auscultation report, that is, selecting different auscultation sound features, and combining different machine learning models to analyze one's own auscultation report.

[0068] In some embodiments, in order to reduce the differences in physical properties such as frequency domain, time domain, period, amplitude, etc. of different pathological sounds produced by different individuals who may suffer from cardiopulmonary diseases, different methods are used for feature extraction and different machine learning models are used for classification. Providing users with a variety of algorithm models to choose from can increase the choices for users to conduct self-diagnosis and improve the accuracy of users' self-examination.

[0069] Optionally, several algorithm models are selected based on the correlation between each candidate algorithm model and each type of cardiopulmonary disease, and the accuracy of each candidate algorithm model, including: determining the candidate algorithm model corresponding to the correlation greater than a first preset threshold as the selected algorithm model; and / or, determining the candidate algorithm model corresponding to the accuracy greater than a second preset threshold as the selected algorithm model.

[0070] Optionally, several algorithm models are selected based on the correlation between each candidate algorithm model and each type of cardiopulmonary disease, as well as the accuracy of each candidate algorithm model, including: sorting each recommended algorithm in descending order based on the correlation between each candidate algorithm model and each type of cardiopulmonary disease, as well as the accuracy of each candidate algorithm model; and selecting several algorithm models from each candidate algorithm model through manual selection by the user.

[0071] In some embodiments, the user can select an algorithm model by manual selection or intelligent recommendation. The intelligent recommendation module of the algorithm model refers to the system reading the self-intelligent analysis results in the second user's auscultation report, matching the disease field or cardiopulmonary disease type in the second user's auscultation report, obtaining several candidate algorithm models, and arranging the list of candidate algorithm models in descending order according to the correlation and accuracy between the cardiopulmonary disease type in the second user's auscultation report and the candidate algorithm model.

[0072] Optionally, before conducting a multi-expert remote consultation on the algorithm model analysis report and the second user's auscultation report, it also includes: obtaining the disease field involved in the second user's auscultation report; screening out a number of candidate expert consultants in the disease field; obtaining the popularity of each candidate expert consultant; and selecting multiple expert consultants based on the popularity of each candidate expert consultant.

[0073] Optionally, obtaining the favorable comments of the candidate expert consultants includes: counting comments of multiple users who have applied for consultations with the candidate expert consultants, and obtaining the favorable comments of each candidate expert consultant.

[0074] Optionally, multiple expert consultants are selected based on the popularity of each candidate expert consultant, including: arranging the candidate expert consultants in descending order according to the popularity of each candidate expert consultant; determining the candidate expert consultant corresponding to the popularity greater than a third preset threshold as the expert consultant; and / or selecting the expert consultant from the candidate expert consultants through manual selection by the user.

[0075] In some embodiments, users can select expert consultants through manual selection or intelligent recommendation. Among them, the intelligent recommendation of expert consultants means that the system reads the self-intelligent analysis results in the second user's auscultation report, and selects several candidate expert consultants in the disease field according to the disease field involved in the analysis results and the expert consultant's expertise. The evaluations of multiple users who have applied for consultations with candidate expert consultants are counted to obtain the favorable ratings of each candidate expert consultant, and the list of optional expert consultants is sorted in descending order according to the favorable ratings of each candidate expert consultant.

[0076] Optionally, a multi-expert remote consultation is conducted on the algorithm model analysis report and the second user auscultation report to obtain the expert analysis report, including: the user applies for a multi-expert remote consultation, and sends the algorithm model analysis report and the second user auscultation report to each expert consultant; after each expert consultant responds to the user's consultation application, the expert analysis report is sent to the user.

[0077] Through multi-expert remote consultation, the user's second-user stethoscope report can get more advice and guidance from expert consultants, achieving many-to-one doctor-patient connection, reducing reliance on the subjectivity and personal experience accumulation of expert consultants, reducing the possibility of misdiagnosis and missed diagnosis, and further ensuring a completely objective assessment of the user's cardiopulmonary health, so that users can obtain more scientific, objective and fair remote diagnosis and treatment results, while also being able to adapt to more complex doctor-patient interaction scenario requirements.

[0078] Optionally, the common user terminal in the mobile portable IoT cardiopulmonary sound auscultation system is open to all users, and the carriers of the common user terminal include but are not limited to smart phones, smart tablets, smart wearable devices, smart homes and other electronic devices.

[0079] Optionally, combined Figure 3As shown, the ordinary user terminal includes an account management module, a user information module, a qualification authentication module, a consultant terminal switching module, a user wireless connection module, a user network communication module, a real-time auscultation module, a history module and a my consultant module. The account management module is used for the user to log in to the account and log out of the account. The user information module is used for the user to fill in, save and display personal information by himself, and the personal information includes name, gender, date of birth and occupation. The qualification authentication module is used to apply for qualification authentication before the user obtains the right to use the consultant terminal. The consultant terminal switching module is used to switch between the ordinary user terminal and the consultant terminal when the user obtains the right to use the consultant terminal after passing the qualification authentication. The user wireless connection module establishes a wireless connection with the device wireless communication module of the signal acquisition terminal to realize the wireless transmission of audio data. The user network communication module establishes a wireless network connection with the cloud service platform to realize the transmission and reception of auscultation reports.

[0080] Optionally, the real-time auscultation module is used to perform real-time auscultation and audio recording of heart sounds or lung sounds, and form a first auscultation report and save it in the history recording module. The real-time auscultation module includes: a heart sound auscultation module and a lung sound auscultation module.

[0081] The heart sound auscultation module is used to perform real-time auscultation and audio recording of heart sounds in five parts, including M-mitral valve auscultation area, P-pulmonary valve auscultation area, A-aortic valve auscultation area, E-aortic valve second auscultation area, and T-tricuspid valve auscultation area, and record the evaluation or opinions on the signals in these parts.

[0082] The heart sound auscultation module includes: a real-time heart sound auscultation display unit, a real-time heart sound auscultation module and a real-time heart sound evaluation module. The real-time heart sound auscultation display unit is used to display the current auscultation site information each time the site auscultation interface is switched. The real-time heart sound auscultation module includes a heart sound auscultation schematic unit, a heart sound site selection unit, and a heart sound recording operation unit. Among them, the heart sound auscultation schematic unit is a picture display, showing a guide schematic diagram of the five sites of heart sound auscultation; the heart sound site selection unit allows users to select the auscultation and recording sites; the heart sound recording operation unit includes 4 control units: start recording, complete recording, re-recording, next site, and end auscultation. The real-time heart sound evaluation module is used to record the auscultation response and condition description of the current site during the real-time auscultation process to increase the completeness of the first auscultation report and provide more comprehensive auxiliary diagnosis reference opinions to the expert consultants of the comprehensive consultation; it can also be left blank.

[0083] The lung sound auscultation module is used to perform real-time auscultation and audio recording of lung sounds in six locations, including 1- the first and second intercostal spaces on the right chest near the manubrium, 2- the first and second intercostal spaces on the left chest near the manubrium, 3- the right side of the spine and the auscultation area of ​​the scapula, 4- the left side of the spine and the auscultation area of ​​the scapula, 5- near the right anterior axillary line, and 6- near the left anterior axillary line, and record the evaluation or opinions on the signals at those locations.

[0084] The lung sound auscultation module includes: a real-time lung sound auscultation display unit, a real-time lung sound auscultation module and a real-time lung sound evaluation module. The real-time lung sound auscultation display unit is used to display the current auscultation site information each time the site auscultation interface is switched. The real-time lung sound auscultation module includes a lung sound auscultation schematic unit, a lung sound site selection unit, and a lung sound recording operation unit. Among them, the lung sound auscultation schematic unit is a picture display, showing a guide schematic diagram of five sites for lung sound auscultation; the lung sound site selection unit allows users to select auscultation and recording sites; the lung sound recording operation unit includes 4 control units: start recording, complete recording, re-recording, next site, and end auscultation. The real-time lung sound evaluation module is used to record the auscultation response and condition description of the current site during the real-time auscultation process to increase the completeness of the first auscultation report and provide more comprehensive auxiliary diagnosis reference opinions to the expert consultants of the comprehensive consultation. It can also be left blank.

[0085] The history record module is used to record the first auscultation report generated by real-time auscultation. The history record module includes a history record display unit and an auscultation report module. The history record display unit arranges and displays the auscultation mode and auscultation time formed by the auscultation report; the auscultation report module is used to record the detailed audio and real-time evaluation opinions of each part recorded by the user each time the auscultation is performed, including a heart sound auscultation report module and a lung sound auscultation report module.

[0086] In some embodiments, the heart sound auscultation report module stores each heart sound auscultation report record, including a heart sound report display unit, a single-site heart sound report module, a heart sound intelligent analysis module, and a heart sound evaluation module. Among them, the heart sound report display unit is used to display the user's personal information recorded in this auscultation recording and a preview of the single-site heart sound report; the single-site heart sound report module is used to record the details of the auscultation of a single site in this heart sound auscultation recording; the heart sound intelligent analysis module is used to analyze whether there are abnormal heart sounds in a single site, and whether the comprehensive situation of multiple sites is related to diseases; the heart sound evaluation module is used to display the real-time evaluation of each site and the evaluation of multiple sites during this real-time auscultation process, and the evaluation is modifiable content.

[0087] The single-site heart sound report module is used to provide the playback function of the audio recorded in this site, as well as waveform display, intelligent analysis and real-time evaluation functions. The single-site heart sound report module includes a single-site heart sound report operation unit, a single-site heart sound information display unit, a single-site heart sound waveform display unit, a single-site heart sound evaluation display unit and a single-site heart sound analysis display unit. The single-site heart sound report operation unit is used to provide the deletion and modification function of the single-site report, and the modifiable content is the real-time evaluation of the single site; the single-site heart sound information display unit is used to display the personal information recorded in this auscultation report; the single-site heart sound waveform display unit is used to display the audio time domain waveform when the single-site recording is played back; the single-site heart sound evaluation display unit is used to display the real-time evaluation recorded when the single-site recording was recorded; the single-site heart sound analysis display unit is used to perform intelligent analysis on whether the auscultation audio of the single site has abnormal heart sounds.

[0088] The intelligent heart sound analysis module calls the algorithm of the artificial intelligence analysis module of the cloud service platform to analyze whether there are abnormal heart sounds in a single auscultation site and whether the comprehensive situation of multiple sites is related to a certain respiratory disease. The intelligent heart sound analysis module includes a multi-site heart sound intelligent analysis module, a multi-site heart sound analysis display unit, a single-site heart sound intelligent analysis module and a single-site heart sound analysis display unit. The multi-site heart sound intelligent analysis module is used to analyze whether the comprehensive situation of multiple sites is related to a certain respiratory disease; the multi-site heart sound analysis display unit is used to display the multi-site intelligent analysis results; the single-site heart sound intelligent analysis module is used to analyze whether there are abnormal heart sounds in a single auscultation site; the single-site heart sound analysis display unit is used to display the single-site intelligent analysis results.

[0089] The heart sound evaluation module is used to provide the user with the function of describing and evaluating the single-site auscultation audio or the overall multi-site auscultation situation in the auscultation report. The heart sound evaluation module includes a multi-site heart sound evaluation module, a multi-site heart sound evaluation display unit, a single-site heart sound evaluation module and a single-site heart sound evaluation display unit. The multi-site heart sound evaluation module is used to provide the function of describing and evaluating the overall multi-site auscultation situation; the multi-site heart sound evaluation display unit is used to display the user's description and evaluation content of the overall multi-site auscultation situation; the single-site heart sound evaluation module is used to provide the function of describing and evaluating the single-site auscultation situation; the single-site heart sound evaluation display unit is used to display the user's description and evaluation content of the single-site auscultation situation.

[0090] In some embodiments, the lung sound auscultation report module is used to store each lung sound auscultation report record. The lung sound auscultation report module includes a lung sound report display unit, a single-site lung sound report module, a lung sound intelligent analysis module, and a lung sound evaluation module. Among them, the lung sound report display unit is used to display the user's personal information recorded in this auscultation recording and a preview of the single-site lung sound report; the single-site lung sound report module is used to record the auscultation details of a single site in this lung sound auscultation recording; the lung sound intelligent analysis module is used to analyze whether there are rales in a single site and whether the comprehensive situation of multiple sites is related to diseases; the lung sound evaluation module is used to display the real-time evaluation of each site and the evaluation of multiple sites during this real-time auscultation process, which is a modifiable content.

[0091] The single-site lung sound reporting module is used to provide the playback function of the recorded audio of the site, as well as waveform display, intelligent analysis and real-time evaluation functions. The single-site lung sound reporting module includes a single-site lung sound reporting operation unit, a single-site lung sound information display unit, a single-site lung sound waveform display unit, a single-site lung sound evaluation display unit and a single-site lung sound analysis display unit.

[0092] The single-site lung sound report operation unit is used to provide the functions of deleting and modifying the single-site report, and the modifiable content is the real-time evaluation of the single site; the single-site lung sound information display unit is used to display the personal information recorded in this auscultation report; the single-site lung sound waveform display unit is used to display the audio time domain waveform during the playback of this single-site recording; the single-site lung sound evaluation display unit is used to display the real-time evaluation recorded during the recording of this single-site recording; the single-site lung sound analysis display unit is used to perform intelligent analysis on the presence or absence of rales in the auscultation audio of a single site.

[0093] The intelligent lung sound analysis module calls the algorithm of the artificial intelligence analysis module of the cloud service platform to analyze whether there are rales in a single auscultation site and whether the comprehensive situation of multiple sites is related to a certain respiratory disease. The intelligent lung sound analysis module includes a multi-site lung sound intelligent analysis module, a multi-site lung sound analysis display unit, a single-site lung sound intelligent analysis module and a single-site lung sound analysis display unit. The multi-site lung sound intelligent analysis module is used to analyze whether the comprehensive situation of multiple sites is related to a certain respiratory disease; the multi-site lung sound analysis display unit is used to display the multi-site intelligent analysis results; the single-site lung sound intelligent analysis module is used to analyze whether there are rales in a single auscultation site; the single-site lung sound analysis display unit is used to display the single-site intelligent analysis results.

[0094] The lung sound evaluation module is used to provide the user with the function of describing and evaluating the single-site auscultation audio or the overall multi-site auscultation situation in the auscultation report. The lung sound evaluation module includes a multi-site lung sound evaluation module, a multi-site lung sound evaluation display unit, a single-site lung sound evaluation module and a single-site lung sound evaluation display unit. The multi-site lung sound evaluation module is used to provide the function of describing and evaluating the overall multi-site auscultation situation; the multi-site lung sound evaluation display unit is used to display the user's description and evaluation content of the overall multi-site auscultation situation; the single-site lung sound evaluation module is used to provide the function of describing and evaluating the single-site auscultation situation; the single-site lung sound evaluation display unit is used to display the user's description and evaluation content of the single-site auscultation situation.

[0095] My Consultant module is used for users to communicate with expert consultants. My Consultant module includes Consultant Address Book module, Consultant Management module, Consultant Instant Communication module and Comprehensive Consultation module. Consultant Address Book module is used to list the added expert consultants; Consultant Management module is used to search, add or delete expert consultants; Consultant Instant Communication module is used for users to have window dialogues with expert consultants; Comprehensive Consultation module is used for users to apply for multi-expert remote consultation or multi-algorithm model analysis, i.e. artificial intelligence analysis, for their own auscultation reports, and manage the analysis results.

[0096] The consultant management module includes the consultant search module, the consultant addition module and the consultant deletion module. The consultant search module is used by users to access the expert consultant information library module of the cloud service platform to search for expert consultants by name; the consultant addition module is used by users to add expert consultants who have not been added to their own consultant address book module; the consultant deletion module is used by users to delete expert consultants who have been added to their own consultant address book module.

[0097] The comprehensive consultation module includes a report selection module, a consultant selection module, an algorithm selection module, an account management module and a consultation opinion management module. The report selection module is used to select an auscultation report from the user's historical records for analysis application; the consultant selection module is used to select one or more expert consultant users from the user's consultant address book module for analysis application; the algorithm selection module is used to select one or more algorithm models from the intelligent algorithm information library module of the cloud service platform for analysis application; the account management module is used for users to settle the comprehensive consultation fee, or to recharge and store cash in their own accounts; the consultation opinion management module is used for users to receive and manage feedback from comprehensive consultations.

[0098] The consultant selection module includes the consultant recommendation module and the self-selected consultant module. The consultant recommendation module is used to call the symptom information entered by the user in the auscultation report applied for, and after analysis by the intelligent matching module of the cloud service platform and the expert consultant information library module, a recommended list of expert consultants for consultation is given; the self-selected consultant module is used for users to select one or more expert consultants in their own consultant address book module for analysis application.

[0099] The algorithm selection module includes an algorithm recommendation module and a self-selected algorithm module. The algorithm recommendation module is used to call the symptom information entered by the user in the auscultation report applied for, and after analysis by the intelligent algorithm information library module of the cloud service platform, a recommended list of algorithm models for selection is given; the self-selected algorithm module is used for users to select one or more algorithm models for analysis and application in the intelligent algorithm information library module of the cloud service platform.

[0100] The account management module includes a fee settlement module, an account storage module and an account recharge module. The fee settlement module is used by the user to calculate the total consultation fee and make payment after the user selects the expert consultant and algorithm model that needs to be analyzed for the comprehensive consultation; the account storage module is used to store the cash recharged by the user; the account recharge module is used by the user to recharge the self-defined amount of cash into the account storage module.

[0101] The consultation opinion management module includes an opinion prompt module, an opinion storage module, an opinion reading module, an opinion reply module and an opinion evaluation module. The opinion storage module is used to feed back consultation opinions to the opinion storage module when there are consultation opinions, and generate new message prompts through the opinion prompt module to remind users to check the consultation results in time; the opinion reading module is used for users to read the opinions of the comprehensive consultation; the opinion reply module is used for users to input their replies to the consultation opinions, and the replies are stored in the opinion history record module of the consultant end of the expert consultant; the opinion evaluation module is used for users to give evaluation opinions on the consultation results.

[0102] Optionally, after responding to the user's consultation request, each expert consultant sends an expert analysis report to the user, including: each expert consultant responds to the user's consultation request, receives the algorithm model analysis report and the second user's auscultation report; analyzes the algorithm model analysis report and the second user's auscultation report, gives expert diagnosis opinions, and forms an expert analysis report; and sends the expert analysis report to the user.

[0103] Optionally, the portable IoT cardiopulmonary sound auscultation system also includes an expert consultant terminal. After qualification certification, the user can switch the module from the consultant terminal of the ordinary user terminal to enter the expert consultant terminal. The consultant terminal switching module includes an identity switching unit and an expert consultant terminal; the user performs the identity switching operation between the user and the expert consultant in the identity switching unit.

[0104] The expert consultant terminal includes the consultant account management module, consultant network communication module, general user terminal switching module, my customer module and consultant account management module. The consultant account management module is used to manage the expert consultant's login or logout operation from the consultant terminal, the consultant network communication module is used by the expert consultant to establish wireless network communication with the cloud service platform; the general user terminal switching module is the same as the expert consultant switching from the consultant terminal to the general user terminal; the my customer module is used by the expert consultant to manage the users he added, and interact with them for instant information exchange or auscultation report consultation; the consultant account management module is used by the expert consultant to collect the consultation application fee from the user after the consultation, and store it in his own network account.

[0105] The consultant account management module includes a fee receiving module and a consultant account storage module. The fee receiving module is used for the expert consultant to collect the corresponding consultation fee after accepting a certain auscultation report; the consultant account storage module is used for storing the consultation fee collected by the expert consultant in this module.

[0106] My Customer Module includes Customer Address Book Module, Customer Management Module, Customer Instant Communication Module and Consultation Management Module. Customer Address Book Module is used to list the added users; Customer Management Module is used to search, add or delete users; Customer Instant Communication Module is used for expert consultants to have window dialogues with users; Consultation Management Module is used for expert consultants to process their own pending consultation applications, manage the received second user's auscultation reports, and record their own consultation opinions.

[0107] The consultation management module includes a consultation application processing module, a user report management module and an opinion history module. The consultation application processing module is used by expert consultants to receive auscultation report consultation applications sent by users through the cloud service platform, and decide whether to accept this consultation application. If the application is accepted, the auscultation report is downloaded to the user report management module for storage and record; if the application is not accepted, the reason for rejection is filled in, and the reason and the corresponding consultation fee are returned to the general user end; the user report management module is used for expert consultants to evaluate or reject the accepted user's auscultation report, and feedback to the general user end through the cloud service platform; the opinion history module is used to record the expert consultant's opinions and evaluations on the accepted consultation applications, and form a historical record for expert consultants to view at any time.

[0108] The consultation application processing module includes a consultation prompt module and a consultation application processing unit. The consultation prompt module is used to generate a new message prompt when there is a new consultation application; the consultation application processing unit is used for the expert consultant to accept or reject the received auscultation report.

[0109] The user report management module includes a user report display unit, a user report storage module, a user report evaluation module and an opinion upload module. The user report display unit is used to display the brief information of the accepted user's auscultation report in the form of a list; the user report storage module is used to store the accepted user's auscultation report; the user report evaluation module is used to make evaluation opinions on the accepted auscultation report; the opinion upload module is used to upload the evaluation opinions on the auscultation report to the cloud service platform and feedback to the user corresponding to the auscultation report.

[0110] Optionally, the mobile portable IoT cardiopulmonary sound auscultation system also includes a cloud service platform. The cloud service platform is used for information exchange between ordinary users and expert consultants, a carrier for wireless transmission of auscultation reports, and provides access to symptom feature information databases, disease information databases, expert consultant information databases, and artificial intelligence information databases. It also provides symptom and disease matching, expert consultant matching, artificial intelligence algorithm model analysis, qualification review, and temporary storage of fees and funds.

[0111] Optionally, the cloud service platform includes: a cloud network communication module, an expert consultant information library module, a stethoscope report library module, an intelligent algorithm information library module, an artificial intelligence analysis module, a qualification review module and a temporary fund storage module.

[0112] The cloud network communication module is used to connect and access the wireless network. The expert consultant information library module is used to provide information access functions for expert consultant end users. In some embodiments, when a user needs to add an expert consultant, the consultant search module will be used to access the corresponding expert consultant information in the expert consultant information library module, or when selecting the consultant recommendation module in the comprehensive consultation, the expert consultant information library module will be accessed according to the user's symptom description label to obtain matching opinions. The auscultation report library module is used to store auscultation reports saved by users. The intelligent algorithm information library module is used to store a large number of packaged and directly usable cardiopulmonary sound processing algorithm models. In some embodiments, when a user uses the algorithm recommendation module in the comprehensive consultation application process, the system matches the recommended algorithm model selection opinions in the intelligent algorithm information library module of the cloud service platform according to the symptom description label selected by the user, and the correlation, solution and praise of different algorithm models with related symptoms, and feedbacks them to the user.

[0113] The artificial intelligence analysis module is used to detect and identify whether there are pathological heart and lung sounds in the auscultation audio of a single part, or whether there is a possibility of a certain heart and lung disease in the whole part. In some embodiments, when the user uses the heart sound intelligent analysis module or the lung sound intelligent analysis module in the auscultation report, or when the algorithm is selected in the comprehensive consultation to apply for analysis, the artificial intelligence analysis module of the cloud service platform calls the algorithm model for analysis, performs signal denoising preprocessing, time-frequency domain comprehensive feature extraction, and machine learning model training on the auscultation audio of a single part, and obtains the recognition result of whether there are pathological heart and lung sounds in the part, and inputs the analysis results of each single part as a comprehensive feature sequence into the dedicated algorithm model for heart and lung disease identification, and obtains the possibility of the whole body suffering from a certain heart and lung disease, and feeds back the single part analysis results and the overall analysis results to the user.

[0114] The qualification review module is used to review the expert consultant qualification certification materials uploaded by the user and feedback the review opinions to the user. In some embodiments, the user uploads the expert consultant certification materials to the qualification review module of the cloud service platform through the qualification certification module, and the system backend administrator reviews them; if the review is passed, the passing opinions are fed back to the applicant's qualification certification module for prompting, and the user is granted the right to use the expert consultant terminal; if the review is not passed, the opinions and reasons for the failure are fed back to the user's qualification certification module for prompting.

[0115] The funds temporary storage module is used to temporarily store the fees paid by the user when applying for a comprehensive consultation. In some embodiments, when a user applies for a comprehensive consultation, the consultation fee is temporarily paid to the funds temporary storage module of the cloud service platform for temporary storage; when the comprehensive consultation application receives an expert consultant's reply or the analysis result of the algorithm model, and is fed back to the user's comprehensive consultation module, the fees pre-stored in the funds temporary storage module are paid to the expert consultant or the platform providing the algorithm model; when the comprehensive consultation application is rejected by the expert consultant or the analysis result of the algorithm model is not obtained, the fees in the funds temporary storage module will be returned to the user's account storage module.

[0116] In some embodiments, in combination Figure 4As shown, a mobile portable IoT heart and lung sound auscultation system includes a signal acquisition terminal, a common user terminal, a cloud service platform, and an expert consultant terminal. The signal acquisition terminal collects the user's heart audio signal and the user's lung audio signal, and sends them to the common user terminal; the common user terminal receives the user's heart audio signal and the user's lung audio signal, and uses the cloud service platform to obtain the algorithm model analysis report and the second user auscultation report; after the common user terminal applies for expert consultation, the algorithm model analysis report and the second user auscultation report are sent to the cloud service platform; after the cloud service platform receives the response from the expert consultant terminal, the algorithm model analysis report and the second user auscultation report are sent to the expert consultant terminal, and the expert consultant of the expert consultant terminal analyzes the algorithm model analysis report and the second user auscultation report to obtain the expert analysis report, and sends the expert analysis report to the user through the cloud service platform.

[0117] Combination Figure 5 As shown, in some embodiments, a payment and recharge process of a mobile portable IoT cardiopulmonary sound auscultation system includes the following steps:

[0118] Step S301, enter the "account management module";

[0119] Step S302: After the user confirms payment, the total consumption is calculated;

[0120] Step S303, compare the total consumption with the balance in the "account storage module";

[0121] Step S304, determine whether the total consumption is less than or equal to the balance; if the total consumption is less than or equal to the balance, execute step S305; otherwise, execute step S310;

[0122] Step S305: Payment is successful, and a message is displayed indicating that the consultation opinion is being replied to. The fee is submitted to the "Fund Temporary Storage Module" of the "Cloud Service Platform";

[0123] Step S306, the comprehensive consultation application is uploaded to the "Auscultation Report Library Module" of the "Cloud Service Platform";

[0124] Step S307, generating a consultation reminder in the "consultation application processing module" in the "expert consultant terminal" of the selected consultant;

[0125] Step S308: The "intelligent algorithm information library module" in the "cloud service platform" calls the selected algorithm, and performs calculation and analysis in the "artificial intelligence analysis module";

[0126] Step S309: Return the generated algorithm analysis result and send it to the "consultation opinion management module" of the "ordinary user terminal" for storage;

[0127] Step S310: If payment fails, a prompt box pops up to remind the user to recharge and go to the "account recharge module";

[0128] Step S311, select a recharge method;

[0129] Step S312, select the recharge amount;

[0130] Step S313: access the bound bank card, WeChat account or Alipay account;

[0131] Step S314, determine whether the recharge amount is less than or equal to the account balance; if the recharge amount ≤ the account balance, execute step S315, otherwise execute step S316;

[0132] Step S315: Recharge is successful, and the total consumption is compared with the balance in the "account storage module" again;

[0133] Step S316: If the recharge fails, a prompt box pops up to remind the user that the balance is insufficient.

[0134] In some embodiments, a real-time auscultation process of a mobile portable IoT cardiopulmonary sound auscultation system includes the following steps:

[0135] Step S401, enter "ordinary user terminal";

[0136] Step S402, complete the login in the "Account Management Module";

[0137] Step S403, entering the "user wireless connection module" to complete the wireless connection between the user and the device;

[0138] Step S404, enter the "real-time auscultation module";

[0139] Step S405, select "heart sound auscultation module" or "lung sound auscultation module" to auscultate heart sounds or lung sounds;

[0140] Step S406: After the auscultation is completed, a first user auscultation report is generated and saved in the "history record module".

[0141] Combination Figure 6 As shown, in some embodiments, a comprehensive consultation application process for a mobile portable IoT cardiopulmonary sound auscultation system includes the following steps:

[0142] Step S501, enter the "ordinary user terminal";

[0143] Step S502, complete the login in the "Account Management Module";

[0144] Step S503, enter "My Consultant Module";

[0145] Step S504, enter the "comprehensive consultation module";

[0146] Step S505, enter the "report selection module";

[0147] Step S506, accessing the "auscultation report module", the user manually selects an auscultation report in the "heart sound auscultation report module" or the "lung sound auscultation report module";

[0148] Step S507, enter the "consultant selection module";

[0149] Step S508, determine whether to select a recommended consultant or a self-selected consultant; if a recommended consultant is selected, execute step S509; if a self-selected consultant is selected, execute step S511;

[0150] Step S509, select "recommendation consultant module";

[0151] Step S510: Comprehensively analyze the activity, title, location, reputation, areas of expertise, and consultation price of consultants in the "Expert Consultant Information Database Module" in the "Cloud Service Platform", and then provide a list of recommended consultants;

[0152] Step S511, call the "consultant address book module" and select the added consultant, or search for other consultants and select them by calling the "consultant addition module"; execute step S514;

[0153] Step S512, select "self-selected consultant module";

[0154] Step S513, calling the "Expert Consultant Information Database Module" in the "Cloud Service Platform" and displaying the list of expert consultants corresponding to the search keyword;

[0155] Step S514, enter "algorithm selection module";

[0156] Step S515, determine whether to select a recommended algorithm or a self-selected algorithm; if the recommended algorithm is selected, execute step S516; if the self-selected algorithm is selected, execute step S519;

[0157] Step S516, select "algorithm recommendation module";

[0158] Step S517: Comprehensively analyze the relevance and accuracy of various algorithm models in the "intelligent algorithm information library module" in the "cloud service platform" and give a recommended algorithm ranking;

[0159] Step S519, select "self-selected algorithm recommendation module";

[0160] Step S520: calling the "intelligent algorithm information library module" in the "cloud service platform" and displaying the algorithm information corresponding to the search keyword;

[0161] Step S521, enter the "account management module", and after successfully completing the payment or recharge process, check the returned consultation opinion in the "consultation opinion management module".

[0162] Users pay a fee to select one or more expert consultants or algorithm models and apply for analysis of their own heart and lung auscultation sounds. This can be used in a variety of complex doctor-patient interaction scenarios to achieve remote diagnosis and treatment and effective communication between users and expert consultants, making rich medical resources and advanced artificial intelligence technology better popularized in users' self-health management, and providing a mobile and more portable IoT heart and lung sound auscultation system.

[0163] Combination Figure 7 As shown, in some embodiments, a comprehensive consultation acceptance and processing process provided by the present invention includes the following steps:

[0164] Step S601, enter the "ordinary user terminal";

[0165] Step S602, complete the login in the "Account Management Module";

[0166] Step S603, switch to the "Expert Advisor Terminal" through the "Advisor Switching Module";

[0167] Step S604, enter "My Customer Module";

[0168] Step S605, entering the "consultation management module";

[0169] Step S606, entering the "consultation application processing module";

[0170] Step S607, previewing the consultation report sent by the user in the "auscultation prompt module";

[0171] Step S608, determine whether to accept the consultation application in the "consultation application processing module"; if the consultation application is accepted, execute step S609; otherwise, execute step S617;

[0172] Step S609, downloading the auscultation report to the "user report storage module";

[0173] Step S610, entering the "user report management module";

[0174] Step S611, reading the user's auscultation report in the "user report display unit";

[0175] Step S612: Make a detailed evaluation and guidance on the user's auscultation report through the "user report evaluation module";

[0176] Step S613, upload the opinions made to the "Auscultation Report Library Module" of the "Cloud Service Platform" through the "Opinion Upload Module", and update the applicant's current auscultation report evaluation;

[0177] Step S614, the "cloud service platform" establishes a connection with the applicant's "ordinary user terminal" through the "user network communication module", and updates the latest evaluation content of the current auscultation report stored in the "auscultation report library module" to the applicant's "opinion storage module";

[0178] Step S615: Generate a new message prompt symbol in the "opinion prompt module" to remind the applicant to check the evaluation opinions in time;

[0179] Step S616: The applicant gives an evaluation of the opinion in the "Opinion Evaluation Module" and uploads it to the "Expert Consultant Information Database Module" of the "Cloud Service Platform" to update the evaluation content of the corresponding consultant in real time;

[0180] Step S617, fill in the reason for rejection of the review application and submit it;

[0181] Step S619, deleting this consultation application, and returning the rejection reason to the applicant's "opinion storage module";

[0182] Step S620: Return the funds of the consultant price from the "Fund Temporary Storage Module" of the "Cloud Service Platform" to the "Account Storage Module" of the applicant, and prompt the applicant in the "Opinion Storage Module" that the refund has been received.

[0183] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A portable IoT heart and lung sound auscultation system, characterized in that: include: Perform auscultation recording of the user's heart and lungs to obtain a first user auscultation report; Perform intelligent analysis on the auscultation report of the first user to obtain intelligent analysis reports for each single part; Obtaining a multi-part intelligent analysis report based on each of the single-part intelligent analysis reports; Obtaining a second user auscultation report according to the first user auscultation report, each of the single-site intelligent analysis reports, and the multi-site intelligent analysis report; Analyzing the auscultation report of the second user using a plurality of algorithm models to obtain an algorithm model analysis report; Conducting multi-expert remote consultation on the algorithm model analysis report and the second user auscultation report to obtain an expert analysis report; Auscultation recording of the user's heart and lungs is performed to obtain the first user auscultation report, including: converting the vibration mechanical energy signals of the user's heart and lungs into corresponding current signals; Performing denoising preprocessing on each of the current signals to obtain each analog electrical signal; Performing digital-to-analog conversion on each of the analog electrical signals to generate a user heart audio signal and a user lung audio signal; determining the user heart audio signal and the user lung audio signal as a first user auscultation report; Perform intelligent analysis on the user's heart audio signal and the user's lung audio signal to obtain intelligent analysis reports for each single part, including: Performing bandpass filtering on each audio signal, and performing noise reduction processing on each filtered audio signal; Perform feature extraction on each audio signal after noise reduction processing to obtain features of each part; Using a preset audio signal processing model to classify and process the features of each part, and obtain an abnormality recognition result of each audio signal; Obtaining intelligent analysis reports for each single part according to the abnormal situation identification results of each of the audio signals; Obtaining a multi-part intelligent analysis report based on each of the single-part intelligent analysis reports includes: Processing each of the single-site intelligent analysis reports according to a preset auscultation order and a preset vector matrix format to obtain a vector matrix; The vector matrix is ​​input into a preset intelligent analysis report processing model to obtain a multi-part intelligent analysis report.

2. The system according to claim 1, characterized in that Auscultation recording of the user's heart and lungs is performed to obtain the first user auscultation report, which also includes: When auscultating the user's heart and lungs, the auscultatory response and condition description of any part of the user's body are recorded to obtain real-time evaluation from the user.

3. The system according to claim 1, characterized in that Analyzing the auscultation report of the second user by using a plurality of algorithm models, before obtaining the algorithm model analysis report, further comprising: Screening out a number of candidate algorithm models according to the auscultation report of the second user; each of the candidate algorithm models includes a number of candidate algorithm models of the same type of cardiopulmonary disease and candidate algorithm models of different types of cardiopulmonary disease; Obtaining the correlation between each candidate algorithm model and each cardiopulmonary disease type, and the accuracy of each candidate algorithm model; Several algorithm models are selected according to the respective relevances and the respective accuracies.

4. The system according to claim 3, characterized in that Several algorithm models are selected according to the respective relevances and the respective accuracies, including: Determine the candidate algorithm model corresponding to the correlation greater than the first preset threshold as the selected algorithm model; and / or, The candidate algorithm model corresponding to the accuracy greater than the second preset threshold is determined as the selected algorithm model.

5. The system according to claim 1, characterized in that Before performing multi-expert remote consultation on the algorithm model analysis report and the second user auscultation report, the method further includes: Obtaining the disease field involved in the auscultation report of the second user; Screening out a number of candidate expert consultants in the disease field; Obtaining the favorable reviews of each of the candidate expert consultants; A plurality of expert consultants are selected based on the popularity of each of the candidate expert consultants.

6. The system according to claim 1, characterized in that Conducting multi-expert remote consultation on the algorithm model analysis report and the second user auscultation report to obtain an expert analysis report, including: The user applies for multi-expert remote consultation, and sends the algorithm model analysis report and the second user auscultation report to each expert consultant; After responding to the user's consultation request, each of the expert consultants sends an expert analysis report to the user.

7. The system according to claim 6, characterized in that After responding to the user's consultation application, each of the expert consultants sends an expert analysis report to the user, including: Each of the expert consultants receives the algorithm model analysis report and the second user auscultation report in response to the user's consultation application; Analyze the algorithm model analysis report and the second user's auscultation report, give expert diagnosis opinions, and form an expert analysis report; The expert analysis report is sent to the user.

Citation Information

Cited By

  • Systems, devices, and methods for computer-aided auscultation

    WO2026174404A1