Intelligent self-service pure tone hearing threshold test method, equipment and medium

Abstract

The invention discloses an intelligent self-service pure tone threshold test method and device and a medium, and relates to the technical field of threshold test, and the method comprises the steps: executing a pure tone threshold test process based on an information gain selection mechanism according to the recognition difficulty level in noise, the matching degree score and a pure tone threshold test starting strategy, the key pressing behavior of the testee is judged in real time in combination with the personal operation rhythm template, the hearing threshold estimation result of each frequency band and the corresponding uncertainty index and behavior event mark are output, and fixed-point deviation correction complementary measurement and weight correction are carried out based on the hearing threshold estimation result of each frequency band, the uncertainty index and the behavior event mark; obtaining a final hearing threshold result and an overall credibility grade; by establishing the candidate test set and introducing the pure tone threshold measurement process based on the information gain selection mechanism, the dynamic adjustment of the test path according to the hearing threshold estimation result and the uncertainty index of each frequency band is realized, and the test efficiency and the result stability are improved.

Description

Technical Field

[0001] This invention relates to the field of hearing threshold testing technology, and in particular to an intelligent self-service pure tone hearing threshold testing method, device and medium. Background Technology

[0002] Pure-tone audiometry is a fundamental testing method in audiological assessment, widely used in hearing screening and hearing loss classification. Traditional pure-tone audiometry often employs manually controlled ascending / descending methods or fixed-step methods. In a relatively quiet environment, professionals adjust the sound intensity level frequency by frequency, determining the hearing threshold for each frequency band based on the subject's subjective response. With the increasing demand for grassroots screening and self-service testing, self-service pure-tone audiometry based on computer terminals or mobile devices is gradually being promoted. This automates the testing process through pre-set workflows, reducing labor costs and increasing testing coverage. Some solutions introduce automatic step-step methods or simple adaptive sound intensity adjustment strategies to improve testing efficiency.

[0003] However, existing self-service pure-tone hearing threshold testing methods mostly use fixed step methods or simple adaptive rules for test path selection, lacking a dynamic decision-making mechanism based on changes in information content. This can easily lead to repeated stimuli or invalid tests, affecting test efficiency and result stability. At the same time, most schemes do not model the rhythm of the test subject's operation, nor do they output uncertainty indicators in the hearing threshold estimation process. It is difficult to distinguish between false triggering behavior and real auditory response, and it is impossible to quantitatively evaluate the reliability of the test results. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides an intelligent self-service pure tone hearing threshold testing method to solve the problem of lacking a dynamic decision-making mechanism based on changes in information content.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides an intelligent self-service pure-tone hearing threshold testing method, which includes, The test subject's personal operating rhythm template and environmental noise profile were obtained, and a two-tone audiometry was performed to obtain the noise recognition difficulty level, cooperation score, and pure tone threshold initiation strategy. Based on the noise identification difficulty level, cooperation score, and pure tone thresholding initiation strategy, a pure tone thresholding process based on information gain selection mechanism is executed. Combined with the personal operation rhythm template, the key press behavior of the test subject is judged in real time, and the hearing threshold estimation results of each frequency band and the corresponding uncertainty index and behavioral event markers are output. Based on the hearing threshold estimation results of each frequency band, uncertainty index and behavioral event markers, targeted correction and supplementary testing and weight adjustment are performed to obtain the final hearing threshold result and overall credibility level. The hearing screening results are determined based on the final hearing threshold and overall reliability level, combined with the level of difficulty in identifying noise, and an automated screening report and referral recommendations are generated.

[0007] As a preferred embodiment of the intelligent self-service pure tone hearing threshold testing method of the present invention, the personal operation rhythm template is constructed by performing a visual button calibration task, recording the button response time sequence, number of early button presses, number of delayed button presses, and consecutive button press intervals after multiple visual stimulus triggers, and based on the button response time sequence, number of early responses, and number of delayed responses. The personal operation rhythm template includes the button response time distribution range, the statistical results of the number of early button presses, and the statistical results of the number of delayed button presses.

[0008] As a preferred embodiment of the intelligent self-service pure tone hearing threshold testing method of the present invention, the environmental noise profile is constructed by starting the environmental sound acquisition process while building a personal operation rhythm template, acquiring the environmental sound signal within a preset time window and extracting the environmental sound pressure level, spectral distribution characteristics and fluctuation amplitude sequence.

[0009] As a preferred embodiment of the intelligent self-service pure-tone hearing threshold testing method of the present invention, the method for obtaining the identification difficulty level, cooperation score, and pure-tone threshold testing starting strategy in noise specifically comprises: Construct a test sequence, play two-part digital speech one after another according to the preset playback intensity adjustment rules, and record the corresponding input results and response time sequence of the test subject for each playback; The number of correct and incorrect recognitions is counted based on the test subject's input, and the recognition difficulty level in the noise is generated by combining the playback intensity adjustment process. Based on the matching analysis between the response time series and the key response time distribution range in the personal operation rhythm template, the number of valid responses, the number of timeout responses and the number of mis-inputs are counted to generate a cooperation score. Based on the mapping between the difficulty level of noise identification and the cooperation score, the initial sound intensity level, initial test frequency band, and test step strategy for pure tone thresholding are determined, thus forming the initial strategy for pure tone thresholding.

[0010] As a preferred embodiment of the intelligent self-service pure-tone hearing threshold testing method of the present invention, the pure-tone threshold testing process based on the information gain selection mechanism specifically comprises: The initial test frequency band in the pure tone thresholding starting strategy is used as the test frequency band in the candidate test set. Multiple sound intensity levels are formed by expanding upward and downward according to the test stepping strategy, with the pure tone thresholding starting sound intensity level as the center. The test frequency bands and corresponding sound intensity levels are combined to form the candidate test set. The test frequency bands in the candidate test set are screened and sorted by combining the identification difficulty level in the noise, the cooperation score, and the environmental noise profile. The uncertainty index of each test frequency band is initialized. The candidate test set includes multiple test frequency bands and corresponding sound intensity levels. Before each round of pure tone stimulation, the degree of reduction of the overall uncertainty index by each test frequency band and sound intensity level is calculated based on the current hearing threshold estimation results and uncertainty index of each frequency band. The test frequency band and sound intensity level with the greatest degree of uncertainty reduction are selected as the next pure tone stimulation. Play the selected pure tone stimulus and record the subject's key press response. Combine the personal operation rhythm template to determine the validity of the key press response and generate the corresponding behavioral event tag. Based on the key response results and behavioral event markers, update the hearing threshold estimation results and uncertainty index of the corresponding test frequency band, and update the candidate test set simultaneously until the uncertainty index of all test frequency bands meets the preset convergence condition, and output the hearing threshold estimation results of each frequency band and the corresponding uncertainty index and behavioral event markers.

[0011] As a preferred embodiment of the intelligent self-service pure-tone hearing threshold testing method of the present invention, the method for obtaining the final hearing threshold result and the overall reliability level specifically includes: Consistency verification is performed on the uncertainty index and behavioral event markers corresponding to the hearing threshold estimation results of each frequency band. Based on the consistency verification results, preset convergence conditions and behavioral event marker triggering conditions, the target test frequency bands that need to be subjected to fixed-point correction and supplementary testing are selected. A new candidate test set was established for the target test frequency band, and pure tone stimulation was played. The key response results and corresponding behavioral event tags during the supplementary test phase were recorded, and the hearing threshold estimation results and uncertainty index of the target test frequency band were updated. The updated hearing threshold estimation results for each frequency band are processed by weight correction. Frequency band weights are assigned according to the uncertainty index and behavioral event markers corresponding to each frequency band, and the final hearing threshold results are calculated. The overall credibility level is generated by comprehensively evaluating the hearing threshold estimation results of each frequency band, the corresponding uncertainty indicators, and behavioral event markers.

[0012] As a preferred embodiment of the intelligent self-service pure-tone hearing threshold testing method of the present invention, the generation of automated screening reports and referral recommendations specifically includes: Extract the final hearing threshold results, overall confidence level, and identification difficulty level in noise to construct a screening result dataset; The final hearing threshold results are compared with the preset hearing grading standards to determine the level, and the validity of the results is generated by combining the overall credibility level. The referral recommendation level is generated by matching the final hearing threshold result with the hearing classification result, the overall confidence level, and the level of difficulty in identifying noise against the preset screening rule table. The final hearing threshold results, overall reliability level, difficulty level of identification in noise, hearing classification results, and referral recommendation level are uniformly packaged to generate automated screening reports and referral recommendations.

[0013] As a preferred embodiment of the intelligent self-service pure-tone hearing threshold testing method of the present invention, the consistency verification refers to the correlation and comparison of the changing trends of hearing threshold estimation results of each frequency band and the corresponding uncertainty indicators and behavioral event markers.

[0014] In a second aspect, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the intelligent self-service pure tone hearing threshold testing method as described in the first aspect of the present invention.

[0015] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the intelligent self-service pure tone hearing threshold testing method as described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: By establishing a candidate test set and introducing a pure tone thresholding process based on an information gain selection mechanism, the test path is dynamically adjusted according to the hearing threshold estimation results and uncertainty index of each frequency band, so that the pure tone stimulus is preferentially applied to the test point with the highest information contribution, avoiding repeated stimulation and invalid tests caused by the fixed step method, thereby improving test efficiency and result stability. At the same time, by constructing a personal operation rhythm template and judging the key press behavior in real time, combined with the output of uncertainty index and behavioral event markers, the invention distinguishes between false triggering behavior and real auditory response, and establishes a quantitative expression mechanism for the stability of hearing threshold estimation, so that the test results not only have numerical output, but also have the ability to evaluate credibility. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1This is a flowchart of an intelligent self-service pure-tone hearing threshold testing method.

[0019] Figure 2 A flowchart for generating personal operation rhythm templates and environmental noise profiles.

[0020] Figure 3 A flowchart for obtaining hearing threshold estimation results for each frequency band and corresponding uncertainty indices and behavioral event labels.

[0021] Figure 4 A flowchart for generating fixed-point correction and supplementary testing and automated screening reports. Detailed Implementation

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0025] Reference Figures 1-4 As one embodiment of the present invention, this embodiment provides an intelligent self-service pure-tone hearing threshold testing method, including the following steps: S1: Obtain the subject's personal operating rhythm template and environmental noise profile, and perform bitonal audiometry to obtain the noise recognition difficulty level, cooperation score, and pure tone threshold initiation strategy.

[0026] S1.1: Obtain the test subject's personal operational rhythm template and environmental noise profile, specifically: Personal Operation Rhythm Template: The visual button calibration task is implemented by presenting a visual stimulus graphic at a fixed brightness on the display interface. The visual stimulus graphic appears in the center of the display interface and remains displayed for a preset duration, such as 300 milliseconds. After the visual stimulus graphic disappears, it enters a preset interval, such as 800 milliseconds. The visual stimulus graphic is presented repeatedly a preset number of times, such as 10 times. The subject performs a button operation when the visual stimulus graphic appears. The button response time sequence is formed by recording the trigger time of each visual stimulus and the corresponding button trigger time. At the same time, it is determined whether the button trigger time is earlier than the visual stimulus trigger time to form the number of early button presses, and whether the button trigger time exceeds the preset response time limit to form the number of delayed button presses. The time difference between two adjacent button trigger times is recorded to form the continuous button press interval. A personal operation rhythm template is generated based on the button response time sequence, the number of early button presses, and the number of delayed button presses. The personal operation rhythm template includes the button response time distribution range, the statistical results of the number of early button presses, and the statistical results of the number of delayed button presses.

[0027] To further explain, the preset response time limit is set based on the statistical range of human visual reaction time, with an optimal value of 800 milliseconds; the 300-millisecond display duration is determined based on the shortest clear perception time range of human visual recognition; the 800-millisecond interval duration is determined based on the time range required for a normal visual reaction to complete; and the 10-repetition value is determined based on the minimum number of samples required to form a stable button response time distribution.

[0028] Environmental noise profiling: The environmental sound acquisition process is initiated within the same time frame as the visual button calibration task. The environmental sound acquisition process continuously acquires environmental sound signals within the same preset time window (10 seconds) as the visual button calibration task. The sound pressure amplitude of the environmental sound signals is extracted to obtain the environmental sound pressure level sequence. The frequency component analysis of the environmental sound signals is performed through fast Fourier transform to obtain the spectral distribution characteristics. The amplitude change of the environmental sound signals over time is statistically analyzed to obtain the fluctuation amplitude sequence. The environmental sound pressure level sequence, spectral distribution characteristics, and fluctuation amplitude sequence together constitute the environmental noise profile.

[0029] S1.2: Obtain the recognition difficulty level in noise, cooperation score, and pure tone thresholding starting strategy, specifically: When constructing the test sequence, a preset number (e.g., 20 groups) of two-part digital speech samples are selected under fixed background noise conditions to form a playlist. Each group of two-part digital speech is played once in the background noise. The preset playback intensity adjustment rule adopts a fixed step adjustment method. For example, the initial playback intensity is 60 dB sound pressure level. After each playback, the playback intensity is adjusted by a fixed amplitude based on the input result of the subject in the previous round, such as an adjustment amplitude of 2 dB sound pressure level. After playing each group of two-part digital speech, the two-digit result input by the subject and the corresponding input completion time are recorded to form a response time sequence. It is determined whether the subject's input result is completely consistent with the played two-part digital speech to count the number of correct recognitions and the number of recognition errors. The recognition difficulty level in the noise is determined based on the changes in the number of correct recognitions and the number of recognition errors during the gradual adjustment of the playback intensity. For example, if the number of recognition errors occurs continuously after the playback intensity is reduced to a certain sound pressure level, the corresponding playback intensity range is used as the basis for judging the recognition difficulty level in the noise. The response time sequence is compared with the key response time distribution range in the personal operation rhythm template. Responses falling within the key response time distribution range are counted as valid responses, responses exceeding the preset response time limit in the personal operation rhythm template are counted as timeout responses, and inputs inconsistent with the played two-part digital voice are counted as mis-inputs. A cooperation score is generated based on the ratio of valid responses, timeout responses, and mis-inputs: a high cooperation score is determined when the proportion of valid responses to total playbacks is 80% or more; a medium cooperation score is determined when the proportion of valid responses to total playbacks is 50% or more but less than 80%; and a low cooperation score is determined when the proportion of valid responses to total playbacks is less than 50%. At the same time, the timeout response and mis-inputs are combined to correct the medium and low cooperation scores. For example, when the proportion of mis-inputs to total playbacks is 30% or more, it is directly determined as a low cooperation score.

[0030] To further explain, the 80% and 50% ratios are determined based on empirical statistical ranges of effective response stability in auditory behavioral tests. Under normal attention, the proportion of effective responses is usually above 80%, so 80% is used as the standard for judging high cooperation, and 50% is used as the minimum stable response ratio limit. The 30% ratio of erroneous inputs to total playbacks is determined based on behavioral consistency control requirements. When the number of erroneous inputs reaches about one-third of the total playbacks, it reflects obvious input deviation, so it is used as a correction threshold for directly judging low cooperation.

[0031] The initial sound intensity level for pure-tone threshold measurement is determined based on the playback intensity range corresponding to the difficulty level of noise identification and the cooperation score. When the cooperation score is high, the initial sound intensity level is set at 5 dB above the upper limit of the playback intensity range corresponding to the difficulty level of noise identification. When the cooperation score is medium, the initial sound intensity level is set at 10 dB above the upper limit of the playback intensity range corresponding to the difficulty level of noise identification. When the cooperation score is low, the initial sound intensity level is set at 15 dB above the upper limit of the playback intensity range corresponding to the difficulty level of noise identification. The pressure level serves as the initial sound intensity level for pure tone thresholding. Simultaneously, the initial test frequency band is selected based on the difficulty level of noise identification, and a test stepping strategy is determined according to the cooperation score. A high cooperation score corresponds to a 5 dB stepping strategy, a medium cooperation score to a 10 dB stepping strategy, and a low cooperation score to a 10 dB stepping strategy, thus forming the initial pure tone thresholding strategy. The test stepping strategy is used to determine the fixed adjustment range of the pure tone stimulus sound intensity level. The fixed adjustment range corresponding to each test stepping strategy is defined as the test step unit, which is the minimum adjustment amount for each change in the pure tone stimulus sound intensity level.

[0032] S2: Based on the difficulty level of noise identification, cooperation score, and pure tone thresholding initiation strategy, execute the pure tone thresholding process based on the information gain selection mechanism, and combine the personal operation rhythm template to judge the subject's key press behavior in real time, output the hearing threshold estimation results of each frequency band and the corresponding uncertainty index and behavioral event markers.

[0033] S2.1: When establishing a candidate test set based on the pure tone threshold starting strategy, the pure tone threshold starting sound intensity level is used as the initial sound intensity level of each test frequency band. Multiple sound intensity levels are formed upward and downward around the pure tone threshold starting sound intensity level according to the test step strategy. The initial test frequency bands are preset standard frequency bands of 500 Hz, 1000 Hz, 2000 Hz, 4000 Hz and 8000 Hz. The candidate test set is composed of test frequency bands and corresponding sound intensity levels in pairs.

[0034] To further explain, the preset standard frequency bands of 500 Hz, 1000 Hz, 2000 Hz, 4000 Hz and 8000 Hz are determined based on the octave band test frequency system commonly used in clinical pure tone audiometry. They are used to cover the low-frequency, mid-frequency and high-frequency hearing range related to speech recognition. Among them, 500 Hz and 1000 Hz reflect low-mid frequency hearing ability, 2000 Hz and 4000 Hz reflect the frequency band related to speech intelligibility, and 8000 Hz reflects high-frequency hearing ability.

[0035] The ranking of the playback intensity range corresponding to the difficulty level of noise recognition is adjusted. When the upper limit of the playback intensity range corresponding to the difficulty level of noise recognition reaches or exceeds 65 dB SPL, the ranking of the 500 Hz and 1000 Hz test bands in the candidate test set is shifted to the back. The 65 dB SPL value is determined based on the boundary standard of moderate background noise environment in speech recognition test, with 65 dB SPL being the preferred value. It can be adaptively adjusted according to the background noise conditions in the actual test environment. The sound intensity level distribution range is determined according to the cooperation score. When the cooperation score is high, the sound intensity level is expanded upward and downward by one test step unit around the pure tone measurement threshold to form a sound intensity level set. When the cooperation score is medium, the sound intensity level is expanded upward and downward by two test step units to form a sound intensity level set. When the cooperation score is low, the sound intensity level is expanded upward by only two test step units to form a sound intensity level set.

[0036] Based on the environmental sound pressure level and spectral distribution characteristics in the environmental noise profile, the concentrated frequency band of environmental sound energy is determined. The spectral distribution characteristics are segmented and statistically analyzed according to the octave bands corresponding to the preset standard frequency bands of 500 Hz, 1000 Hz, 2000 Hz, 4000 Hz, and 8000 Hz. When the environmental sound pressure level in a certain octave band reaches or exceeds the preset environmental interference judgment value, the corresponding octave band is determined as the concentrated frequency band of environmental sound energy. The preset environmental interference judgment value is determined based on the background noise limit recommended by the quiet hearing threshold test, with a preferred value of 45 dB sound pressure level. The center frequency of the concentrated frequency band of environmental sound energy is... When the center frequency of the test band matches that in the candidate test set, the corresponding test band is shifted to the next position in the candidate test set, thus forming a candidate test set arranged according to test priority. An initial uncertainty index is assigned to each test band. The uncertainty index is determined based on the difference between the initial sound intensity level of the pure tone threshold measurement and the preset normal hearing threshold reference range. The initial sound intensity level of the pure tone threshold measurement is expressed in decibel hearing level, and the preset normal hearing threshold reference range is expressed in hearing level from 0 to 25 decibels. The decibel sound pressure level is only used for environmental noise determination and is not involved in the calculation of the numerical difference between the hearing threshold estimation result and the uncertainty index.

[0037] To further clarify, the preset normal hearing threshold reference range is determined based on the adult pure-tone hearing threshold normal hearing judgment standard, and the hearing level of 0 to 25 dB is taken as the normal hearing threshold reference range according to the clinical pure-tone hearing classification standard.

[0038] Before each round of pure-tone stimulation, the current hearing threshold estimation results and corresponding uncertainty indices for each frequency band are read. The current hearing threshold estimation results for each frequency band refer to the initial sound intensity level assigned to each test frequency band based on the initial sound intensity level of the pure-tone threshold measurement, and the corresponding sound intensity level values ​​for each test frequency band are gradually corrected based on the key response results after each pure-tone stimulation. For each test frequency band and corresponding sound intensity level in the candidate test set, the difference between the current sound intensity level and the current hearing threshold estimation results for the corresponding test frequency band is calculated. When the difference is equal to one test step unit, the corresponding test frequency band is marked as a priority evaluation object. When the value is not equal to a test step unit, the corresponding test frequency band is not marked as a priority evaluation object, and the original sorting position of the corresponding test frequency band in the candidate test set is maintained. All priority evaluation objects are sorted from largest to smallest according to the corresponding uncertainty index value, and the test frequency band with the largest uncertainty index value and the corresponding sound intensity level are selected as the next pure tone stimulus. In this way, the test path is dynamically adjusted as the hearing threshold estimation result changes. Compared with the fixed-order frequency-by-frequency test method, the pure tone stimulus order of the present invention is no longer arranged by the traditional preset frequency band method, but is adaptively selected around the change of uncertainty index.

[0039] S2.2: After playing the pure tone stimulus of the selected test frequency band and corresponding sound intensity level, record the subject's key press response results. Compare the key press response time with the key press response time distribution range in the personal operation rhythm template. Key press responses that fall within the key press response time distribution range are judged as valid responses, and key press responses that exceed the preset response time limit in the personal operation rhythm template are judged as invalid responses. Generate behavioral event markers based on valid or invalid responses. Behavioral event markers are used to distinguish between real auditory perception behavior and non-auditory triggered behavior.

[0040] Based on the key response results and behavioral event markers, the hearing threshold estimation results and corresponding uncertainty indices for the corresponding test frequency bands are updated. When two consecutive valid responses occur at the same sound intensity level, the hearing threshold estimation results are converged to the current sound intensity level. When a valid response occurs and the current sound intensity level is higher than the current hearing threshold estimation results for each frequency band, the sound intensity level of the corresponding test frequency band in the next round of candidate test sets is adjusted down by one test step unit. When an invalid response occurs, the sound intensity level of the corresponding test frequency band in the next round of candidate test sets is adjusted up by one test step unit, and a new uncertainty index range corresponding to one test step unit is added to the current uncertainty index value of the corresponding test frequency band. The ranking position in the candidate test set is updated simultaneously, so that unstable test frequency bands are given priority to enter the next round of pure tone stimulus selection process.

[0041] When the uncertainty index of all test frequency bands reaches the preset convergence condition, the pure tone threshold measurement process ends and the hearing threshold estimation results of each frequency band, as well as the corresponding uncertainty index and behavioral event markers, are output. The dynamic test path driven by uncertainty index can reduce the number of repeated ineffective stimuli compared with the traditional fixed step method, and keep the hearing threshold estimation results stable even when the subject's cooperation fluctuates.

[0042] To further explain, the preset convergence condition is determined based on the smallest resolvable sound intensity change unit in clinical pure tone hearing threshold determination. Specifically, when the sound intensity level fluctuation range represented by the uncertainty index corresponding to each test frequency band does not exceed one test step unit, the corresponding test frequency band is determined to have reached a stable hearing threshold state, thus serving as the convergence condition for ending the pure tone threshold measurement process.

[0043] Preferably, compared with existing pure-tone audiometry methods that use fixed order and fixed step method, the present invention constructs a dynamic candidate test set by combining the difficulty level of noise identification, cooperation score and environmental noise profile, and drives the pure-tone stimulus selection path with uncertainty index to achieve adaptive adjustment of test frequency band and sound intensity level; at the same time, it improves the ability to identify invalid responses by using personal operation rhythm template to judge key behavior in real time, thereby improving the ability to steadily converge hearing threshold estimation results under complex environment and cooperation fluctuation conditions, and improving test efficiency and result reliability.

[0044] S3: Based on the hearing threshold estimation results of each frequency band, uncertainty indicators and behavioral event markers, perform fixed-point correction and supplementary testing and weight adjustment to obtain the final hearing threshold result and overall credibility level.

[0045] S3.1: When verifying the consistency between the uncertainty index and behavioral event markers corresponding to the hearing threshold estimation results of each frequency band, the hearing threshold estimation results corresponding to 500 Hz, 1000 Hz, 2000 Hz, 4000 Hz and 8000 Hz are read one by one and arranged in order of frequency from low to high to form a hearing threshold change sequence. At the same time, the corresponding uncertainty index values ​​and the number of behavioral event markers are read. The difference between the hearing threshold estimation results of adjacent frequency bands is correlated and compared with the corresponding uncertainty index values. When the difference between the hearing threshold estimation results of adjacent frequency bands exceeds two test step units, and the corresponding uncertainty index value is in a non-converged state or there are invalid response times in the behavioral event markers, the corresponding frequency band is marked as a candidate frequency band that needs further verification. Among the candidate frequency bands, the test frequency bands whose uncertainty index values ​​exceed the preset convergence conditions or whose invalid response times in the behavioral event markers reach the trigger conditions are selected as the target test frequency bands.

[0046] To further explain, the triggering condition for the number of invalid responses in the behavior event marker is determined based on the behavior consistency control principle. It is used to determine whether there is repetitive invalid feedback in the test process. The preferred value is to trigger when there are two consecutive invalid responses or when there are a total of two invalid responses in the same test frequency band.

[0047] When re-establishing a candidate test set for the target test frequency band, the current hearing threshold estimation result of the target test frequency band is used as the central sound intensity level. The set of supplementary sound intensity levels is formed by expanding upwards and downwards by one test step unit. Pure tone stimuli are played, and the button response results and corresponding behavioral event markers during the supplementary testing phase are recorded. The hearing threshold estimation result of the target test frequency band is corrected based on two consecutive valid or invalid responses during the supplementary testing phase. When there are two consecutive valid responses during the supplementary testing phase, the hearing threshold estimation result of the target test frequency band converges to the current supplementary sound intensity level. When there are two consecutive invalid responses during the supplementary testing phase, the hearing threshold estimation result of the target test frequency band is adjusted upwards by one test step unit, and the corresponding uncertainty index is updated, causing the uncertainty index of the target test frequency band to re-enter the convergence judgment process.

[0048] S3.2: When performing weight correction processing on the updated hearing threshold estimation results for each frequency band, the frequency band weight is determined based on the uncertainty index value and the number of invalid responses in the behavioral event markers for each frequency band. Frequency bands with convergent uncertainty index values ​​and zero invalid responses are assigned a standard weight (i.e., 1). When the uncertainty index value of a certain test frequency band has not converged or there are invalid responses, the weight of the frequency band is reduced. The frequency band weight of the current test frequency band is calculated by attenuating according to the uncertainty index value and the number of invalid responses, as expressed in the following expression: ; in, Indicates the first The frequency band weights of each test frequency band Indicates the first The uncertainty index values ​​corresponding to each test frequency band. This indicates a preset convergence condition. Indicates the first The number of invalid responses corresponding to each test frequency band. Greater than or equal to zero Greater than zero, The value must be greater than or equal to zero to ensure that the frequency band weight is positive and the denominator of the expression is not zero.

[0049] The hearing threshold estimation results for each frequency band are weighted and corrected according to the corresponding frequency band weights to obtain the final hearing threshold results for each frequency band. The expression for the final hearing threshold results for each frequency band is as follows: ; in, Indicates the first The final hearing threshold results for each test frequency band, Indicates the first Hearing threshold estimation results for each test frequency band.

[0050] The final hearing threshold results for each frequency band together constitute the final hearing threshold result set.

[0051] A comprehensive evaluation is conducted based on the hearing threshold estimation results for each frequency band, the corresponding uncertainty indicators, and behavioral event markers. The number of frequency bands in a convergent state and the number of frequency bands with invalid responses are counted. When all test frequency bands are in a convergent state and the number of invalid responses for all test frequency bands is zero, a Level 1 overall credibility level is generated. When one test frequency band is not in a convergent state or when one test frequency band has a greater than zero number of invalid responses, a Level 2 overall credibility level is generated. When two or more test frequency bands are not in a convergent state or when two or more test frequency bands have a greater than zero number of invalid responses, a Level 3 overall credibility level is generated, thus completing the output of the final hearing threshold result and the overall credibility level.

[0052] S4: Based on the final hearing threshold results and overall reliability level, combined with the level of difficulty in identifying noise, determine the hearing screening results and generate an automated screening report and referral recommendations.

[0053] When extracting the final hearing threshold results, the final hearing threshold result values ​​for each frequency band corresponding to 500 Hz, 1000 Hz, 2000 Hz, 4000 Hz and 8000 Hz are read. At the same time, the overall confidence level and the level of difficulty in identifying noise are read and written into the screening result data set. The screening result data set includes the frequency band identifier, the final hearing threshold result value for each frequency band, the overall confidence level value and the level of difficulty in identifying noise identifier.

[0054] The final hearing threshold values ​​for each frequency band are compared one by one with the grading intervals in the preset hearing grading standard. According to the clinical pure-tone audiometry grading standard, hearing levels of 0 to 25 dB are classified as normal hearing, 26 to 40 dB as mild hearing loss, 41 to 55 dB as moderate hearing loss, 56 to 70 dB as moderate to severe hearing loss, 71 to 90 dB as severe hearing loss, and greater than 90 dB as profound hearing loss. The overall hearing grading result is determined based on the highest grade result in each frequency band. A result validity label is generated based on the overall credibility level. When the overall credibility level is Level 1, a "Result Valid" label is generated; when the overall credibility level is Level 2, a "Retest Recommended" label is generated; and when the overall credibility level is Level 3, a "Result Should Be Used with Caution" label is generated.

[0055] The system reads a pre-defined screening rule table. The screening rule table uses the hearing classification result as the first criterion, the overall credibility level as the second criterion, and the difficulty level of identification in noise as the third criterion. When the hearing classification result is normal hearing level and the overall credibility level is Level 1, a referral recommendation level of "no referral required" is generated. When the hearing classification result is mild hearing loss level and the overall credibility level is Level 1, a referral recommendation level of "recommendation for regular follow-up" is generated. When the hearing classification result is moderate or above hearing loss level, a referral recommendation level of "recommendation for specialist evaluation" is generated. When the overall credibility level is Level 3, a referral recommendation level of "recommendation for retesting in a quiet environment" is uniformly generated.

[0056] The final hearing threshold results, overall confidence level, noise identification difficulty level, hearing classification results, and referral recommendation level are uniformly packaged and structured text content is generated according to the preset report format. The report content includes a table of final hearing threshold results corresponding to the tested frequency band, a description of the hearing classification results, a description of the overall confidence level, and a description of the referral recommendation level. The above content is written into the automated screening report and referral recommendation text, thereby completing the generation of the automated screening report and referral recommendation.

[0057] This embodiment also provides a computer device applicable to the intelligent self-service pure tone hearing threshold testing method, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the intelligent self-service pure tone hearing threshold testing method proposed in the above embodiment.

[0058] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0059] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the intelligent self-service pure-tone hearing threshold testing method proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0060] In summary, this invention achieves the following: By establishing a candidate test set and introducing a pure-tone thresholding process based on an information gain selection mechanism, the test path is dynamically adjusted according to the hearing threshold estimation results and uncertainty indicators of each frequency band. This ensures that pure-tone stimuli preferentially act on the test points with the highest information contribution, avoiding repetitive stimulation and invalid tests caused by the fixed step method, thereby improving test efficiency and result stability. Simultaneously, by constructing a personal operation rhythm template and judging key press behavior in real time, combined with the output of uncertainty indicators and behavioral event markers, the invention distinguishes between false triggering behaviors and real auditory responses, and establishes a quantitative expression mechanism for the stability of hearing threshold estimation. Ultimately, the test results not only have numerical output but also the ability to evaluate credibility.

[0061] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A smart self-service pure-tone hearing threshold testing method, characterized in that: include, The test subject's personal operating rhythm template and environmental noise profile were obtained, and a two-tone audiometry was performed to obtain the noise recognition difficulty level, cooperation score, and pure tone threshold initiation strategy. Based on the noise identification difficulty level, cooperation score, and pure tone thresholding initiation strategy, a pure tone thresholding process based on information gain selection mechanism is executed. Combined with the personal operation rhythm template, the key press behavior of the test subject is judged in real time, and the hearing threshold estimation results of each frequency band and the corresponding uncertainty index and behavioral event markers are output. Based on the hearing threshold estimation results of each frequency band, uncertainty index and behavioral event markers, targeted correction and supplementary testing and weight adjustment are performed to obtain the final hearing threshold result and overall credibility level. The hearing screening results are determined based on the final hearing threshold and overall reliability level, combined with the level of difficulty in identifying noise, and an automated screening report and referral recommendations are generated.

2. The intelligent self-service pure-tone hearing threshold testing method as described in claim 1, characterized in that: The personal operation rhythm template is constructed by performing a visual key calibration task, recording the key response time sequence, number of early key presses, number of delayed key presses, and consecutive key press intervals after multiple visual stimulus triggers, and based on the key response time sequence, number of early responses, and number of delayed responses. The personal operation rhythm template includes the key response time distribution range, the statistical results of the number of early key presses, and the statistical results of the number of delayed key presses.

3. The intelligent self-service pure-tone hearing threshold testing method as described in claim 1, characterized in that: The environmental noise profile is constructed by starting the environmental sound acquisition process while building a personal operation rhythm template, acquiring environmental sound signals within a preset time window and extracting environmental sound pressure level, spectral distribution characteristics and fluctuation amplitude sequence.

4. The intelligent self-service pure-tone hearing threshold testing method as described in claim 1, characterized in that: The obtained noise recognition difficulty level, cooperation score, and pure tone thresholding starting strategy are as follows: Construct a test sequence, play two-part digital speech one after another according to the preset playback intensity adjustment rules, and record the corresponding input results and response time sequence of the test subject for each playback; The number of correct and incorrect recognitions is counted based on the test subject's input, and the recognition difficulty level in the noise is generated by combining the playback intensity adjustment process. Based on the matching analysis between the response time series and the key response time distribution range in the personal operation rhythm template, the number of valid responses, the number of timeout responses and the number of mis-inputs are counted to generate a cooperation score. Based on the mapping between the difficulty level of noise identification and the cooperation score, the initial sound intensity level, initial test frequency band, and test step strategy for pure tone thresholding are determined, thus forming the initial strategy for pure tone thresholding.

5. The intelligent self-service pure-tone hearing threshold testing method as described in claim 1, characterized in that: The pure tone thresholding process based on the information gain selection mechanism is as follows: The initial test frequency band in the pure tone thresholding starting strategy is used as the test frequency band in the candidate test set. Multiple sound intensity levels are formed by expanding upward and downward according to the test stepping strategy, with the pure tone thresholding starting sound intensity level as the center. The test frequency bands and corresponding sound intensity levels are combined to form the candidate test set. The test frequency bands in the candidate test set are screened and sorted by combining the identification difficulty level in the noise, the cooperation score, and the environmental noise profile. The uncertainty index of each test frequency band is initialized. The candidate test set includes multiple test frequency bands and corresponding sound intensity levels. Before each round of pure tone stimulation, the degree of reduction of the overall uncertainty index by each test frequency band and sound intensity level is calculated based on the current hearing threshold estimation results and uncertainty index of each frequency band. The test frequency band and sound intensity level with the greatest degree of uncertainty reduction are selected as the next pure tone stimulation. Play the selected pure tone stimulus and record the subject's key press response. Combine the personal operation rhythm template to determine the validity of the key press response and generate the corresponding behavioral event tag. Based on the key response results and behavioral event markers, update the hearing threshold estimation results and uncertainty index of the corresponding test frequency band, and update the candidate test set simultaneously until the uncertainty index of all test frequency bands meets the preset convergence condition, and output the hearing threshold estimation results of each frequency band and the corresponding uncertainty index and behavioral event markers.

6. The intelligent self-service pure-tone hearing threshold testing method as described in claim 1, characterized in that: The final hearing threshold result and overall reliability level are obtained as follows: Consistency verification is performed on the uncertainty index and behavioral event markers corresponding to the hearing threshold estimation results of each frequency band. Based on the consistency verification results, preset convergence conditions and behavioral event marker triggering conditions, the target test frequency bands that need to be subjected to fixed-point correction and supplementary testing are selected. A new candidate test set was established for the target test frequency band, and pure tone stimulation was played. The key response results and corresponding behavioral event tags during the supplementary test phase were recorded, and the hearing threshold estimation results and uncertainty index of the target test frequency band were updated. The updated hearing threshold estimation results for each frequency band are processed by weight correction. Frequency band weights are assigned according to the uncertainty index and behavioral event markers corresponding to each frequency band, and the final hearing threshold results are calculated. The overall credibility level is generated by comprehensively evaluating the hearing threshold estimation results of each frequency band, the corresponding uncertainty indicators, and behavioral event markers.

7. The intelligent self-service pure-tone hearing threshold testing method as described in claim 1, characterized in that: The generation of automated screening reports and referral recommendations specifically includes: Extract the final hearing threshold results, overall confidence level, and identification difficulty level in noise to construct a screening result dataset; The final hearing threshold results are compared with the preset hearing grading standards to determine the level, and the validity of the results is generated by combining the overall credibility level. The referral recommendation level is generated by matching the final hearing threshold result with the hearing classification result, the overall confidence level, and the level of difficulty in identifying noise against the preset screening rule table. The final hearing threshold results, overall reliability level, difficulty level of identification in noise, hearing classification results, and referral recommendation level are uniformly packaged to generate automated screening reports and referral recommendations.

8. The intelligent self-service pure-tone hearing threshold testing method as described in claim 6, characterized in that: The consistency verification refers to the correlation and comparison of the changing trends of the hearing threshold estimation results of each frequency band, as well as the corresponding uncertainty indicators and behavioral event markers.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the intelligent self-service pure tone hearing threshold testing method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the intelligent self-service pure tone hearing threshold testing method according to any one of claims 1 to 8.

Images