System and method for recognizing early Alzheimer's disease through mobile phone voice

Abstract

The invention relates to the technical field of medical software, in particular to a system and method capable of recognizing early Alzheimer's disease through mobile phone voice, the system capable of recognizing early Alzheimer's disease through mobile phone voice comprises intelligent mobile phone software and a hospital background, and the mobile phone software builds an interactive platform based on the hospital background. Through intervention monitoring of mobile phone software, when voice communication, chatting and instruction input are carried out in daily life, monitoring can be carried out through system monitoring of software and characteristics such as smoothness and continuity of voice, so that analysis and diagnosis of the Alzheimer's disease can be carried out according to comparison with a normal state, and moreover, in the monitoring process, the monitoring efficiency is improved. Diversified collection is carried out in different time periods and different statement numbers, language recognition and dialect recognition are added when a user inputs voice, and keyword recognition for determining the Alzheimer's disease condition can be carried out through approximate comparison with mandarin after the user speaks out statements.

Description

Technical Field

[0001] The present invention relates to the field of medical software technology, and in particular to a system and method for identifying early-stage Alzheimer's disease through mobile phone voice. Background Art

[0002] Alzheimer's disease is a progressive neurodegenerative disorder that primarily affects the elderly. The most common early symptom is short-term memory loss. Patients may repeat themselves, forget appointments or daily tasks, and experience a decline in thinking, judgment, and problem-solving abilities. Patients may become lost in familiar environments and have difficulty completing daily tasks. They may also have trouble finding the right words or expressing their thoughts.

[0003] After searching, based on the authorization announcement number CN116327117A, it discloses the assessment of Parkinson's disease condition based on speech recognition. Provided are systems and methods for assessing Parkinson's disease (PD) condition or its progression and for treating PD condition by using speech analysis. The assessment of PD condition or its progression can be performed immediately or over time in the following manner: by analyzing the current speech sample of the PD subject, and optionally analyzing the historical speech samples stored in the data storage unit, determining the relevant speech features of the PD subject based on the analysis of the speech sample, and assessing the PD condition or its progression by comparing the current speech features with similar speech features. Treating PD condition can include, for example, adjusting drug delivery parameters to a therapeutically beneficial level, for example, beneficial to preventing or delaying the onset of the "off" period or shortening the duration of the "off" period, or improving the symptoms of hypokinetic dysarthria (HD) or other PD symptoms.

[0004] At present, Alzheimer's disease is difficult to detect in its early stages and can be easily confused with senile forgetfulness, mental overwork, personality habits and work style. People of a certain age may also ignore the occurrence of Alzheimer's disease because they think their memory is declining due to age. In addition, routine physical examinations record less information, making it difficult to accurately diagnose Alzheimer's disease in its early stages. Continuous observation is often used as a monitoring method, allowing patients to monitor themselves. However, in the early stages of Alzheimer's disease, the cognitive memory of things is no different from that of normal people. Except for those with a family history of Alzheimer's disease who pay attention to self-monitoring, it is difficult for the rest of the patients to meet this standard.

[0005] With the popularity of mobile phones, most middle-aged and elderly people cannot leave their phones. They communicate through mobile phones for communication and business. Voice communication on smartphones is the most frequently used communication method besides text input. The early symptoms of Alzheimer's disease are closely related to factors such as situational memory, naming, orientation in time and space, vocabulary, speaking speed, and language fluency. Based on the global prevention and control consensus of Alzheimer's disease "early identification, early intervention, and early benefit", the mobile phone voice analysis system can dynamically identify community people, especially high-risk groups, which can improve the early screening and identification rate of the disease at a low cost, and help medical institutions accurately improve the early intervention rate of the disease. Summary of the Invention

[0006] In response to the defects in the existing technology, the present invention provides a system that can identify early Alzheimer's disease through mobile phone voice. The system for identifying early Alzheimer's disease through mobile phone voice includes smart phone software and a hospital backend. The mobile phone software builds an interactive platform based on the hospital backend. The hospital backend is set up in a designated hospital. The hospital backend includes a database module, a doctor information module, a doctor working hours module, a user information filing module, a user information protection module, a monitoring information timed recording and summary module, a monitoring information analysis and processing module, a problem information prompt and warning module, a suspected patient appointment module, a misdiagnosed person warning elimination module, and a confirmed person continuous monitoring module.

[0007] Furthermore, the steps for setting up the hospital backend and mobile phone software are as follows:

[0008] Step 1: System design;

[0009] Step 2: Mobile software development;

[0010] Front-end development: developing the user interface of mobile software, supporting iOS and Android platforms;

[0011] Back-end development: Develop the business logic and data processing functions of the hospital's back-end system;

[0012] Interface implementation: realize the interface communication between mobile phone software and hospital backend system;

[0013] Integrated development: Integrate front-end and back-end systems and conduct joint debugging;

[0014] Step 3: Monitoring process deployment;

[0015] Server configuration: configure server hardware and operating system, and install necessary software environment;

[0016] Database deployment: deploy the database system and import initial data;

[0017] Application deployment: deploy the backend application to the server and configure related parameters;

[0018] Monitoring and maintenance: Set up system monitoring and logging, and perform regular system maintenance and updates;

[0019] Step 4: Continuous maintenance of the software and background during use.

[0020] Furthermore, based on the user interface of the second step, when registering and logging in, the information to be filled in includes name, gender, age, designated hospital selection, ID card information, medical history information, mobile phone number and login password.

[0021] Furthermore, in the front-end development, several segment nodes of monitoring period are set, A(A1,A N )、B(B1,B n )、C(C1,C n ), to achieve full-day monitoring, segmentation from midnight to twelve o'clock, segmentation from twelve o'clock to midnight of the next day, or set three groups of node monitoring for any time period, with the node time not less than two hours.

[0022] Furthermore, based on the third step, the monitoring scope includes incoming mobile phone calls, voice calls of various chat software, voice message releases, and voice operations.

[0023] Furthermore, based on the Python design, language recognition and dialect recognition can be set to make approximate translation judgments for users of different regional languages.

[0024] Furthermore, based on the third step, the monitoring scope includes incoming mobile phone calls, voice calls using various chat software, voice message publishing, and voice operations on mobile phones.

[0025] Further, based on the categories of language monitoring:

[0026] L1, monitoring the repetition frequency of relevant sensitive words;

[0027] L2, monitoring of pause frequency in speech, i.e., controlling speech rate, including identifying complete sentences starting with five or more words (fewer than five words will not be recorded);

[0028] In a complete sentence, if there are more than three pauses or the interval time exceeds one second, it will be recorded as a frequency count. Every eight hours is set as the time node for frequency count calculation. Within an eight-hour interval, more than ten voice sentences are used as the starting measurement standard. When the pause frequency exceeds the set upper limit, it will be reported once to the hospital background. If it exceeds but not including three times, an early warning will be activated.

[0029] Furthermore, when the hospital backend calculates the frequency of occurrence, it first uses conventional frequency calculation and then predicts it through frequency estimation. If the frequency in the subsequent time reaches the predicted value, it is confirmed to be Alzheimer's disease.

[0030] Furthermore, the method for identifying early-stage Alzheimer's disease through mobile phone voice is as follows:

[0031] J1. Language fluency analysis;

[0032] J1.1. Vocabulary: Alzheimer's patients often experience a decrease in vocabulary. By analyzing the user's vocabulary usage in daily conversations, we can preliminarily determine whether there is a language barrier.

[0033] J1.2. Grammatical structure: Patients may make grammatical errors or have disorganized sentence structures. Speech recognition technology analyzes the user's grammatical usage to assess their language ability.

[0034] J2. Speech feature analysis;

[0035] J2.1. Speech speed: Alzheimer's patients typically speak slowly and pause frequently. By analyzing the user's speech speed and pause frequency, we can preliminarily determine whether there is a speech disorder.

[0036] J2.2. Pitch and Volume: Patients may have a monotonous voice or unstable volume. Speech recognition technology analyzes changes in the user's pitch and volume to assess their language ability.

[0037] J3, emotion recognition;

[0038] J3.1. Emotional Expression: Alzheimer's patients may have difficulty expressing their emotions. By analyzing the user's emotional changes, we can preliminarily determine whether their emotional expression ability is affected.

[0039] J3.2. Mood swings: Patients may experience significant mood swings. Voice recognition technology analyzes the user's mood changes to assess their emotional state.

[0040] J4, memory and cognitive function tests;

[0041] J4.1. Voice Memory Test: Using voice recognition technology, we design some voice memory test questions, asking users to repeat or recall specific words or sentences to assess their memory;

[0042] J4.2. Voice Command Execution: Using voice recognition technology, design some simple voice commands and have users execute them to assess their understanding and execution capabilities.

[0043] J5, daily conversation monitoring;

[0044] J5.1. Natural Conversation Analysis: By analyzing the user's performance in daily conversations, we can preliminarily determine whether their language ability and cognitive function are affected.

[0045] J5.2. Voice Diary: Encourage users to record their lives with voice every day. By monitoring their voice changes over a long period of time, potential language and cognitive dysfunctions can be discovered.

[0046] The beneficial effects of the present invention are as follows: 1. Through the intervention monitoring of mobile phone software, during daily voice calls, chats, and command input, the software system can monitor characteristics such as voice fluency and coherence, thereby performing Alzheimer's disease analysis and diagnosis based on comparison with normal conditions. In addition, during the monitoring process, diversified data collection is performed through different time periods and different numbers of sentences. When the user inputs voice, language recognition and dialect recognition are also added. After the user speaks a sentence, it can be compared with Mandarin to identify keywords related to Alzheimer's disease, thereby increasing the adaptability and extensiveness of monitoring and improving the accuracy of monitoring recognition.

[0047] 2. As described in 1, by calculating the repetition frequency of sensitive words and the pause frequency of sentences, conventional basic frequency calculation is performed using the normal distribution method, and then future frequency prediction is performed using the Bayesian frequency calculation method. In subsequent monitoring, the frequency conditions of real-time monitoring can be discretely compared with the future frequencies, thereby quickly and accurately determining whether the user's Alzheimer's disease is misdiagnosed or confirmed. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly describes the drawings required for the specific embodiments or the description of the prior art. Similar elements or parts are generally identified by similar reference numerals throughout the drawings. Elements or parts in the drawings are not necessarily drawn to scale.

[0049] Figure 1 This is a schematic diagram of the hospital backend and mobile phone software construction of the present invention;

[0050] Figure 2 This is a schematic diagram of the mobile phone software registration information of the present invention;

[0051] Figure 3 This is a schematic diagram of mobile phone voice recognition of Alzheimer's disease according to the present invention. DETAILED DESCRIPTION

[0052] The following embodiments of the technical solution of the present invention will be described in detail with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention and are therefore only examples and are not intended to limit the scope of protection of the present invention.

[0053] It should be noted that, unless otherwise specified, the technical 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.

[0054] like Figure 1-3 As shown, a system that can identify early Alzheimer's disease through mobile phone voice recognition, the system that can identify early Alzheimer's disease through mobile phone voice recognition includes smart phone software and hospital background, the mobile phone software builds an interactive platform based on the hospital background, the hospital background is set up in a designated hospital, the hospital background includes a database module, a doctor information module, a doctor working time module, a user information filing module, a user information protection module, a monitoring information timed recording and summary module, a monitoring information analysis and processing module, a problem information prompt and warning module, a suspected patient appointment module, a misdiagnosed person warning elimination module, and a confirmed person continuous monitoring module.

[0055] The steps for setting up the hospital backend and mobile software are as follows:

[0056] Step 1: System design;

[0057] Architecture design: Select appropriate system architecture, such as client-server architecture and microservice architecture;

[0058] Database design: Design database structure, including user information, appointment records, and medical records data;

[0059] Interface design: define the interface protocol between mobile software and hospital backend system, such as RESTful API and SOAP;

[0060] Security design: Consider security measures such as data encryption, identity authentication, and permission management;

[0061] Step 2: Mobile software development;

[0062] Front-end development: developing the user interface of mobile software, supporting iOS and Android platforms;

[0063] Back-end development: Develop the business logic and data processing functions of the hospital's back-end system;

[0064] Interface implementation: realize the interface communication between mobile phone software and hospital backend system;

[0065] Integrated development: Integrate front-end and back-end systems and conduct joint debugging;

[0066] Step 3: Monitoring process deployment;

[0067] Server configuration: configure server hardware and operating system, and install necessary software environment;

[0068] Database deployment: deploy the database system and import initial data;

[0069] Application deployment: deploy the backend application to the server and configure related parameters;

[0070] Monitoring and maintenance: Set up system monitoring and logging, and perform regular system maintenance and updates;

[0071] Step 4: Continuous maintenance of software and background during use;

[0072] User feedback: Collect user feedback and continuously improve system functions and user experience;

[0073] Version updates: New versions are released regularly to fix bugs and add new features.

[0074] Based on the user interface in the second step, when registering and logging in, the information to be filled in includes name, gender, age, designated hospital selection, ID card information, medical history information, mobile phone number and login password. For daily login, enter ID card number, mobile phone number and login password;

[0075] In front-end development, since the background software consumes power, in order to reduce power loss, you can also set up several segment nodes for monitoring periods, A(A1,A N )、B(B1,B n )、C(C1,C n ), can achieve full-day monitoring, segmented monitoring from midnight to 12 o'clock, segmented monitoring from 12 o'clock to 0:00 the next day, or set three groups of node monitoring for any time period, with the node time not less than two hours;

[0076] Based on Python design, language recognition and dialect recognition can be set to make approximate translation judgments for users of different regional languages;

[0077] I. Language Identification Method:

[0078] Establish a complete language word database and language ranking information, by recording the frequency of occurrence of language words and language ranking information as f(m) and r(m) respectively, where m represents the sequence number of the word in the language information to be identified, and m takes values of 1, 2, 3, ..., M, and M is the maximum value corresponding to the sequence number of the word. When m is 1, it represents the first word, when m is 2, it represents the second word, and when m takes the value of M, it represents the Mth word;

[0079] Collect language data of known languages to obtain a known language database corresponding to each language; wherein yi represents the language, i is 1, 2, ..., Y; y1 represents the first language, y2 represents the second language, and so on; Y represents the maximum value corresponding to the language number in the known language database;

[0080] Based on the known language database, count and record the mth word cm contained in the language information to be recognized in the corresponding i-th known language y i The known frequency f(y i , c m ) and known sort r(y i , c m ); where m is 1, 2, ..., M; f(y i , c m ) represents the mth word c m The corresponding known frequency in language i, r(y i , c m ) represents the mth word c m The order of frequency of occurrence in language i corresponds to the known order;

[0081] The frequency of occurrence f(m) and the frequency of occurrence f(y i , c m ), and use formula (1) to calculate the frequency feature distance F(i) between the language to be identified and language i:

[0082]

[0083] Ranking information r(m) and ranking information r(y i , c m ), using formula (2), calculate the ranking feature distance R(i) between the language information to be identified and language i:

[0084]

[0085] The frequency feature distance F(i) and the ranking feature distance R(i) are used to calculate the language probability P(i) that the language information to be identified is language i using formula (3):

[0086]

[0087] By identifying the probability of a certain language, the category of the language is determined;

[0088] II. Dialect Recognition Methods

[0089] Determine the characteristics of the sound produced by the user, including short-time energy, zero-crossing rate and vocal tract characteristics;

[0090] The information evaluation parameters are obtained through the following formula:

[0091] S=λ1·S1+λ2·S2+λ3·S3

[0092]

[0093] S represents the information evaluation parameter; S1, S2 and S3 represent the first evaluation factor, the second evaluation factor and the third evaluation factor respectively;

[0094] λ1, λ2 and λ3 represent the preset weights corresponding to the first evaluation factor, the second evaluation factor and the third evaluation factor respectively;

[0095] n represents the number of frames corresponding to a speech information division, and each frame corresponds to a time window;

[0096] E 1i Represents the energy value corresponding to the speech information of the i-th time window;

[0097] E0 represents the preset energy threshold;

[0098] N i Indicates the number of zero crossings of the speech information corresponding to the i-th time window;

[0099] N0 represents the preset zero-crossing threshold;

[0100] N max Indicates the maximum number of zero crossings corresponding to the time window contained in the voice information input by the user;

[0101] f i Represents the formant frequency of the speech information corresponding to the i-th time window;

[0102] f0 represents the preset resonance peak frequency threshold;

[0103] f max Indicates the maximum formant frequency of valid voice information input by the user;

[0104] H i Represents the formant amplitude of the speech information corresponding to the i-th time window;

[0105] H0 represents the preset resonance peak amplitude threshold;

[0106] H max Indicates the maximum formant amplitude of the valid speech information input by the user.

[0107] Based on the third step, the monitoring scope includes incoming mobile phone calls, voice calls using various chat software, voice message publishing, and voice operations on mobile phones.

[0108] The following are some ways to identify early-stage Alzheimer's disease through mobile phone voice:

[0109] J1. Language fluency analysis;

[0110] J1.1. Vocabulary: Alzheimer's patients often experience a decrease in vocabulary. By analyzing the vocabulary used by users in daily conversations, we can preliminarily determine whether there is a language barrier.

[0111] J1.2. Grammatical structure: Patients may make grammatical errors or have disorganized sentence structures. Speech recognition technology can analyze the user's grammatical usage to assess their language ability.

[0112] J2. Speech feature analysis;

[0113] J2.1. Speech speed: Alzheimer's patients typically speak slowly and pause frequently. By analyzing the user's speech speed and pause frequency, we can preliminarily determine whether there is a speech disorder.

[0114] J2.2. Pitch and Volume: Patients may have a monotonous voice or unstable volume. Speech recognition technology can analyze the user's pitch and volume changes to assess their language ability.

[0115] J3, emotion recognition;

[0116] J3.1. Emotional Expression: Alzheimer's patients may have difficulty expressing their emotions. By analyzing the user's emotional changes, we can preliminarily determine whether their emotional expression ability is affected.

[0117] J3.2 Mood swings: Patients may experience significant mood swings. Voice recognition technology can analyze the user's mood changes to assess their emotional state.

[0118] J4, memory and cognitive function tests;

[0119] J4.1. Voice Memory Test: Using voice recognition technology, we can design voice memory test questions that require users to repeat or recall specific words or sentences to assess their memory.

[0120] J4.2. Voice Command Execution: Using voice recognition technology, we can design some simple voice commands for users to execute in order to assess their understanding and execution capabilities.

[0121] J5, daily conversation monitoring;

[0122] J5.1. Natural Conversation Analysis: By analyzing the user's performance in daily conversations, we can preliminarily determine whether their language ability and cognitive function are affected.

[0123] J5.2. Voice Diary: Encourage users to record their lives with voice every day. By monitoring their voice changes over a long period of time, potential language and cognitive dysfunctions can be discovered.

[0124] Categories based on language monitoring:

[0125] L1. Monitoring the repetition frequency of relevant sensitive words, including "forget", "can't remember", "can't recall", "it seems so", etc., which are often used to express forgetfulness;

[0126] L2, monitoring of pause frequency in speech, i.e., controlling speech rate, including identifying complete sentences starting with five or more words (fewer than five words will not be recorded);

[0127] In a complete sentence, if there are more than three pauses or the interval time exceeds one second, it will be recorded as a frequency count. Every eight hours is set as the time node for frequency count calculation. Within an eight-hour interval, more than ten voice sentences are used as the starting measurement standard. When the pause frequency exceeds the set upper limit, it will be reported to the hospital backend once. If it exceeds but not includes three times, an alarm will be activated;

[0128] When calculating the frequency of occurrence, the hospital backend first uses conventional frequency calculation and then frequency estimation to make a prediction. If the frequency in the subsequent time reaches the predicted value, it is confirmed to be Alzheimer's disease;

[0129] The calculation of regular frequencies is based on the normal distribution:

[0130]

[0131] The normal distribution has two parameters, the mean μ and the standard deviation σ, which can be written as N(μ,σ²): the mean μ determines the center position of the normal curve; the standard deviation σ determines the steepness or flatness of the normal curve. The smaller σ is, the steeper the curve is; the larger σ is, the flatter the curve is.

[0132] u-transformation: For ease of description and application, normal variables are often transformed. μ is the location parameter of the normal distribution, which describes the location of the central tendency of the normal distribution. The normal distribution is completely symmetrical with X = μ as the axis of symmetry. The mean, median, and mode of the normal distribution are the same and are all equal to μ.

[0133] σ describes the degree of dispersion of the data distribution of the normal distribution. The larger σ is, the more dispersed the data distribution is, and the smaller σ is, the more concentrated the data distribution is. It is also called the shape parameter of the normal distribution. The larger σ is, the flatter the curve is, and conversely, the smaller σ is, the taller the curve is.

[0134] The predicted frequencies are based on the conventional frequency criterion, obtained using Bayes' theorem:

[0135]

[0136] (P(A|B)) is the posterior probability of event A occurring given event B;

[0137] (P(B|A)) is the likelihood of event B occurring given event A;

[0138] (P(A)) is the prior probability of event A;

[0139] (P(B)) is the total probability of event B.

[0140] 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 above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions 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 description of the present invention.

Claims

1. A system for identifying early-stage Alzheimer's disease through mobile phone voice recognition, characterized by: The system capable of recognizing early Alzheimer's disease through mobile phone voice comprises a smartphone software and a hospital backend, wherein the mobile phone software builds an interactive platform based on the hospital backend, and the hospital backend is set up in a designated hospital, and the hospital backend comprises a database module, a doctor information module, a doctor working hours module, a user information filing module, a user information protection module, a monitoring information timing recording and summary module, a monitoring information analysis and processing module, a problem information prompt and warning module, a suspected patient appointment appointment module, a misdiagnosed person warning elimination module, and a confirmed person continuous monitoring module.

2. The system for identifying early-stage Alzheimer's disease through mobile phone voice according to claim 1, characterized in that: The steps for setting up the hospital backend and mobile phone software are as follows: Step 1: System design; Step 2: Mobile software development; Front-end development: developing the user interface of mobile software, supporting iOS and Android platforms; Back-end development: Develop the business logic and data processing functions of the hospital's back-end system; Interface implementation: realize the interface communication between mobile phone software and hospital backend system; Integrated development: Integrate front-end and back-end systems and conduct joint debugging; Step 3: Monitoring process deployment; Server configuration: configure server hardware and operating system, and install necessary software environment; Database deployment: deploy the database system and import initial data; Application deployment: deploy the backend application to the server and configure related parameters; Monitoring and maintenance: Set up system monitoring and logging, and perform regular system maintenance and updates; Step 4: Continuous maintenance of the software and background during use.

3. The system for identifying early-stage Alzheimer's disease through mobile phone voice according to claim 2, characterized in that: Based on the user interface in the second step, when registering and logging in, the information to be filled in includes name, gender, age, designated hospital selection, ID card information, medical history information, mobile phone number and login password.

4. The system for identifying early-stage Alzheimer's disease through mobile phone voice according to claim 3, characterized in that: In front-end development, set up several segment nodes for monitoring periods, A(A1,A N )、B(B1,B n )、C(C1,C n ), to achieve full-day monitoring, segmentation from midnight to twelve o'clock, segmentation from twelve o'clock to midnight of the next day, or set three groups of node monitoring for any time period, with the node time not less than two hours.

5. The system for identifying early-stage Alzheimer's disease through mobile phone voice according to claim 4, characterized in that: Based on the third step, the monitoring scope includes incoming mobile phone calls, voice calls of various chat software, voice message releases, and voice operations.

6. The system for identifying early-stage Alzheimer's disease through mobile phone voice according to claim 5, characterized in that: Designed based on Python, it can set language recognition and dialect recognition, and make approximate translation judgments for users of different regional languages.

7. The system for identifying early-stage Alzheimer's disease through mobile phone voice according to claim 6, characterized in that: Based on the third step, the monitoring scope includes incoming mobile phone calls, voice calls using various chat software, voice message publishing, and voice operations on mobile phones.

8. The system for identifying early-stage Alzheimer's disease through mobile phone voice according to claim 7, characterized in that: Categories based on language monitoring: L1, monitoring the repetition frequency of relevant sensitive words; L2, monitoring of pause frequency in speech, i.e., controlling speech rate, including identifying complete sentences starting with five or more words (fewer than five words will not be recorded); In a complete sentence, if there are more than three pauses or the interval time exceeds one second, it will be recorded as a frequency count. Every eight hours is set as the time node for frequency count calculation. Within an eight-hour interval, more than ten voice sentences are used as the starting measurement standard. When the pause frequency exceeds the set upper limit, it will be reported once to the hospital background. If it exceeds but not including three times, an early warning will be activated.

9. The system for identifying early-stage Alzheimer's disease through mobile phone voice according to claim 8, characterized in that: When the hospital backend calculates the frequency of occurrence, it first uses conventional frequency calculation and then makes a prediction through frequency estimation. If the frequency in the subsequent time reaches the predicted value, it is confirmed to be Alzheimer's disease.

10. The method for identifying early-stage Alzheimer's disease through mobile phone voice according to any one of claims 1 to 9, characterized in that: The method for identifying early-stage Alzheimer's disease through mobile phone voice is as follows: J1. Language fluency analysis; J1.

1. Vocabulary: Alzheimer's patients often experience a decrease in vocabulary. By analyzing the user's vocabulary usage in daily conversations, we can preliminarily determine whether there is a language barrier. J1.

2. Grammatical structure: Patients may make grammatical errors or have disorganized sentence structures. Speech recognition technology analyzes the user's grammatical usage to assess their language ability. J2. Speech feature analysis; J2.

1. Speech speed: Alzheimer's patients typically speak slowly and pause frequently. By analyzing the user's speech speed and pause frequency, we can preliminarily determine whether there is a speech disorder. J2.

2. Pitch and Volume: Patients may experience a monotonous voice and unstable volume. Speech recognition technology analyzes changes in the user's pitch and volume to assess their language ability. J3, emotion recognition; J3.

1. Emotional Expression: Alzheimer's patients may have difficulty expressing their emotions. By analyzing the user's emotional changes, we can preliminarily determine whether their emotional expression ability is affected. J3.

2. Mood swings: Patients may experience significant mood swings. Voice recognition technology analyzes the user's mood changes to assess their emotional state. J4, memory and cognitive function tests; J4.

1. Voice Memory Test: Using voice recognition technology, we design some voice memory test questions, asking users to repeat or recall specific words or sentences to assess their memory; J4.

2. Voice Command Execution: Using voice recognition technology, design some simple voice commands and have users execute them to assess their understanding and execution capabilities. J5, daily conversation monitoring; J5.

1. Natural Conversation Analysis: By analyzing the user's performance in daily conversations, we can preliminarily determine whether their language ability and cognitive function are affected. J5.

2. Voice Diary: Encourage users to record their lives with voice every day. By monitoring their voice changes over a long period of time, potential language and cognitive dysfunctions can be discovered.

Images