Intelligent earphone-based tinnitus monitoring method and system

Abstract

The invention belongs to the technical field of electronic equipment, and particularly relates to a tinnitus monitoring method and system based on an intelligent earphone, the design of the intelligent earphone enables tinnitus monitoring to be simple and easy, and the convenience and timeliness of monitoring are greatly improved. Through a high-precision spectrum analysis chip and a self-adaptive filtering algorithm, environmental noise can be effectively inhibited, and the accuracy of tinnitus frequency and loudness data is ensured. A personalized control signal is generated by using a deep learning model, and a sound signal is customized according to the tinnitus characteristics of each patient, so that the effectiveness and pertinence of analysis are remarkably improved. In addition, the data encryption and remote transmission functions ensure the safety and privacy protection of patient data, and meanwhile, the doctor can remotely monitor and adjust a personalized analysis scheme conveniently. The problem that convenient and accurate tinnitus self-monitoring and personalized analysis are difficult to achieve in the prior art is effectively solved, and a more scientific and efficient health management means is brought to tinnitus patients.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of electronic equipment, and particularly relates to a tinnitus monitoring method and system based on a smart earphone. BACKGROUND

[0002] Tinnitus, as a common auditory phenomenon, has a significant negative impact on the daily life of many people. Existing tinnitus monitoring methods mainly rely on professional medical equipment and the operation of professional personnel, such as pure tone audiometers or otoacoustic emission testers. These devices are usually large in size, complex in operation and need to be performed in hospitals or clinics, limiting the possibility of daily self-monitoring by patients. In addition, traditional methods often only provide basic information on tinnitus frequency and loudness, lack personalized analysis schemes, and cannot be dynamically adjusted according to the specific conditions of each patient.

[0003] Specifically, one of the main problems in the prior art is that it is difficult to achieve convenient and accurate self-monitoring and personalized analysis of tinnitus. Since traditional tinnitus monitoring devices do not have portability and ease of use, patients are difficult to continuously monitor their tinnitus status at home, which not only affects the effectiveness of early detection and intervention, but also limits the feasibility of long-term management. At the same time, traditional analysis methods are mostly general strategies that do not fully consider individual differences, resulting in inaccurate analysis results. SUMMARY

[0004] The purpose of the present application is to provide a tinnitus monitoring method and system based on a smart earphone to solve the problems raised in the background.

[0005] To achieve the above-mentioned purpose, the present application adopts the following technical solution: a tinnitus monitoring method based on a smart earphone, comprising the following steps: Pressing the operation button starts the tinnitus detection module, and the ear sound signal is collected by the spectrum analysis chip. The time domain signal collected by the spectrum analysis chip is converted into a frequency domain signal by fast Fourier transform, and the tinnitus frequency range is identified; The spectrum analysis chip obtains tinnitus loudness data through a sound pressure sensor and transmits the tinnitus frequency range and loudness data to a main control module; The main control module stores the tinnitus frequency range and loudness data in the data storage module through the AES encryption standard, and sends the tinnitus frequency range and loudness data to a rehabilitation sound synthesis module; The rehabilitation sound synthesis module generates a personalized control signal based on the tinnitus frequency range and loudness data, and transmits the personalized control signal to a programmable waveform generator to generate a rehabilitation sound. The programmable waveform generator outputs the rehabilitation sound through a sound output unit.

[0006] Preferably, the time-domain signal collected by the spectrum analysis chip is converted into a frequency-domain signal by a fast Fourier transform, including: The collected ear sound signal is subjected to frame processing, and each frame contains a fixed number of sampling points; A Hamming window function is applied to each frame of signal to reduce spectral leakage; The frequency domain components of each frame of signal are calculated by a fast Fourier transform, and the frequency range with energy higher than the background noise is extracted.

[0007] Preferably, the spectrum analysis chip acquires tinnitus loudness data through a sound pressure sensor, including: The short-time energy of the ear sound signal is calculated to obtain the instantaneous sound pressure value; The instantaneous sound pressure value is smoothed by a moving average filter to eliminate transient noise interference; The smoothed sound pressure value is compared with the reference sound pressure value to calculate the relative change amplitude of the tinnitus loudness.

[0008] Preferably, the tinnitus frequency range and loudness data are stored in the data storage module by the main control module through the AES encryption standard, including: The tinnitus frequency range and loudness data are divided into fixed-size data blocks; An AES encryption algorithm is applied to each data block to generate corresponding ciphertext data; The ciphertext data is written to the designated storage area of the data storage module.

[0009] Preferably, the generation of the individualized control signal includes: The tinnitus frequency range is matched with the rehabilitation sound frequency through a preset mapping rule; The amplitude of the rehabilitation sound is adjusted according to the tinnitus loudness data to maintain dynamic balance between the rehabilitation sound intensity and the tinnitus loudness; The control signal parameters are optimized through historical rehabilitation records.

[0010] Preferably, the generation of the rehabilitation sound includes: A composite signal of white noise and reverse-phase masking sound is synthesized by a programmable waveform generator; The amplitude ratio of white noise and masking sound is adjusted according to the individualized control signal; The synthesized signal is subjected to low-pass filtering by a digital signal processing module to remove high-frequency noise components.

[0011] Preferably, the following steps are further included: The tinnitus frequency range and loudness data are transmitted by the main control module to the mobile phone end application program through the Bluetooth module; The mobile phone terminal application program shows the change trend of tinnitus frequency with time in the form of a broken line graph, and uploads data to a cloud server through a Wi-Fi module.

[0012] Preferably, the tinnitus frequency range and loudness data are transmitted by the master module to the mobile phone terminal application program through a Bluetooth module, including: The tinnitus frequency range and loudness data are encapsulated as Bluetooth data packets. The data packets are sent to the mobile phone terminal device through a Bluetooth protocol stack. The mobile phone terminal device receives the data packets, parses and stores them in a local database.

[0013] Preferably, the data is uploaded to the cloud server through the Wi-Fi module, including: The locally stored tinnitus data is sent to the cloud server through an HTTP protocol. The cloud server performs integrity checking on the received data. A long-term trend report of the patient's tinnitus condition is generated through a data visualization tool.

[0014] In another aspect, the present application proposes a tinnitus monitoring system based on a smart earphone, including: An operation button is used to start the tinnitus detection module. A spectrum analysis chip is connected to the operation button and is used to collect ear sound signals and perform spectrum analysis. A sound pressure sensor is connected to the spectrum analysis chip and is used to obtain tinnitus loudness data. A master module is connected to the spectrum analysis chip and the sound pressure sensor and is used to receive and process tinnitus data. A data storage module is connected to the master module and is used to store encrypted tinnitus data. A rehabilitation sound synthesis module is connected to the master module and is used to generate individualized control signals. A programmable waveform generator is connected to the rehabilitation sound synthesis module and is used to generate rehabilitation sounds. A sound emitting unit is connected to the programmable waveform generator and is used to output rehabilitation sounds. A Bluetooth module is connected to the master module and is used to transmit tinnitus data to a mobile phone terminal application program. A Wi-Fi module is connected to the master module and is used to upload data to a cloud server.

[0015] The technical effects and advantages of the present application are as follows: The present application provides a convenient and accurate tinnitus self-monitoring and personalized analysis solution. First, the design of the intelligent earphone makes tinnitus monitoring simple and easy, and patients do not need to go to a medical institution and can complete real-time monitoring of the tinnitus state at home, greatly improving the convenience and timeliness of monitoring. Second, through the high-precision spectrum analysis chip and the adaptive filtering algorithm, environmental noise can be effectively suppressed to ensure the accuracy of the tinnitus frequency and loudness data. Third, the deep learning model is used to generate personalized control signals, and the sound signals are customized according to the tinnitus characteristics of each patient, which significantly improves the effectiveness and pertinence of the analysis. In addition, data encryption and remote transmission function ensure the safety and privacy protection of patient data, and also facilitate remote monitoring and adjustment of the personalized analysis scheme by doctors. In summary, the present application effectively solves the problem of difficult implementation of convenient and accurate tinnitus self-monitoring and personalized analysis in the prior art, and brings a more scientific and efficient health management method for tinnitus patients. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 A flowchart of a tinnitus monitoring method based on an intelligent earphone according to the present application. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The specific embodiments described herein are only used to explain the present application, and are not used to limit the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0018] The present application provides a tinnitus monitoring method based on an intelligent earphone as shown in Figure 1 The present application provides a tinnitus monitoring method based on an intelligent earphone as shown in The operation button is pressed to start the tinnitus detection module, the ear sound signal is collected by the spectrum analysis chip, the time domain signal collected by the spectrum analysis chip is converted into a frequency domain signal by fast Fourier transform, the tinnitus frequency range is identified, including: the collected ear sound signal is subjected to frame processing, each frame contains a fixed number of sampling points; a Hamming window function is applied to each frame of signal to reduce spectral leakage; the frequency domain components of each frame of signal are calculated by fast Fourier transform, and the frequency range with energy higher than the background noise is extracted.

[0019] The present application realizes efficient and accurate identification of the tinnitus frequency range by pressing the operation button to start the tinnitus detection module and using the spectrum analysis chip to collect and process the ear sound signal. The specific technical effects are as follows: The patient only needs to press the operation button to start the tinnitus detection module, without the need for complex equipment or professional operation, greatly simplifying the process of tinnitus monitoring and improving the user's experience and convenience. The spectrum analysis chip can collect ear sound signals in real time and perform frame processing on them, with each frame containing a fixed number of sampling points. This frame processing method makes subsequent frequency domain analysis more accurate, avoiding the problem of computational complexity caused by long continuous signal processing. In addition, frame processing helps to capture transient changes in tinnitus signals, improving detection sensitivity and accuracy.

[0020] In each frame of signal processing, a Hamming window function is applied to reduce spectral leakage. Spectral leakage refers to the energy leakage phenomenon caused by signal truncation in frequency domain analysis. By applying the Hamming window function, the leakage can be effectively reduced, improving the accuracy of the spectrum analysis results. This step ensures clear extraction of tinnitus frequency components and reduces background noise interference, allowing more accurate identification of the tinnitus frequency range.

[0021] Each frame of signal is converted from time domain to frequency domain by Fast Fourier Transform, and its frequency domain components are calculated. This process is not only fast, but also effectively reveals the frequency characteristics of tinnitus signals. By comparing the energy levels of each frequency band, the frequency range with higher energy than background noise, i.e. the main frequency component of tinnitus, can be extracted. This method significantly improves the accuracy of tinnitus frequency identification, helping users better understand their own tinnitus conditions.

[0022] Finally, the system identifies the tinnitus frequency range based on the extracted frequency domain components. Compared with traditional methods, this scheme can still maintain high recognition accuracy in complex environments (such as in the presence of environmental noise). In this way, users can obtain more detailed and accurate tinnitus frequency information, providing a solid foundation for subsequent personalized analysis.

[0023] The spectrum analysis chip obtains tinnitus loudness data through a sound pressure sensor and transmits the tinnitus frequency range and loudness data to the main control module; including: performing short-time energy calculation on ear sound signals to obtain instantaneous sound pressure values; smoothing the instantaneous sound pressure values through a sliding average filter to eliminate transient noise interference; comparing the smoothed sound pressure values with reference sound pressure values to calculate the relative change amplitude of tinnitus loudness.

[0024] The sound pressure sensor performs short-time energy calculation on the ear sound signals to obtain instantaneous sound pressure values, so that the loudness of tinnitus is no longer a subjective and vague description, but a measurable and recordable physical quantity. This method improves the objectivity and consistency of loudness information, providing a reliable basis for subsequent data analysis and long-term trend tracking.

[0025] After obtaining the instantaneous sound pressure value, a sliding average filter is used for smoothing processing. This processing method can effectively suppress the transient interference caused by environmental sudden noise, ear canal micro-motion or signal collection jitter, and avoid the sharp fluctuations of loudness data. The smoothed sound pressure value can better reflect the true level of tinnitus loudness, and enhance the continuity and credibility of the data.

[0026] The smoothed sound pressure value is compared with the preset reference sound pressure value to calculate the relative change amplitude of tinnitus loudness. This method eliminates the absolute sound pressure difference caused by individual ear canal structure, sensor fitment and other factors, so that the loudness data at different times or between different users is comparable. This relative processing provides basic support for cross-period trend analysis and personalized data modeling.

[0027] The spectrum analysis chip synchronously acquires the loudness data while completing the tinnitus frequency range recognition, and integrates and transmits the two to the main control module. This integrated data acquisition mechanism ensures the strict alignment of frequency and loudness information in time, avoids the data misplacement problem caused by multi-module asynchronous acquisition, and improves the coordination and integrity of the overall monitoring result.

[0028] The frequency and loudness data are packaged and transmitted to the main control module to provide structured input for data encryption, storage and further analysis. This design reduces the communication overhead and timing disorder risk caused by scattered data processing, and improves the overall system operation efficiency.

[0029] In summary, the present application realizes stable, objective and comparable acquisition of tinnitus loudness through short-time energy calculation, sliding average filtering and relative amplitude comparison, and forms a coordinated output with frequency information. This technical means effectively overcomes the problems of tinnitus loudness data being easily disturbed, difficult to quantify and lacking consistency in traditional monitoring, significantly improves the accuracy and practicality of tinnitus monitoring, and provides a solid technical support for users to track the tinnitus state for a long time.

[0030] The main control module stores the tinnitus frequency range and loudness data in the data storage module through the AES encryption standard, including: dividing the tinnitus frequency range and loudness data into fixed-size data blocks; applying the AES encryption algorithm to each data block to generate corresponding ciphertext data; writing the ciphertext data into the specified storage area of the data storage module. And send the tinnitus frequency range and loudness data to the rehabilitation sound synthesis module; Tinnitus frequency and loudness data belong to personal health information and have high sensitivity. By dividing the data into fixed-size data blocks and independently applying the AES encryption algorithm to each data block to generate ciphertext data, it is ensured that the original data does not exist in plaintext form during storage. Even if the data storage module is accessed or physically extracted without authorization, the original tinnitus information cannot be restored, effectively preventing privacy leakage.

[0031] The tinnitus data is divided into data blocks of a fixed size, facilitating the master module to uniformly manage and schedule the storage space. This approach supports segmented writing and reading of data, improving the efficiency and stability of storage operations, avoiding fragmentation problems caused by different data lengths, and enhancing the long-term operation reliability of the data storage module.

[0032] The AES encryption standard is used as the encryption algorithm, which has extensive technical support and hardware acceleration capabilities, and can be efficiently executed under the limited computing resources of the intelligent earphone. The encrypted ciphertext data conforms to the general security specifications, providing a secure foundation for subsequent uploading to external devices or the cloud through the communication module, ensuring the confidentiality of end-to-end data transmission.

[0033] While completing the encrypted storage, the master module directly sends the original tinnitus frequency range and loudness data to the rehabilitation sound synthesis module, avoiding the introduction of processing delays due to encryption processes that affect the real-time performance of subsequent sound generation. This parallel processing mechanism decouples "secure storage" and "function flow", meeting the data security requirements while ensuring system response speed.

[0034] Ciphertext data is written to the designated storage area of the data storage module, forming an ordered data record sequence. This design supports data retrieval and playback based on timestamps or other identifiers, providing a data foundation for users to view historical tinnitus states and analyze trends, enhancing the system's long-term monitoring capabilities.

[0035] The rehabilitation sound synthesis module generates personalized control signals based on the tinnitus frequency range and loudness data, including: matching the tinnitus frequency range with the rehabilitation sound frequency through a pre-set mapping rule; adjusting the amplitude of the rehabilitation sound according to the tinnitus loudness data to maintain dynamic balance between the rehabilitation sound intensity and the tinnitus loudness; optimizing control signal parameters through historical rehabilitation records. And transmit the personalized control signal to the programmable waveform generator to generate rehabilitation sound, including: synthesizing the composite signal of white noise and reverse phase masking sound through the programmable waveform generator; adjusting the amplitude ratio of white noise and masking sound according to the personalized control signal; removing high-frequency noise components through low-pass filtering of the synthesized signal by the digital signal processing module. The programmable waveform generator outputs the rehabilitation sound through the sound output unit.

[0036] The detected tinnitus frequency range is matched with the rehabilitation sound frequency through a pre-set mapping rule, ensuring that the generated sound signal matches the user's perceived tinnitus components in the frequency domain. This mapping mechanism avoids the inconsistency between general sound solutions and individual tinnitus characteristics, making the output sound more consistent with the user's auditory perception characteristics, improving the relevance and adaptability of sound output.

[0037] The amplitude of the rehabilitation sound is adjusted according to the acquired tinnitus loudness data, so that the intensity of the rehabilitation sound and the loudness of the tinnitus are kept in dynamic balance. This way avoids the situation that the sound is too strong to cause auditory discomfort or too weak to have an effect, so that the output sound is in the user's acceptable and effective perception interval, enhancing the comfort and consistency during use.

[0038] The control signal parameters are optimized by calling historical rehabilitation records, so that the newly generated control signal is adjusted on the basis of continuing the past effective settings. This mechanism avoids starting from zero every time and improves the stability and continuity of the sound output strategy, which helps to form a regular sound exposure pattern and supports the perception adaptation process in the long term.

[0039] The composite signal of white noise and reverse phase masking sound is synthesized by a programmable waveform generator, which comprehensively utilizes the masking effect of wideband noise and the cancellation potential of specific frequency reverse signals. This composite structure enhances the intervention ability of tinnitus perception without relying on high-intensity output, making the sound output more layered and covering a wider range of effects.

[0040] The amplitude ratio between white noise and masking sound is adjusted according to the individualized control signal, so that the composition of the composite signal can be flexibly adjusted according to different tinnitus characteristics. For wideband tinnitus, the proportion of white noise can be increased to enhance the overall masking effect; for narrowband high-frequency tinnitus, the weight of reverse phase masking sound can be increased to achieve targeted intervention, improving the flexibility and individual adaptation ability of sound generation.

[0041] The synthesized signal is subjected to low-pass filtering by a digital signal processing module to remove high-frequency noise components and avoid introducing sharp frequencies that may cause auditory fatigue. This processing makes the output sound smoother and more natural, reduces the stimulation of healthy hearing areas, and improves the tolerance and long-term usability of wearing and using.

[0042] The individualized control signal is converted into a specific sound waveform by a programmable waveform generator and output to the user's ear in real time by a sound output unit, realizing a complete closed loop from tinnitus characteristic recognition to sound response. The entire process does not require external intervention, and users can independently complete sound matching and output in daily environments, significantly improving the independent operation ability and convenience of the system.

[0043] Further, the main control module transmits the tinnitus frequency range and loudness data to the mobile phone application through the Bluetooth module, including: packaging the tinnitus frequency range and loudness data into a Bluetooth data packet; sending the data packet to the mobile phone device through the Bluetooth protocol stack; the mobile phone device receives the data packet and parses and stores it to the local database.

[0044] The tinnitus frequency range and loudness data are packaged as Bluetooth data packets and sent to the mobile terminal device through the Bluetooth protocol stack, thus breaking away from the dependence on physical connection. The user can complete data uploading while wearing the smart earphone, without the need to plug in a cable or manually export, thus improving the convenience and continuity of data transmission and supporting continuous tracking of the tinnitus state.

[0045] Ensure the structuring and integrity of data transmission: through a standardized data packaging method, the frequency and loudness information is organized into a data packet format that conforms to the Bluetooth communication specification, ensuring that the information does not become misaligned, lost or confused during wireless transmission. This mechanism improves the reliability of data transmission over the air interface, providing a basic guarantee for the mobile terminal to accurately restore the original monitoring results.

[0046] The mobile terminal device receives and analyzes the Bluetooth data packets and stores the analyzed data in the local database, realizing the terminal storage and management of tinnitus information. The user can view historical records and analyze trends on the mobile phone without relying on the local display function of the earphone, thus expanding the accessibility and use scenarios of the data.

[0047] After the data enters the mobile terminal, the tinnitus frequency and loudness change curve can be generated in combination with the timestamp to visually present the tinnitus fluctuation law. The user can understand the evolution of their own state through intuitive charts, enhancing their understanding of tinnitus and self-management awareness, and providing a basis for subsequent behavior adjustment.

[0048] The Bluetooth transmission mechanism reserves interfaces for subsequent function expansion of the system. On the basis of basic data uploading, the mobile terminal can further support parameter setting feedback, sound mode switching instruction issuance and other bidirectional interaction, so that the smart earphone and the mobile terminal form a collaborative working system, improving the flexibility and intelligent level of the overall system.

[0049] In summary, the present application realizes wireless transmission of tinnitus frequency and loudness data to the mobile terminal through the Bluetooth module, completing the data connection from the acquisition terminal to the user interface. This technical means effectively solves the problems of data isolation, difficulty in viewing and inability to track long-term in traditional tinnitus monitoring, enhances the usability and user engagement of the system, and provides reliable technical support for realizing the daily and visual management of tinnitus state.

[0050] The mobile terminal application displays the change trend of tinnitus frequency over time in the form of a line chart and uploads the data to the cloud server through the Wi-Fi module, including: sending the locally stored tinnitus data to the cloud server through the HTTP protocol; the cloud server performs integrity check on the received data; generating a long-term trend report of the patient's tinnitus condition through a data visualization tool.

[0051] The distribution of tinnitus frequency at different time points is displayed through a line chart, enabling the user to clearly identify the fluctuation pattern of tinnitus frequency, such as whether it tends to be stable, gradually shifts or experiences sudden changes. This visualization method converts abstract numerical data into easily understood graphical information, significantly improving the user's cognitive ability of their own tinnitus state.

[0052] The line chart organizes data in chronological order, covering a range of days, weeks or even longer, helping users and relevant personnel identify the evolution of tinnitus characteristics. By observing the frequency trend, it can be determined whether there are periodic changes or a continuous deterioration trend, providing a reference for daily management.

[0053] The tinnitus data stored in the local database of the mobile phone is sent to the cloud server through the HTTP protocol, and the data upload is completed using standard network communication mechanisms. This method is compatible with existing network infrastructure and does not require special equipment or complex configuration, allowing remote collection of data and ensuring centralized management of data.

[0054] After receiving the tinnitus data, the cloud server performs integrity verification to check whether the data packet is complete, the timestamp is continuous, and the value is within a reasonable range. This mechanism can effectively identify packet loss, out-of-order or abnormal data that may occur during transmission, preventing incorrect information from entering the long-term record and improving the credibility of the remote data set.

[0055] The data visualization tool deployed on the cloud integrates the accumulated tinnitus data and automatically generates trend reports containing frequency change trajectories, fluctuation intervals, extreme points and other information. This report supports multi-dimensional retrospective analysis, providing systematic support for users to review historical status and assess change processes.

[0056] The data stored on the cloud and the generated trend reports can be accessed by authorized personnel. With user permission, doctors or hearing professionals can remotely view the long-term evolution of their tinnitus frequency, providing data support for subsequent personalized recommendations and enhancing the system's collaborative role in health management.

[0057] The cloud-based data architecture supports subsequent functional expansion, such as multi-user data comparison, statistical analysis model access, and abnormal warning mechanisms. This design leaves room for future system upgrades, enhancing the sustainable development capabilities of the overall solution.

[0058] In summary, the present application displays the tinnitus frequency change trend through a line chart and combines Wi-Fi upload and cloud processing, achieving the visualization, long-termization and remote management of monitoring data.

[0059] In another aspect, the present application proposes a tinnitus monitoring system based on smart earphones. This system not only accurately captures tinnitus signals but also generates personalized rehabilitation sounds based on the patient's tinnitus characteristics, achieving continuous and effective tinnitus management. The entire process involves the coordinated work of multiple modules, including tinnitus detection, data processing, rehabilitation sound synthesis, communication, and power supply, etc.

[0060] Firstly, in the tinnitus detection phase, the patient presses the operation button to start the tinnitus detection module. The spectrum analysis chip collects and processes the ear sound signals, converts the time domain signals into frequency domain signals through the Fast Fourier Transform (FFT) algorithm, and thus identifies the tinnitus frequency range. At the same time, the sound pressure sensor monitors the real-time changes in ear sound intensity and obtains the tinnitus loudness data. These data are then transmitted to the main control module for storage and further processing.

[0061] In the data processing phase, the main control module receives data from the tinnitus detection module and uses the AES encryption standard to securely store it in the flash memory chip. In addition, the main control module also sends data to the rehabilitation sound synthesis module and the communication module. The rehabilitation sound synthesis module analyzes the tinnitus data based on a pre-trained deep learning model, generates personalized control signals, and guides the programmable waveform generator to produce rehabilitation sounds. This step takes into account the specific values of tinnitus frequency and loudness, as well as the patient's historical rehabilitation records, to ensure the effectiveness of the rehabilitation program.

[0062] Next, the communication module transmits the tinnitus detection results and rehabilitation records in real time to the mobile phone application through Bluetooth technology. The application displays the trends of tinnitus frequency and loudness in intuitive charts, helping patients understand their own condition. At the same time, doctors can access patient data through a cloud server to evaluate the current rehabilitation effect and make adjustment suggestions. When the earphone is in a Wi-Fi network environment, data will be uploaded to the cloud for remote medical monitoring.

[0063] Finally, the power supply module of the system uses high-energy-density rechargeable lithium batteries to ensure long-term stable operation. The power management chip monitors the battery status in real time and reminds users to charge in time through the status indicator light. Users can charge the earphone through the Type-C interface or wireless charging equipment to ensure the continuous availability of the system.

[0064] Through the above steps, the present application provides a complete tinnitus monitoring and rehabilitation process, achieving closed-loop management from detection to rehabilitation and feedback adjustment, greatly improving the rehabilitation experience of tinnitus patients.

[0065] The tinnitus monitoring system based on smart earphones specifically includes: An operation button for starting the tinnitus detection module; A spectrum analysis chip is connected with the operation button, used for collecting ear sound signals and performing spectrum analysis; A sound pressure sensor is connected with the spectrum analysis chip, used for obtaining tinnitus loudness data; A main control module is connected with the spectrum analysis chip and the sound pressure sensor, used for receiving and processing tinnitus data; A data storage module is connected with the main control module, used for storing encrypted tinnitus data; A rehabilitation sound synthesis module is connected with the main control module, used for generating personalized control signals; A programmable waveform generator is connected with the rehabilitation sound synthesis module, used for generating rehabilitation sounds; A sound output unit is connected with the programmable waveform generator, used for outputting rehabilitation sounds; A Bluetooth module is connected with the main control module, used for transmitting tinnitus data to a mobile phone application program; A Wi-Fi module is connected with the main control module, used for uploading data to a cloud server.

[0066] In addition, the above components are also used to implement other steps of the above-mentioned tinnitus monitoring method based on the smart earphone when they are executed, as follows: Tinnitus detection starts and signal collection: The patient presses the operation button on the smart earphone to start the tinnitus detection module. This operation triggers the built-in spectrum analysis chip to start working. After the chip is initialized, it immediately enters standby mode and is ready to collect ear sound signals.

[0067] The spectrum analysis chip starts to collect ear sound signals in real time. Assuming that the collection time is seconds, the sampling rate is , then a total of sample points are collected.

[0068] During the collection process, the adaptive filtering algorithm dynamically adjusts the filtering parameters, and the formula is as follows: ; wherein, represents the transfer function of the filter, and are the numerator and denominator coefficients respectively. This algorithm automatically adjusts the coefficients according to the real-time collected signal characteristics to suppress environmental noise.

[0069] The filtered signal needs to be further denoised and smoothed. A window function is used for weighted averaging, and the formula is as follows: ; wherein, is the window function weight, is the input signal, is the output signal. This step helps to remove high-frequency noise and retain effective tinnitus signal components.

[0070] The time-domain signal is converted to a frequency-domain signal using the Fast Fourier Transform (FFT) for subsequent spectral analysis. The FFT formula is: ; where is the th frequency component in the frequency domain, is the th sample point in the time domain. This conversion allows for clear visualization of the amplitude of each frequency component.

[0071] In the frequency-domain signal, look for frequency components significantly higher than background noise as the tinnitus frequency. Let the tinnitus frequency be , then: ; where is a set threshold value to distinguish between background noise and tinnitus signal. Through this process, the main frequency range of tinnitus can be accurately identified.

[0072] Tinnitus loudness measurement: The sound pressure sensor monitors the ear sound intensity changes in real time to obtain tinnitus loudness data. The sound pressure level The calculation formula is: ; where is the actual measured sound pressure value, is the reference sound pressure (20\muPa). Through this calculation, the loudness level of tinnitus can be quantified.

[0073] The detected tinnitus frequency and loudness data are integrated into a data packet, formatted and transmitted to the main control module through the internal bus. Assuming the data packet size is D bytes and the transmission rate is Rbps, the transmission time t is: This process ensures that the data is transmitted to the main control module without errors, preparing for the next step of data processing.

[0074] After receiving the data packet transmitted by the tinnitus detection module, the main control module immediately decodes and stores it to the data storage module. The data storage module uses the AES encryption standard to protect the security of the data. The basic principle of AES encryption is to transform the original data into unreadable ciphertext through a series of complex mathematical operations.

[0075] Specifically, the AES encryption algorithm uses a block cipher mechanism, where each data block is 128 bits (16 bytes) long, and the encryption key length can be 128 bits, 192 bits, or 256 bits. The encryption process can be represented by the following formula: ; Where C represents ciphertext, E_k represents encryption function, and P represents plaintext. This step ensures the secure storage of sensitive data, preventing unauthorized access.

[0076] After storage is complete, the main control module performs further analysis on the tinnitus data. First, the tinnitus frequency and loudness data are classified for subsequent processing. Assuming there is a set of N tinnitus data, each data includes tinnitus frequency and loudness , a classification function can be defined as follows: ; Where and are the frequency and loudness thresholds, respectively. Through this classification function, severe tinnitus cases that require special attention can be filtered out, allowing for the development of more targeted rehabilitation programs.

[0077] After classification, the main control module sends the data to the rehabilitation sound synthesis module. In this process, the data undergoes certain compression and encoding processing to reduce transmission bandwidth occupancy. Assuming the original data size is D bytes and the compression ratio is r, the compressed data size is: ; The transmission rate remains Rbps, so the transmission time becomes ; This optimization improves data transmission efficiency, shortens waiting time, and ensures that the rehabilitation sound synthesis module can obtain the required data as soon as possible.

[0078] To ensure data integrity and reliability, the main control module also regularly synchronizes important data to external storage devices or cloud servers. The synchronization process involves data verification, one of the commonly used verification algorithms is CRC (Cyclic Redundancy Check), whose formula is: ; Where M(x) is the message polynomial, G(x) is the generator polynomial, and n is the order of the generator polynomial. Through CRC verification, errors in the transmission process can be detected and corrected, ensuring data accuracy.

[0079] After receiving the data sent by the main control module, the rehabilitation sound synthesis module first parses it. Assuming the received data packet is , containing tinnitus frequency and loudness information. The parsing process mainly restores the data packet to its original format for subsequent processing. The parsing formula is as follows: ; Where P represents the parsed data set, containing specific tinnitus frequency and loudness values. This step ensures data integrity and correctness, laying the foundation for subsequent generation of personalized rehabilitation programs.

[0080] The parsed data is fed into a pre-trained deep learning model for analysis. This model is trained on a large amount of tinnitus patient data and can generate personalized control signals based on different patients' tinnitus frequencies, loudness, and historical rehabilitation records. The core of the model is a multi-layer neural network, and its forward propagation process can be represented by the following formula: ; Where Y is the output control signal, is the activation function (such as ReLU), W is the weight matrix, X is the input data (tinnitus frequency and loudness), and b is the bias term. Through learning on a large amount of data, the model can accurately predict the rehabilitation strategy suitable for each patient.

[0081] Based on the output of the deep learning model, the corresponding control signal is generated to guide the programmable waveform generator to generate rehabilitation sound. For example, for a high-frequency tinnitus patient, the model may generate instructions for the programmable waveform generator to generate low-frequency masking sound. Assuming the tinnitus frequency is , the control signal S is represented as: ; Where is the control signal generation function, which selects the appropriate masking sound frequency according to the tinnitus frequency. For different types of tinnitus, this function dynamically adjusts the generated control signal to maximize rehabilitation effectiveness.

[0082] According to the generated control signal, the programmable waveform generator synthesizes the corresponding sound waveform. Assuming that a composite rehabilitation sound needs to be generated, combining white noise and reverse phase masking sound, its synthesis formula can be represented as: ; Where and are the amplitudes of white noise and masking sound, and are the frequencies of white noise and masking sound, is the phase difference. In this way, a sound waveform with rehabilitation effect can be generated to help patients alleviate tinnitus symptoms.

[0083] Finally, the synthesized sound waveform is played through the sound emitting unit. Assuming the output power of the sound emitting unit is , the output sound pressure level can be represented as: ; Where is the reference power (taking ). By adjusting the output power, the intensity of the sound can be flexibly adjusted to ensure that the patient receives rehabilitation within a comfortable range. If the patient feels that the sound volume is not appropriate, they can adjust it through the operation of the buttons or the mobile app.

[0084] The communication module plays a crucial interactive role throughout the process. First, tinnitus detection data and rehabilitation records are transmitted in real-time to the mobile application via Bluetooth technology. The basic principle of Bluetooth transmission is short-range wireless communication using the 2.4GHz ISM band, with data transmission rates typically between 1Mbps and 3Mbps. Assuming the amount of data transmitted each time is... bytes, transmission time is Seconds, then the transmission rate for: ; This efficient transmission method ensures the real-time nature and integrity of the data, allowing patients to check their tinnitus status immediately.

[0085] After receiving the data, the mobile application will display the tinnitus frequency over time and the fluctuations in tinnitus loudness on a data visualization interface in the form of line charts, bar charts, etc. Assuming there is a period of time ( Tinnitus frequency data (seconds) The formula for drawing a line chart is: ; in, It is a unit impulse function. The time intervals are used to represent the time points. This intuitive presentation helps patients better understand their tinnitus condition and its changing trends, thereby enhancing their confidence and motivation for recovery.

[0086] In addition to data visualization, the mobile application also provides rehabilitation plan management functions. Patients can view historical rehabilitation records, understand their past rehabilitation progress, and set personalized rehabilitation parameters, such as preferred rehabilitation sound types. For example, if a patient selects a certain type of rehabilitation sound, their preference setting can be represented as: ; in, Different types of rehabilitation sounds are available. This personalized setting allows patients to choose the most suitable rehabilitation program based on their preferences and comfort, improving rehabilitation outcomes.

[0087] If a patient wants to switch rehabilitation modes or adjust the volume, they can do so via buttons or a mobile application. For example, if a patient wants to switch from mode 1 to mode 2, the process can be represented as follows: ; The main control module controls the rehabilitation sound synthesis module to switch between different rehabilitation sound generation methods according to instructions. Similarly, volume adjustment can also be achieved through similar operations, ensuring that patients can adjust to the most suitable state at any time during the rehabilitation process.

[0088] When the earphone is in a Wi-Fi network environment, the Wi-Fi module of the communication module uploads the data in the data storage module to the cloud server. Assuming that the data upload speed is U Mbps, the total amount of data is D bytes, and the upload time T is: The doctor accesses the patient data in the cloud server through a special medical data management platform, and comprehensively analyzes the tinnitus condition and rehabilitation effect of the patient. For example, the doctor can observe the change trend of the tinnitus frequency and loudness of the patient within a period of time, evaluate the effectiveness of the current rehabilitation plan, and propose individualized rehabilitation suggestions and plan adjustment notifications according to the analysis results.

[0089] Through the above steps, the communication module not only realizes efficient transmission and display of data, but also provides a convenient interactive platform for patients and doctors, promoting the individualization and scientific management of tinnitus rehabilitation.

[0090] The power supply module of the system uses high-energy-density rechargeable lithium batteries to ensure long-term stable operation. Assuming that the battery capacity is Q mAh and the rated voltage is V volts, the energy E of the battery can be represented as: This design provides sufficient power support for each module of the earphone, ensuring stable performance even in long-term use.

[0091] The power management chip monitors the charging and discharging state of the battery in real time to ensure that the battery is always in the best working state. Assuming that the remaining battery capacity is S percent, its remaining energy ; When the battery level is below a certain threshold, the power management chip will remind the user to charge in time through the state indicator light in color or flashing mode. For example, red flashing may indicate that the battery is low and needs to be charged immediately; blue constant indicates that the battery is sufficient and can be used normally.

[0092] Users can charge the earphone through the Type-C interface using a charging cable, or place the earphone on a charging device that supports wireless charging to charge through the wireless charging receiving coil.

[0093] The operation button, touch sensing area and state indicator light provide users with a convenient interaction method. The operation button not only realizes basic functions such as on-off, detection start, mode switching and volume adjustment, but also can realize some advanced functions such as quick start rehabilitation mode through different key combinations.

[0094] ​​​​The touch sensing area supports the user to control the function by sliding, double-clicking and other touch operations, for example, the user can adjust the volume by sliding on the touch sensing area when listening to music, without the need to find the operation button specially. The status indicator light clearly shows the working status of the earphone by different colors and flashing frequencies, such as blue constant light indicating that the Bluetooth is connected, red flashing indicating that it is charging, etc.

[0095] Through the above steps, the present application not only provides a comprehensive tinnitus monitoring and rehabilitation solution, but also makes careful design in power management and user interaction, ensuring the high efficiency, reliability and ease of use of the system.

[0096] Finally, it should be noted that: the above only for the preferred embodiments of the present application, and not for limiting the present application, although the foregoing detailed description of the present application is made with reference to the foregoing embodiments, for those skilled in the art, it still can be modified to the technical solutions recorded in the foregoing embodiments, or equivalent replacement of some technical features, any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application, should be included in the protection scope of the present application.

Claims

1. A tinnitus monitoring method based on smart headphones, characterized in that, Includes the following steps: Press the operation button to start the tinnitus detection module. The spectrum analysis chip collects the sound signal from the ear. The spectrum analysis chip converts the collected time-domain signal into a frequency-domain signal through a fast Fourier transform to identify the tinnitus frequency range. The spectrum analysis chip acquires tinnitus loudness data through a sound pressure sensor and transmits the tinnitus frequency range and loudness data to the main control module. The main control module stores the tinnitus frequency range and loudness data in the data storage module using the AES encryption standard, and sends the tinnitus frequency range and loudness data to the rehabilitation sound synthesis module; The rehabilitation sound synthesis module generates a personalized control signal based on the tinnitus frequency range and loudness data, and transmits the personalized control signal to the programmable waveform generator to generate rehabilitation sound. The programmable waveform generator outputs the rehabilitation sound through the sound generation unit.

2. The tinnitus monitoring method based on smart headphones according to claim 1, characterized in that, The spectrum analysis chip converts the acquired time-domain signal into a frequency-domain signal using a fast Fourier transform, including: The collected ear sound signals are processed by frame segmentation, with each frame containing a fixed number of sampling points; Apply a Hamming window function to each frame of the signal to reduce spectral leakage; The frequency domain components of each frame of signal are calculated by Fast Fourier Transform, and the frequency range with energy higher than the background noise is extracted.

3. The tinnitus monitoring method based on smart headphones according to claim 1, characterized in that, The spectrum analysis chip acquires tinnitus loudness data through a sound pressure sensor, including: Short-time energy calculation is performed on the sound signal to the ear to obtain the instantaneous sound pressure value; The instantaneous sound pressure value is smoothed by using a moving average filter to eliminate transient noise interference; The smoothed sound pressure level is compared with the reference sound pressure level to calculate the relative change in tinnitus loudness.

4. The tinnitus monitoring method based on smart headphones according to claim 1, characterized in that, The main control module stores the tinnitus frequency range and loudness data to the data storage module using the AES encryption standard, including: The tinnitus frequency range and loudness data are divided into data blocks of fixed size; Apply the AES encryption algorithm to each data block to generate the corresponding ciphertext data; Write the encrypted data to the designated storage area of ​​the data storage module.

5. The tinnitus monitoring method based on smart headphones according to claim 1, characterized in that, The generation of personalized control signals includes: The tinnitus frequency range is matched with the rehabilitation sound frequency through a preset mapping rule; The amplitude of the rehabilitation sound is adjusted according to the tinnitus loudness data to maintain a dynamic balance between the intensity of the rehabilitation sound and the tinnitus loudness. Optimize control signal parameters using historical rehabilitation records.

6. The tinnitus monitoring method based on smart headphones according to claim 1, characterized in that, The generation of rehabilitation sounds includes: A composite signal of white noise and inverse phase masking tone is synthesized using a programmable waveform generator; Adjust the amplitude ratio of white noise and masking tone according to the personalized control signal; The synthesized signal is low-pass filtered by a digital signal processing module to remove high-frequency noise components.

7. The tinnitus monitoring method based on smart headphones according to claim 1, characterized in that, It also includes the following steps: The main control module transmits the tinnitus frequency range and loudness data to the mobile application via Bluetooth. The mobile application displays the trend of tinnitus frequency over time in the form of a line graph and uploads the data to the cloud server via a Wi-Fi module.

8. The tinnitus monitoring method based on smart headphones according to claim 7, characterized in that, The main control module transmits the tinnitus frequency range and loudness data to the mobile application via Bluetooth, including: Encapsulate the tinnitus frequency range and loudness data into a Bluetooth data packet; Data packets are sent to the mobile device via the Bluetooth protocol stack; The mobile device receives the data packet, parses it, and stores it in the local database.

9. A tinnitus monitoring method based on smart headphones according to claim 7, characterized in that, Uploading data to the cloud server via the Wi-Fi module includes: The locally stored tinnitus data is sent to the cloud server via the HTTP protocol; The cloud server performs integrity checks on the received data; Long-term trend reports on patients' tinnitus conditions are generated using data visualization tools.

10. A tinnitus monitoring system based on smart headphones for implementing the method as described in any one of claims 1-9, characterized in that, include: The operation button is used to activate the tinnitus detection module; The spectrum analysis chip, connected to the operation buttons, is used to collect ear sound signals and perform spectrum analysis. A sound pressure sensor, connected to a spectrum analysis chip, is used to acquire tinnitus loudness data; The main control module, connected to the spectrum analysis chip and sound pressure sensor, is used to receive and process tinnitus data; The data storage module, connected to the main control module, is used to store encrypted tinnitus data; The rehabilitation sound synthesis module, connected to the main control module, is used to generate personalized control signals; A programmable waveform generator, connected to a rehabilitation sound synthesis module, is used to generate rehabilitation sounds; The sound-generating unit, connected to a programmable waveform generator, is used to output rehabilitation sounds; The Bluetooth module, connected to the main control module, is used to transmit tinnitus data to the mobile application. The Wi-Fi module connects to the main control module and is used to upload data to the cloud server.

Images