Embedded health management system based on voice analysis for traditional Chinese medicine diagnosis

Abstract

The invention provides an embedded health management system based on voice analysis for traditional Chinese medicine diagnosis. The embedded health management system comprises a sound collection module used for collecting illness state description voice of a user and preprocessing the illness state description voice; the feature extraction module is used for carrying out analysis and feature extraction on the non-interference voice; the traditional Chinese medicine diagnosis module is used for obtaining a traditional Chinese medicine diagnosis result through a pre-training diagnosis model based on the sound features and the description symptoms of the user; the result feedback module is used for providing preliminary traditional Chinese medicine dialectical suggestions to the user on the basis of the traditional Chinese medicine diagnosis result and in combination with the traditional Chinese medicine dialectical relationship describing the symptoms; the data storage and transmission module is used for storing and transmitting voice data and a diagnosis report of a user; and the health management module is used for formulating a personalized health management scheme. The traditional Chinese medicine diagnosis theory is combined with the modern voice analysis technology and the embedded technology, a new and objective method is provided for traditional Chinese medicine diagnosis, and the accuracy and efficiency of traditional Chinese medicine diagnosis are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of health management systems, and in particular to an embedded health management system based on speech analysis for traditional Chinese medicine diagnosis. Background Art

[0002] Traditional Chinese medicine diagnosis plays an important role in the medical field. Traditional Chinese medicine diagnosis mainly relies on doctors' observation, smell, questioning, palpation and other means. Among them, sound, as an external manifestation of human life activities, has a close relationship with internal organs, qi and blood in traditional Chinese medicine theory. However, the existing Chinese medicine diagnosis methods have some limitations in practical applications. For example, the doctor's diagnostic efficiency is limited by time and space, and it is difficult to monitor and diagnose a large number of users in real time; the judgment of the patient's voice characteristics is highly subjective, and there is a lack of objective and quantitative analysis methods; at the same time, the existing health management systems mostly focus on the monitoring of Western medicine indicators, and the integration of traditional Chinese medicine characteristics is insufficient. With the development of information technology, speech analysis technology has gradually matured, but its application in the embedded health management system of traditional Chinese medicine diagnosis is still imperfect. Therefore, the present invention proposes an embedded health management system based on speech analysis for traditional Chinese medicine diagnosis. Summary of the invention

[0003] The present invention provides an embedded health management system based on speech analysis for TCM diagnosis, which combines traditional TCM diagnosis theory with modern speech analysis technology and embedded technology, provides a new and objective method for TCM diagnosis, and improves the accuracy and efficiency of TCM diagnosis.

[0004] The present invention provides an embedded health management system based on speech analysis for traditional Chinese medicine diagnosis, comprising:

[0005] A sound collection module, used to collect the user's condition description voice based on a high-sensitivity microphone, and pre-process the voice to obtain interference-free voice;

[0006] The feature extraction module is used to analyze and extract features of non-interference speech through digital signal processing algorithms to obtain the user's voice features and describe symptoms;

[0007] The sound features include, but are not limited to, the fundamental frequency, harmonic structure, intonation, speech speed, loudness and timbre of the voice;

[0008] The TCM diagnosis module is used to obtain TCM diagnosis results based on the user's voice features and described symptoms through a pre-trained diagnosis model combined with the correspondence between voice features and TCM symptoms pre-stored in a large TCM diagnosis database;

[0009] The result feedback module is used to provide users with preliminary TCM syndrome differentiation suggestions based on the TCM diagnosis results and the TCM syndrome differentiation relationships of various symptoms, and output the corresponding diagnosis report for display;

[0010] Data storage and transmission module, used to store and transmit user's voice data and diagnostic reports;

[0011] The health management module is used to develop personalized health management plans for users based on their diagnostic storage records.

[0012] Preferably, in an embedded health management system based on speech analysis for TCM diagnosis, the feature extraction module comprises:

[0013] A speech recognition and analysis unit is used to recognize non-interference speech based on a preset speech recognition model, obtain the user's description semantics, analyze the user's description semantics based on the syndrome differentiation and treatment vocabulary library corresponding to the TCM diagnosis, and determine the description symptoms of the user's condition;

[0014] The sound feature extraction unit is used to extract sound signs of non-interference speech based on a preset sound feature removal model, and generate a user sound map according to the sound feature extraction result.

[0015] Preferably, in an embedded health management system based on speech analysis for TCM diagnosis, the TCM diagnosis module comprises:

[0016] The model diagnosis unit is used to analyze the user's voice spectrum and voice features based on the pre-trained diagnosis model, and to screen the voice features pre-stored in the large database of traditional Chinese medicine diagnosis to obtain similar voice features;

[0017] A result determination unit, used to obtain a preliminary TCM diagnosis result of the user based on the correspondence between the sound features and TCM symptoms pre-stored in the large TCM diagnosis database;

[0018] Based on the preliminary TCM diagnosis results, combined with the user's described symptoms and the TCM dialectical relationship between the various described symptoms, the TCM diagnosis result of the current user is determined.

[0019] Preferably, in an embedded health management system based on speech analysis for TCM diagnosis, the result feedback module includes:

[0020] A result analysis unit, used to determine the severity of the user's current illness based on the TCM diagnosis results and the user's description of the symptoms;

[0021] A feedback reminder unit, used to send a hospital diagnosis and treatment reminder to the user when the current severity of the disease is greater than a preset degree;

[0022] A feedback display unit is used to determine the user's current physical condition and potential health problems based on the TCM diagnosis and treatment results and the current severity of the disease, combined with the development trend of the user's description of the symptoms, when the current severity of the disease is less than or equal to a preset degree;

[0023] And based on the current physical condition and potential health problems, combined with clinical diagnosis and treatment records in a large database of traditional Chinese medicine diagnosis, generate traditional Chinese medicine treatment recommendations for the user;

[0024] Generates and displays a diagnostic report based on current physical condition, potential health issues, and TCM treatment recommendations.

[0025] Preferably, in an embedded health management system based on speech analysis for TCM diagnosis, the result analysis unit comprises:

[0026] The symptom classification subunit is used to determine the symptom attributes of the current user based on the TCM diagnosis results, and based on the symptom attributes, combined with the pre-stored symptom data in the TCM diagnosis large database, determine the possible accompanying symptoms corresponding to the TCM diagnosis results and the severity of the possible accompanying symptoms, and classify the accompanying symptoms according to the corresponding relationship between the accompanying symptoms and the TCM diagnosis elements to obtain sub-accompanying symptoms of multiple dimensions;

[0027] The result evaluation subunit is used to compare the similarity between the user's voice features and the matching similar features, the TCM symptoms corresponding to each matching similar feature, and the user's description features with each sub-companion symptom, and determine the symptom evaluation results of each dimension and their corresponding symptom manifestation scores according to the comparison results;

[0028] According to the preset weights corresponding to each dimension and the symptom performance scores corresponding to each dimension, the user's current symptom score result is calculated, and based on the score range corresponding to each symptom in the symptom severity assessment table and the current symptom score result, the user's current symptom severity is determined;

[0029] Among them, multiple dimensions include the nature of the pathogenic factors, the degree of damage to the internal organs, the strength of the positive energy, the complexity of the symptoms and the impact on daily life.

[0030] Preferably, in an embedded health management system based on speech analysis for TCM diagnosis, the feedback display unit comprises:

[0031] The diagnosis analysis subunit is used to infer the user's current physical state based on the user's diagnosis result and the current severity of the disease, combined with the user's initial physical state assessment result, and obtain the user's described symptoms and the performance association between the described symptoms;

[0032] According to the manifestation correlation between the described symptoms, the development trend of the user's described symptoms is predicted, and based on the development trend and the disease development process corresponding to the diagnosis result, new manifestations of the user's disease are predicted. Based on the TCM diagnosis and treatment logic, the user's potential health risk analysis is conducted in combination with the described symptoms and new manifestations to obtain the user's potential health problems;

[0033] It is suggested to generate a subunit for screening clinical diagnosis and treatment records in a large database of traditional Chinese medicine diagnosis based on the diagnosis results of traditional Chinese medicine, obtain relevant clinical diagnosis and treatment records, and obtain user body data corresponding to each personality clinical diagnosis and treatment record respectively;

[0034] Based on the user's physical data, determine the physical condition of the treated user corresponding to each relevant clinical diagnosis and treatment record, compare the physical condition of the treated user with the current physical condition of the current user, and determine users with the same physical condition as the current user;

[0035] Based on the target diagnosis and treatment records corresponding to the users with the same constitution, determine the potential similarity between the potential health problems of the users with the same constitution and the potential health problems of the current user, sort the availability of each target diagnosis and treatment record based on the similarity of the potential problems, and obtain a target sequence;

[0036] Obtain the TCM treatment suggestion error between each target diagnosis and treatment record in the target sequence and its corresponding adjacent target diagnosis and treatment record. When the TCM treatment suggestion error of the target sequence is always kept within a preset range, the TCM treatment suggestion of the user with the same constitution with the highest potential problem similarity in the target sequence is used as the TCM treatment suggestion for the current user;

[0037] Otherwise, the target sequence is divided into multiple subsequence sets based on a preset similarity interval, and the TCM treatment recommendations corresponding to the multiple subsequence sets are compared to obtain the differences in treatment recommendations between the subsequence sets. Based on the differences in treatment recommendations, the impact of potential health problem differences on the treatment recommendations is determined. According to the impact, the TCM treatment recommendations for the user with the same constitution who has the highest similarity to the potential problem in the target sequence are adjusted to obtain the TCM treatment recommendations for the current user.

[0038] Preferably, in an embedded health management system based on speech analysis for TCM diagnosis, the feedback display unit further includes:

[0039] The data management subunit is used for users to upload the latest physical examination data in various forms, and to evaluate the user's physical condition based on the latest physical examination data and the latest TCM diagnosis results stored in the cloud to obtain basic physical condition evaluation results;

[0040] Among them, the most recent TCM diagnosis result refers to the result of the most recent diagnosis from the current TCM diagnosis.

[0041] Preferably, in an embedded health management system based on speech analysis for TCM diagnosis, the data storage and transmission module includes:

[0042] A data transmission unit, used to send the user's voice data and diagnosis report to the user's family physician or health guardian;

[0043] The data storage unit is used to send the user's voice data and diagnostic report to the cloud for storage.

[0044] Preferably, in an embedded health management system based on speech analysis for TCM diagnosis, it is characterized in that the health management module comprises:

[0045] The record identification management unit is used to obtain all the uncured medical records stored by the user in the cloud, and determine the user's current symptoms and current disease type based on the final TCM diagnosis record;

[0046] And based on the user's current symptoms and current disease type, all unrecovered medical records are identified to obtain recovered medical records and unrecovered medical records, and the labels of the recovered unrecovered medical records are modified to recovered labels;

[0047] The health management unit is used to determine the duration of the current symptoms based on the unrecovered diagnosis and treatment records, classify various symptoms based on the duration of each symptom, and obtain the user's long-term persistent symptom group and short-term manifestation symptom group;

[0048] Based on the unrecovered diagnosis and treatment records and TCM treatment theories, we determined the targeted treatment suggestions corresponding to various symptoms, clustered and integrated the targeted treatment suggestions corresponding to the long-term persistent symptom group and the short-term manifestation symptom group, and obtained the continuous treatment suggestions and short-term treatment suggestions;

[0049] According to the dialectical relationship between various symptoms in TCM treatment, the continuous treatment suggestions and short-term treatment suggestions are integrated within the short-term symptom prediction treatment time to obtain a fusion plan. Based on the user's current physical condition and the principle of balanced nutrition, combined with the dietary taboos recommended by TCM treatment in the fusion plan, a recommended diet strategy for the user is generated;

[0050] At the same time, according to the user's current physical condition and the recommended amount of exercise for a normal human body, the user's appropriate exercise type is determined and a recommended exercise strategy is generated for the user;

[0051] Based on the fusion scheme, the user's recommended diet strategy and the user's recommended exercise strategy, a personalized health management plan for the user is generated and sent to the embedded wearable device display page for display.

[0052] Preferably, in an embedded health management system based on speech analysis for TCM diagnosis, the health management module further includes:

[0053] Automatic review unit, used to remind users to have a TCM review when the predicted treatment time of short-term symptoms is reached, to determine whether each symptom in the short-term symptom group has been alleviated;

[0054] If so, the user's physical condition is re-evaluated based on the user's current symptoms, and the user's recommended exercise strategy and the user's recommended diet strategy are adjusted according to the evaluation results and continuous treatment recommendations to obtain the latest recommended exercise strategy and recommended diet strategy;

[0055] The user's health management plan is updated and displayed based on ongoing treatment recommendations, the latest recommended exercise strategies, and recommended dietary strategies.

[0056] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures specifically pointed out in this application document.

[0057] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0059] Figure 1 This is a structural diagram of an embedded health management system based on speech analysis for traditional Chinese medicine diagnosis according to the present invention;

[0060] Figure 2 This is a structural diagram of a feature extraction module of an embedded health management system based on speech analysis for traditional Chinese medicine diagnosis according to the present invention;

[0061] Figure 3 This is a structural diagram of a TCM diagnosis module of an embedded health management system based on speech analysis for TCM diagnosis of the present invention;

[0062] Figure 4 This is a structural diagram of a result feedback module of an embedded health management system based on speech analysis for traditional Chinese medicine diagnosis according to the present invention;

[0063] Figure 5 This is a structural diagram of a data storage and transmission module of an embedded health management system based on speech analysis for traditional Chinese medicine diagnosis according to the present invention. DETAILED DESCRIPTION

[0064] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0065] In embodiment 1:

[0066] The present invention provides an embedded health management system based on speech analysis for traditional Chinese medicine diagnosis, such as Figure 1 As shown, including:

[0067] A sound collection module, used to collect the user's condition description voice based on a high-sensitivity microphone, and pre-process the voice to obtain interference-free voice;

[0068] The feature extraction module is used to analyze and extract features of non-interference speech through digital signal processing algorithms to obtain the user's voice features and describe symptoms;

[0069] The sound features include, but are not limited to, the fundamental frequency, harmonic structure, intonation, speech speed, loudness and timbre of the voice;

[0070] The TCM diagnosis module is used to obtain TCM diagnosis results based on the user's voice features and described symptoms through a pre-trained diagnosis model combined with the correspondence between voice features and TCM symptoms pre-stored in a large TCM diagnosis database;

[0071] The result feedback module is used to provide users with preliminary TCM syndrome differentiation suggestions based on the TCM diagnosis results and the TCM syndrome differentiation relationships of various symptoms, and output the corresponding diagnosis report for display;

[0072] Data storage and transmission module, used to store and transmit user's voice data and diagnostic reports;

[0073] The health management module is used to develop personalized health management plans for users based on their diagnostic storage records.

[0074] In this embodiment, the sound collection module uses voice activation detection (VAD) technology to determine the validity of the sound, reduce invalid data collection, and ensure that only voice segments related to diagnosis are collected.

[0075] In this embodiment, the entire system runs on an embedded hardware platform, which has the characteristics of low power consumption, miniaturization, high reliability, etc. The hardware platform can adopt a high-performance system using an embedded microcontroller such as an ARM Cortex-M series processor, and be equipped with sufficient storage units for storing system software, diagnostic models and user data. At the same time, it can also include communication interfaces such as WiFi, Bluetooth, etc., so as to interact with external devices (such as hospital information systems, doctors' mobile terminals, health guardians, etc.) for data.

[0076] The beneficial effects of the above technical solution are as follows: the present invention uses a sound collection module to collect the user's condition description voice based on a high-sensitivity microphone, and pre-processes the voice to obtain interference-free voice; then uses a feature extraction module to analyze and extract features of the interference-free voice through a digital signal processing algorithm to obtain the user's voice features and description of symptoms; then uses a traditional Chinese medicine diagnosis module based on the user's voice features and description of symptoms, through a pre-trained diagnosis model, combined with the correspondence between sound features and traditional Chinese medicine symptoms pre-stored in a large-scale traditional Chinese medicine diagnosis database, to obtain a traditional Chinese medicine diagnosis result; through an embedded sound analysis module, the efficiency of consultation in the traditional Chinese medicine diagnosis process is greatly improved, and the diagnosis time is reduced; and through a result feedback module, based on the traditional Chinese medicine diagnosis result, combined with the traditional Chinese medicine dialectical relationship of each symptom description, the traditional Chinese medicine diagnosis result is obtained. The system proposes preliminary TCM syndrome differentiation suggestions to the user, and outputs the corresponding diagnosis report for display, which is conducive to the user to timely understand the development of his own disease and the treatment of the disease, and realizes the automatic tracking of the user's physical health. Then, the user's voice data and diagnosis report are stored and transmitted through the data storage and transmission module, so as to realize long-term storage and tracking of the user's health data, and can better observe the changing trend of the user's health status, and provide strong support for the doctor to adjust the treatment plan and the user's self-care; finally, the health management module formulates a personalized health management plan for the user according to the user's diagnosis storage record, and makes personalized plans for the patient's treatment, diet and exercise according to the user's own situation, which is conducive to the user to better control the disease and improve the pertinence and effect of TCM treatment suggestions. The present invention combines TCM theory with modern AI technology to realize the automated sound diagnosis function, improve the standardization and consistency of the TCM "smell" diagnosis stage, and while assisting doctors in rapid syndrome differentiation and user health tracking, it also breaks the space limitation of TCM diagnosis, and provides a basis for remote health monitoring and TCM diagnosis services for users in remote areas.

[0077] Embodiment 2:

[0078] Based on Example 1, the feature extraction module, such as Figure 2 As shown, including:

[0079] A speech recognition and analysis unit is used to recognize non-interference speech based on a preset speech recognition model, obtain the user's description semantics, analyze the user's description semantics based on the syndrome differentiation and treatment vocabulary library corresponding to the TCM diagnosis, and determine the description symptoms of the user's condition;

[0080] The sound feature extraction unit is used to extract sound signs of non-interference speech based on a preset sound feature removal model, and generate a user sound map according to the sound feature extraction result.

[0081] The beneficial effects of the above technical solution are as follows: the present invention recognizes interference-free speech based on a preset speech recognition model through a speech recognition and analysis unit to obtain the user's description semantics, analyzes the user's description semantics based on the syndrome differentiation and treatment vocabulary library corresponding to the traditional Chinese medicine diagnosis, determines the descriptive symptoms of the user's condition, and realizes automatic recognition of the user's speech content; and extracts sound signs from the interference-free speech based on a preset sound feature extraction model through a sound feature extraction unit, and generates a user sound map based on the sound feature extraction result, providing a basis for the speech analysis of the traditional Chinese medicine diagnosis module.

[0082] Embodiment 3:

[0083] Based on Example 1, a TCM diagnosis module, such as Figure 3 As shown, including:

[0084] The model diagnosis unit is used to analyze the user's voice spectrum and voice features based on the pre-trained diagnosis model, and to screen the voice features pre-stored in the large database of traditional Chinese medicine diagnosis to obtain similar voice features;

[0085] A result determination unit, used to obtain a preliminary TCM diagnosis result of the user based on the correspondence between the sound features and TCM symptoms pre-stored in the large TCM diagnosis database;

[0086] Based on the preliminary TCM diagnosis results, combined with the user's described symptoms and the TCM dialectical relationship between the various described symptoms, the TCM diagnosis result of the current user is determined.

[0087] The beneficial effects of the above technical solution: the present invention analyzes the user's voice spectrum and voice features based on the pre-trained diagnosis model through the model diagnosis unit, screens the sound features pre-stored in the large-scale TCM diagnosis database, obtains similar sound features, and then obtains the user's preliminary TCM diagnosis results based on the correspondence between the sound features pre-stored in the large-scale TCM diagnosis database and the TCM symptoms through the result determination unit; based on the preliminary TCM diagnosis results, combined with the user's described symptoms and the TCM dialectical relationship between each described symptom, the TCM diagnosis result of the current user is determined, assisting doctors to quickly complete the diagnosis of the "smelling" diagnosis stage in the TCM diagnosis process, while realizing the unified standardization of the TCM "smelling" diagnosis stage, improving the efficiency of TCM consultation, and also providing a basis for remote TCM consultation.

[0088] Embodiment 4:

[0089] Based on Example 1, the result feedback module includes:

[0090] A result analysis unit, used to determine the severity of the user's current illness based on the TCM diagnosis results and the user's description of the symptoms;

[0091] A feedback reminder unit, used to send a hospital diagnosis and treatment reminder to the user when the current severity of the disease is greater than a preset degree;

[0092] A feedback display unit is used to determine the user's current physical condition and potential health problems based on the TCM diagnosis and treatment results and the current severity of the disease, combined with the development trend of the user's description of the symptoms, when the current severity of the disease is less than or equal to a preset degree;

[0093] And based on the current physical condition and potential health problems, combined with clinical diagnosis and treatment records in a large database of traditional Chinese medicine diagnosis, generate traditional Chinese medicine treatment recommendations for the user;

[0094] Generates and displays a diagnostic report based on current physical condition, potential health issues, and TCM treatment recommendations.

[0095] The beneficial effects of the above technical solution are as follows: the present invention first determines the severity of the user's current illness based on the TCM diagnosis results and the user's description of the symptoms through the result analysis unit, and then sends a hospital diagnosis and treatment reminder to the user through the feedback reminder unit when the current severity of the illness is greater than the preset degree, reminding the user to go to the hospital for follow-up diagnosis and treatment in time, avoiding delays in the user's condition and increasing the probability of treatment, reducing the high costs incurred by patients due to emergency medical treatment and hospitalization; when the current severity of the illness is less than or equal to the preset degree, based on the TCM diagnosis and treatment results and the current severity of the illness, combined with the development trend of the user's description of the symptoms, the user's current physical state and potential health problems are determined; and based on the current physical state and potential health problems, combined with the clinical diagnosis and treatment records in the large-scale TCM diagnosis database, TCM treatment suggestions are generated for the user; a diagnostic report is generated based on the current physical state, potential health problems and TCM treatment suggestions for display, providing a reference basis for the patient's TCM treatment, which is conducive to the timely control of the patient's condition and delaying the progression of the disease.

[0096] Embodiment 5:

[0097] On the basis of Example 4, the result analysis unit comprises:

[0098] The symptom classification subunit is used to determine the symptom attributes of the current user based on the TCM diagnosis results, and based on the symptom attributes, combined with the pre-stored symptom data in the TCM diagnosis large database, determine the possible accompanying symptoms corresponding to the TCM diagnosis results and the severity of the possible accompanying symptoms, and classify the accompanying symptoms according to the corresponding relationship between the accompanying symptoms and the TCM diagnosis elements to obtain sub-accompanying symptoms of multiple dimensions;

[0099] The result evaluation subunit is used to compare the similarity between the user's voice features and the matching similar features, the TCM symptoms corresponding to each matching similar feature, and the user's description features with each sub-companion symptom, and determine the symptom evaluation results of each dimension and their corresponding symptom manifestation scores according to the comparison results;

[0100] According to the preset weights corresponding to each dimension and the symptom performance scores corresponding to each dimension, the user's current symptom score result is calculated, and based on the score range corresponding to each symptom in the symptom severity assessment table and the current symptom score result, the user's current symptom severity is determined;

[0101] Among them, multiple dimensions include the nature of the pathogenic factors, the degree of damage to the internal organs, the strength of the positive energy, the complexity of the symptoms and the impact on daily life.

[0102] In this embodiment, pathogenic factors are various pathogenic factors that lead to the occurrence of diseases, including exogenous pathogenic factors and endogenous pathogenic factors. Pathogenic factors of different natures will cause different disease manifestations.

[0103] The human body is an organic whole, and the various organs are interconnected and influence each other. During the course of disease, pathogenic factors will damage the function and structure of the organs, and the degree of damage will affect the prognosis and treatment of the disease.

[0104] Zhengqi is the body's functional activity and ability to resist disease and recover. The strength of Zhengqi plays a key role in the occurrence, development and prognosis of disease.

[0105] The symptoms of a disease can be single or multiple symptoms can appear simultaneously. The complexity of the symptoms is related to many factors such as the nature of the pathogenic factors and the damage to the internal organs.

[0106] The disease will have varying degrees of impact on the patient's daily life, including physical activities, work, study, social interaction, etc.

[0107] The sum of the preset weights of the nature of the pathogenic factors, the degree of damage to the internal organs, the strength of the vital energy, the complexity of the symptoms and the impact on daily life is 1, which can be adjusted according to the doctor's diagnostic needs. For example, the nature of the pathogenic factors is 35%, the degree of damage to the internal organs is 35%, the strength of the vital energy is 15%, the complexity of the symptoms is 10% and the impact on daily life is 5%.

[0108] In this embodiment, the severity of the disease can be divided into three levels: mild, moderate and severe. Mild diseases are characterized by low symptom intensity, short duration, little impact on daily life, and no progression or a trend of improvement; moderate diseases are characterized by medium symptom intensity, lasting for a certain period of time, having a certain impact on daily life, and the condition is stable or has a slow progression; severe diseases are characterized by high symptom intensity, long duration, serious impact on daily life, and a clear trend of worsening.

[0109] Among them, the score range corresponding to each symptom in the disease severity assessment table is [70, 100] for severe, [30, 70) for moderate, and [0, 30) for mild.

[0110] The beneficial effects of the above technical solution: the present invention realizes intelligent classification of the symptoms of the user's current disease manifestation through the symptom classification sub-unit to obtain sub-accompanying symptoms of multiple dimensions, and then evaluates the severity of the sub-accompanying symptoms of multiple dimensions through the result evaluation sub-unit, and then calculates the user's current disease score result according to the preset weight corresponding to each dimension and the symptom manifestation score corresponding to each dimension, which provides a basis for evaluating the severity of the user's disease. Finally, based on the score range corresponding to each symptom in the disease severity evaluation table and combined with the current disease score result, the severity of the user's current disease is determined, thereby realizing automatic and objective evaluation of the user's disease.

[0111] Embodiment 6:

[0112] Based on Embodiment 4, the feedback display unit includes:

[0113] The diagnosis analysis subunit is used to infer the user's current physical state based on the user's diagnosis result and the current severity of the disease, combined with the user's initial physical state assessment result, and obtain the user's described symptoms and the performance association between the described symptoms;

[0114] According to the manifestation correlation between the described symptoms, the development trend of the user's described symptoms is predicted, and based on the development trend and the disease development process corresponding to the diagnosis result, new manifestations of the user's disease are predicted. Based on the TCM diagnosis and treatment logic, the user's potential health risk analysis is conducted in combination with the described symptoms and new manifestations to obtain the user's potential health problems;

[0115] It is suggested to generate a subunit for screening clinical diagnosis and treatment records in a large database of traditional Chinese medicine diagnosis based on the diagnosis results of traditional Chinese medicine, obtain relevant clinical diagnosis and treatment records, and obtain user body data corresponding to each personality clinical diagnosis and treatment record respectively;

[0116] Based on the user's physical data, determine the physical condition of the treated user corresponding to each relevant clinical diagnosis and treatment record, compare the physical condition of the treated user with the current physical condition of the current user, and determine users with the same physical condition as the current user;

[0117] Based on the target diagnosis and treatment records corresponding to the users with the same constitution, determine the potential similarity between the potential health problems of the users with the same constitution and the potential health problems of the current user, sort the availability of each target diagnosis and treatment record based on the similarity of the potential problems, and obtain a target sequence;

[0118] Obtain the TCM treatment suggestion error between each target diagnosis and treatment record in the target sequence and its corresponding adjacent target diagnosis and treatment record. When the TCM treatment suggestion error of the target sequence is always kept within a preset range, the TCM treatment suggestion of the user with the same constitution with the highest potential problem similarity in the target sequence is used as the TCM treatment suggestion for the current user;

[0119] Otherwise, the target sequence is divided into multiple subsequence sets based on a preset similarity interval, and the TCM treatment recommendations corresponding to the multiple subsequence sets are compared to obtain the differences in treatment recommendations between the subsequence sets. Based on the differences in treatment recommendations, the impact of potential health problem differences on the treatment recommendations is determined. According to the impact, the TCM treatment recommendations for the user with the same constitution who has the highest similarity to the potential problem in the target sequence are adjusted to obtain the TCM treatment recommendations for the current user.

[0120] The beneficial effects of the above technical solution: The present invention realizes the automatic formulation of TCM treatment suggestions for the current user, provides a basis for the doctor's formulation of treatment plans, and also provides a basis for the patient's condition control in the process of the doctor issuing a formal treatment plan, effectively reducing the probability of disease worsening.

[0121] Embodiment 7:

[0122] Based on Embodiment 6, the feedback display unit further includes:

[0123] The data management subunit is used for users to upload the latest physical examination data in various forms, and to evaluate the user's physical condition based on the latest physical examination data and the latest TCM diagnosis results stored in the cloud to obtain basic physical condition evaluation results;

[0124] Among them, the most recent TCM diagnosis result refers to the result of the most recent diagnosis from the current TCM diagnosis.

[0125] The latest physical examination data in various forms can be text data or physical examination report pictures.

[0126] The beneficial effects of the above technical solution: the present invention allows users to upload the latest physical examination data in various forms through the data management subunit, and evaluates the user's physical condition based on the latest physical examination data in combination with the most recent TCM diagnosis results stored in the cloud, to obtain basic physical condition evaluation results, which provides a basis for the analysis of the user's current physical condition, makes the evaluation of the current physical condition closer to the user's actual physical condition, and provides a basis for the user's TCM treatment recommendations and accurate formulation of health management plans.

[0127] Embodiment 8:

[0128] Based on Example 1, the data storage and transmission module includes:

[0129] A data transmission unit, used to send the user's voice data and diagnosis report to the user's family physician or health guardian;

[0130] The data storage unit is used to send the user's voice data and diagnostic report to the cloud for storage.

[0131] The beneficial effects of the above technical solution: The present invention sends the user's voice data and diagnosis report to the user's family physician or health guardian through a data transmission unit, which provides a basis for the implementation of remote Chinese medicine diagnosis, is beneficial for the patient's responsible physician and the patient's family to track and understand the patient's physical condition, and also provides a reference for the responsible physician to adjust the patient's treatment plan.

[0132] Embodiment 9:

[0133] Based on Example 1, the health management module includes:

[0134] The record identification management unit is used to obtain all the uncured medical records stored by the user in the cloud, and determine the user's current symptoms and current disease type based on the final TCM diagnosis record;

[0135] And based on the user's current symptoms and current disease type, all unrecovered medical records are identified to obtain recovered medical records and unrecovered medical records, and the labels of the recovered unrecovered medical records are modified to recovered labels;

[0136] The health management unit is used to determine the duration of the current symptoms based on the unrecovered diagnosis and treatment records, classify various symptoms based on the duration of each symptom, and obtain the user's long-term persistent symptom group and short-term manifestation symptom group;

[0137] Based on the unrecovered diagnosis and treatment records and TCM treatment theories, we determined the targeted treatment suggestions corresponding to various symptoms, clustered and integrated the targeted treatment suggestions corresponding to the long-term persistent symptom group and the short-term manifestation symptom group, and obtained the continuous treatment suggestions and short-term treatment suggestions;

[0138] According to the dialectical relationship between various symptoms in TCM treatment, the continuous treatment suggestions and short-term treatment suggestions are integrated within the short-term symptom prediction treatment time to obtain a fusion plan. Based on the user's current physical condition and the principle of balanced nutrition, combined with the dietary taboos recommended by TCM treatment in the fusion plan, a recommended diet strategy for the user is generated;

[0139] At the same time, according to the user's current physical condition and the recommended amount of exercise for a normal human body, the user's appropriate exercise type is determined and a recommended exercise strategy is generated for the user;

[0140] Based on the fusion scheme, the user's recommended diet strategy and the user's recommended exercise strategy, a personalized health management plan for the user is generated and sent to the embedded wearable device display page for display.

[0141] In this embodiment, the long-term persistent symptom group refers to a group consisting of long-term persistent symptoms caused by chronic diseases or other diseases; the short-term symptom group refers to a group consisting of symptoms caused by short-term diseases (for example, nasal congestion and cough caused by colds, etc.).

[0142] The beneficial effects of the above technical solution: the record identification management unit of the present invention realizes the automatic classification management of traditional Chinese medicine diagnosis records in the cloud, and also completes the preliminary screening of data used in the formulation of health management plans by the health management unit, which is conducive to improving the accuracy of the formulation of health management plans. The health management unit classifies the patient's symptoms according to the duration of the patient's unrecovered symptoms. In combination with the theory of traditional Chinese medicine treatment, targeted treatment suggestions corresponding to various symptoms are determined, and continuous treatment suggestions and short-term treatment suggestions are obtained, which provide a basis for adjusting the treatment suggestions of users at different treatment stages, which is conducive to improving the determination of traditional Chinese medicine treatment suggestions, and within the short-term symptom prediction treatment time, the continuous treatment suggestions and the short-term treatment suggestions are integrated to obtain a fusion plan. Based on the user's current physical condition and the principle of balanced nutrition, combined with the dietary taboos of the traditional Chinese medicine treatment suggestions in the fusion plan, a user recommended diet strategy is generated; at the same time, according to the user's current physical condition, with reference to the recommended exercise amount for a normal human body, the user's suitable exercise type is determined, and a user recommended exercise strategy is generated; according to the fusion plan, the user's recommended diet strategy and the user's recommended exercise strategy, a personalized health management plan is generated for the user and sent to the embedded wearable device display page for display, realizing the formulation of a health plan during the short-term symptom treatment period, ensuring that the continuous symptom treatment is taken into account during the short-term symptom treatment process, avoiding the possibility of ignoring the patient's chronic diseases during the short-term symptom treatment process, and helping to improve the diagnostic security of the system.

[0143] Embodiment 10:

[0144] Based on Example 9, the health management module further includes:

[0145] The automatic review unit is used to remind the user to undergo a TCM review when the short-term symptom predicted treatment time is reached, and re-evaluate the user's physical condition based on the user's current symptoms, and adjust the user's recommended exercise strategy and user's recommended diet strategy based on the evaluation results and continuous treatment recommendations to obtain the latest recommended exercise strategy and recommended diet strategy;

[0146] The user's health management plan is updated and displayed based on ongoing treatment recommendations, the latest recommended exercise strategies, and recommended dietary strategies.

[0147] The beneficial effects of the above technical solution are as follows: when the short-term symptom predicted treatment time is reached, the present invention reminds the user to undergo a TCM review, and re-evaluates the user's physical condition based on the user's current symptoms, and adjusts the user's recommended exercise strategy and user's recommended diet strategy according to the evaluation results and continuous treatment recommendations to obtain the latest recommended exercise strategy and recommended diet strategy; based on the continuous treatment recommendations, the latest recommended exercise strategy and recommended diet strategy, the user's health management plan is updated and displayed, which improves the adaptability of the health management plan, helps the user to better control the disease, and improves the pertinence and effectiveness of TCM treatment recommendations.

[0148] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An embedded health management system based on speech analysis for traditional Chinese medicine diagnosis, characterized in that: include: A sound collection module, used to collect the user's condition description voice based on a high-sensitivity microphone, and pre-process the voice to obtain interference-free voice; The feature extraction module is used to analyze and extract features of non-interference speech through digital signal processing algorithms to obtain the user's voice features and describe symptoms; The sound features include, but are not limited to, the fundamental frequency, harmonic structure, intonation, speech speed, loudness and timbre of the voice; The TCM diagnosis module is used to obtain TCM diagnosis results based on the user's voice features and described symptoms through a pre-trained diagnosis model combined with the correspondence between voice features and TCM symptoms pre-stored in a large TCM diagnosis database; The result feedback module is used to provide users with preliminary TCM syndrome differentiation suggestions based on the TCM diagnosis results and the TCM syndrome differentiation relationships of various symptoms, and output the corresponding diagnosis report for display; Data storage and transmission module, used to store and transmit user's voice data and diagnostic reports; The health management module is used to develop personalized health management plans for users based on their diagnostic storage records.

2. According to claim 1, an embedded health management system based on speech analysis for traditional Chinese medicine diagnosis is characterized in that: Feature extraction module, including: A speech recognition and analysis unit is used to recognize non-interference speech based on a preset speech recognition model, obtain the user's description semantics, analyze the user's description semantics based on the syndrome differentiation and treatment vocabulary library corresponding to the TCM diagnosis, and determine the description symptoms of the user's condition; The sound feature extraction unit is used to extract sound signs of non-interference speech based on a preset sound feature removal model, and generate a user sound map according to the sound feature extraction result.

3. The embedded health management system based on speech analysis for TCM diagnosis according to claim 1 is characterized in that: Traditional Chinese Medicine diagnosis module, including: The model diagnosis unit is used to analyze the user's voice spectrum and voice features based on the pre-trained diagnosis model, and to screen the voice features pre-stored in the large database of traditional Chinese medicine diagnosis to obtain similar voice features; A result determination unit, used to obtain a preliminary TCM diagnosis result of the user based on the correspondence between the sound features and TCM symptoms pre-stored in the large TCM diagnosis database; Based on the preliminary TCM diagnosis results, combined with the user's described symptoms and the TCM dialectical relationship between the various described symptoms, the TCM diagnosis result of the current user is determined.

4. The embedded health management system based on speech analysis for TCM diagnosis according to claim 1 is characterized in that: Result feedback module, including: A result analysis unit, used to determine the severity of the user's current illness based on the TCM diagnosis results and the user's description of the symptoms; A feedback reminder unit, used to send a hospital diagnosis and treatment reminder to the user when the current severity of the disease is greater than a preset degree; A feedback display unit is used to determine the user's current physical condition and potential health problems based on the TCM diagnosis and treatment results and the current severity of the disease, combined with the development trend of the user's description of the symptoms, when the current severity of the disease is less than or equal to a preset degree; And based on the current physical condition and potential health problems, combined with clinical diagnosis and treatment records in a large database of traditional Chinese medicine diagnosis, generate traditional Chinese medicine treatment recommendations for the user; Generates and displays a diagnostic report based on current physical condition, potential health issues, and TCM treatment recommendations.

5. The embedded health management system based on speech analysis for TCM diagnosis according to claim 4 is characterized in that: Result analysis unit, including: The symptom classification subunit is used to determine the symptom attributes of the current user based on the TCM diagnosis results, and based on the symptom attributes, combined with the pre-stored symptom data in the TCM diagnosis large database, determine the possible accompanying symptoms corresponding to the TCM diagnosis results and the severity of the possible accompanying symptoms, and classify the accompanying symptoms according to the corresponding relationship between the accompanying symptoms and the TCM diagnosis elements to obtain sub-accompanying symptoms in multiple dimensions; The result evaluation subunit is used to compare the similarity between the user's voice features and the matching similar features, the TCM symptoms corresponding to each matching similar feature, and the user's description features with each sub-companion symptom, and determine the symptom evaluation results of each dimension and their corresponding symptom manifestation scores according to the comparison results; According to the preset weights corresponding to each dimension and the symptom performance scores corresponding to each dimension, the user's current symptom score result is calculated, and based on the score range corresponding to each symptom in the symptom severity assessment table and the current symptom score result, the user's current symptom severity is determined; Among them, multiple dimensions include the nature of the pathogenic factors, the degree of damage to the internal organs, the strength of the positive energy, the complexity of the symptoms and the impact on daily life.

6. The embedded health management system based on speech analysis for TCM diagnosis according to claim 4 is characterized in that: Feedback display unit, comprising: The diagnosis analysis subunit is used to infer the user's current physical state based on the user's diagnosis result and the current severity of the disease, combined with the user's initial physical state assessment result, and obtain the user's described symptoms and the performance association between the described symptoms; According to the manifestation correlation between the described symptoms, the development trend of the user's described symptoms is predicted, and based on the development trend and the disease development process corresponding to the diagnosis result, new manifestations of the user's disease are predicted. Based on the TCM diagnosis and treatment logic, the user's potential health risk analysis is conducted in combination with the described symptoms and new manifestations to obtain the user's potential health problems; It is suggested to generate a subunit for screening clinical diagnosis and treatment records in a large database of traditional Chinese medicine diagnosis based on the diagnosis results of traditional Chinese medicine, obtain relevant clinical diagnosis and treatment records, and obtain user body data corresponding to each personality clinical diagnosis and treatment record respectively; Based on the user's physical data, determine the physical condition of the treated user corresponding to each relevant clinical diagnosis and treatment record, compare the physical condition of the treated user with the current physical condition of the current user, and determine users with the same physical condition as the current user; Based on the target diagnosis and treatment records corresponding to the users with the same constitution, determine the potential similarity between the potential health problems of the users with the same constitution and the potential health problems of the current user, sort the availability of each target diagnosis and treatment record based on the similarity of the potential problems, and obtain a target sequence; Obtain the TCM treatment suggestion error between each target diagnosis and treatment record in the target sequence and its corresponding adjacent target diagnosis and treatment record. When the TCM treatment suggestion error of the target sequence is always kept within a preset range, the TCM treatment suggestion of the user with the same constitution with the highest potential problem similarity in the target sequence is used as the TCM treatment suggestion for the current user; Otherwise, the target sequence is divided into multiple subsequence sets based on a preset similarity interval, and the TCM treatment recommendations corresponding to the multiple subsequence sets are compared to obtain the differences in treatment recommendations between the subsequence sets. Based on the differences in treatment recommendations, the impact of potential health problem differences on the treatment recommendations is determined. According to the impact, the TCM treatment recommendations for the user with the same constitution who has the highest similarity to the potential problem in the target sequence are adjusted to obtain the TCM treatment recommendations for the current user.

7. The embedded health management system based on speech analysis for TCM diagnosis according to claim 6 is characterized in that: The feedback display unit also includes: The data management subunit is used for users to upload the latest physical examination data in various forms, and to evaluate the user's physical condition based on the latest physical examination data and the latest TCM diagnosis results stored in the cloud to obtain basic physical condition evaluation results; Among them, the most recent TCM diagnosis result refers to the result of the most recent diagnosis from the current TCM diagnosis.

8. The embedded health management system based on speech analysis for TCM diagnosis according to claim 1, characterized in that: Data storage and transmission module, including: A data transmission unit, used to send the user's voice data and diagnosis report to the user's family physician or health guardian; The data storage unit is used to send the user's voice data and diagnostic report to the cloud for storage.

9. The embedded health management system based on speech analysis for TCM diagnosis according to claim 1, characterized in that: Health management module, including: The record identification management unit is used to obtain all the uncured medical records stored by the user in the cloud, and determine the user's current symptoms and current disease type based on the final TCM diagnosis record; And based on the user's current symptoms and current disease type, all unrecovered medical records are identified to obtain recovered medical records and unrecovered medical records, and the labels of the recovered unrecovered medical records are modified to recovered labels; The health management unit is used to determine the duration of the current symptoms based on the unrecovered diagnosis and treatment records, classify various symptoms based on the duration of each symptom, and obtain the user's long-term persistent symptom group and short-term manifestation symptom group; Based on the unrecovered diagnosis and treatment records and TCM treatment theories, we determined the targeted treatment suggestions corresponding to various symptoms, clustered and integrated the targeted treatment suggestions corresponding to the long-term persistent symptom group and the short-term manifestation symptom group, and obtained the continuous treatment suggestions and short-term treatment suggestions; According to the dialectical relationship between various symptoms in TCM treatment, the continuous treatment suggestions and short-term treatment suggestions are integrated within the short-term symptom prediction treatment time to obtain a fusion plan. Based on the user's current physical condition and the principle of balanced nutrition, combined with the dietary taboos recommended by TCM treatment in the fusion plan, a recommended diet strategy for the user is generated; At the same time, according to the user's current physical condition and the recommended amount of exercise for a normal human body, the user's appropriate exercise type is determined and a recommended exercise strategy is generated for the user; Based on the fusion scheme, the user's recommended diet strategy and the user's recommended exercise strategy, a personalized health management plan for the user is generated and sent to the embedded wearable device display page for display.

10. The embedded health management system based on speech analysis for TCM diagnosis according to claim 9, characterized in that: The health management module also includes: Automatic review unit, used to remind users to have a TCM review when the predicted treatment time of short-term symptoms is reached, to determine whether each symptom in the short-term symptom group has been alleviated; If so, the user's physical condition is re-evaluated based on the user's current symptoms, and the user's recommended exercise strategy and the user's recommended diet strategy are adjusted according to the evaluation results and continuous treatment recommendations to obtain the latest recommended exercise strategy and recommended diet strategy; The user's health management plan is updated and displayed based on ongoing treatment recommendations, the latest recommended exercise strategies, and recommended dietary strategies.

Images