Hearing loss prediction method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention discloses a hearing loss prediction method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a first data set and a plurality of initial kurtosis equivalent sound level models, determining a kurtosis adjustment equivalent sound level group and an estimated permanent hearing threshold displacement group corresponding to each initial kurtosis equivalent sound level model, determining a steady-state high-frequency noise hearing loss ratio corresponding to each steady-state noise equivalent sound level and an estimated high-frequency noise hearing loss ratio corresponding to each kurtosis adjustment equivalent sound level, and determining a target kurtosis equivalent sound level model based on a preset model screening method, and adjusting the equivalent sound level of the test object by using the target kurtosis equivalent sound level model. Therefore, the problem that a traditional equivalent noise index is underestimated to impact and intermittent noise hazards is solved, and the accuracy of hearing loss prediction is improved.

Description

Technical Field

[0001] The present disclosure relates to a hearing displacement model optimization technology, and in particular to a hearing loss prediction method, device, electronic device and storage medium. Background Art

[0002] The current occupational noise exposure assessment system generally uses the 8-hour equivalent A sound level (L EX,8h ) as the core evaluation indicator. The technical principle of this indicator is based on the time-averaged characteristics of noise energy, and uses the A-weighted network to simulate the human ear's auditory response to evaluate hearing loss. International standards (such as "ISO1999:2013Acoustics.Estimation of noise-induced hearing loss") and my country's occupational health standards (such as "GBZ / T 229.4-2012 Occupational Disease Hazards in the Workplace Part 4: Noise") both use L EX,8h To construct a dose-response relationship model between noise exposure dose and hearing loss risk. EX,8h Estimating hearing loss has the advantages of convenient measurement and standardized calculation. However, in working environments such as stamping, forging, and impact tools, noise signals often have significant time domain fluctuation characteristics. Related studies have shown that under the same L EX,8h At the same level, high-peak noise (such as impulse noise) is more likely to cause hearing loss than steady-state noise. EX,8h The kurtosis of the noise is not taken into account, which makes the L EX,8h The estimated hearing loss is inaccurate. Experimental data show that the EX,8h The estimated predicted hearing loss value was significantly lower than the actual clinically detected hearing loss. Summary of the Invention

[0003] In order to solve the above problems, the embodiments of the present disclosure provide a hearing loss prediction method, device, electronic device and storage medium.

[0004] In one aspect of an embodiment of the present disclosure, a method for predicting hearing loss is provided, comprising: obtaining a first data set and a plurality of initial kurtosis equivalent sound level models, wherein the first data set includes a plurality of training data, each training data including a test equivalent sound level, a test noise exposure kurtosis, a test permanent hearing threshold shift, a noise exposure duration, a steady-state noise equivalent sound level, and a steady-state permanent hearing threshold shift, each initial kurtosis equivalent sound level model is used to represent a correspondence between an equivalent sound level and a kurtosis-adjusted equivalent sound level; based on the test equivalent sound level, the test noise exposure kurtosis, the test permanent hearing threshold shift, the noise exposure duration, the steady-state noise equivalent sound level, and the steady-state permanent hearing threshold shift, each initial kurtosis equivalent sound level model is used to represent a correspondence between an equivalent sound level and a kurtosis-adjusted equivalent sound level; The kurtosis of sound exposure and the duration of noise exposure are used to determine the kurtosis-adjusted equivalent sound level group and the estimated permanent hearing threshold shift group corresponding to each initial kurtosis equivalent sound level model, the kurtosis-adjusted equivalent sound level group includes multiple kurtosis-adjusted equivalent sound levels, and the estimated permanent hearing threshold shift group includes multiple estimated permanent hearing threshold shifts; determine the proportion of steady-state high-frequency noise hearing loss corresponding to each steady-state noise equivalent sound level, and the estimated proportion of high-frequency noise hearing loss corresponding to each kurtosis-adjusted equivalent sound level; based on the preset model screening method, determine from the multiple initial kurtosis equivalent sound level models. Determining a target kurtosis equivalent sound level model, the preset model screening method includes: determining the underestimation and improvement degree corresponding to each initial kurtosis adjustment model based on the estimated permanent hearing threshold shift groups corresponding to each initial kurtosis adjustment model, as well as the test permanent hearing threshold shift and steady-state permanent hearing threshold shift based on each training data; performing logistic regression curve fitting on each steady-state noise equivalent sound level and the steady-state high-frequency noise hearing loss percentage corresponding to each steady-state noise equivalent sound level to obtain a steady-state curve, and performing logistic regression curve fitting on each kurtosis-adjusted equivalent sound level and the estimated high-frequency noise hearing loss percentage corresponding to each kurtosis-adjusted equivalent sound level to obtain a logistic regression curve corresponding to each initial kurtosis adjustment model; determining a target kurtosis equivalent sound level model based on the logistic regression curve, steady-state curve, improvement degree, and underestimation corresponding to each initial kurtosis adjustment model; and adjusting the equivalent sound level of the test subject using the target kurtosis equivalent sound level model to obtain the kurtosis-adjusted equivalent sound level of the test subject, so as to assess the hearing loss of the test subject based on the kurtosis-adjusted equivalent sound level.

[0005] Another aspect of the embodiments of the present disclosure provides a hearing loss prediction device, including a first data acquisition module for acquiring a first data set and multiple initial kurtosis equivalent sound level models, wherein the first data set includes multiple training data, each training data includes a test equivalent sound level, a test noise exposure kurtosis, a test permanent hearing threshold shift, a noise exposure duration, a steady-state noise equivalent sound level, and a steady-state permanent hearing threshold shift, and each initial kurtosis equivalent sound level model is used to represent a correspondence between an equivalent sound level and a kurtosis-adjusted equivalent sound level; a first data determination module for determining a first data set based on the test data in each training data. The test equivalent sound level, the test noise exposure kurtosis and the noise exposure duration are used to determine the kurtosis-adjusted equivalent sound level group and the estimated permanent hearing threshold shift group corresponding to each initial kurtosis equivalent sound level model, the kurtosis-adjusted equivalent sound level group includes multiple kurtosis-adjusted equivalent sound levels, and the estimated permanent hearing threshold shift group includes multiple estimated permanent hearing threshold shifts; the second data determination module is used to determine the steady-state high-frequency noise hearing loss ratio corresponding to each steady-state noise equivalent sound level, and the estimated high-frequency noise hearing loss ratio corresponding to each kurtosis-adjusted equivalent sound level; the model screening module is used to select the model based on the preset model screening method. The target kurtosis equivalent sound level model is determined from the multiple initial kurtosis equivalent sound level models, and the preset model screening method includes: determining the underestimation and improvement degree corresponding to each initial kurtosis adjustment model based on the estimated permanent hearing threshold shift group corresponding to each initial kurtosis adjustment model, and the test permanent hearing threshold shift and steady-state permanent hearing threshold shift based on each training data; performing logistic regression curve fitting on each steady-state noise equivalent sound level and the steady-state high-frequency noise hearing loss ratio corresponding to each steady-state noise equivalent sound level to obtain a steady-state curve, and performing logistic regression curve fitting on each kurtosis adjusted equivalent sound level and each Logistic regression curve fitting is performed on the estimated high-frequency noise hearing loss proportion values corresponding to the kurtosis-adjusted equivalent sound levels to obtain the logistic regression curves corresponding to the initial kurtosis-adjusted models; based on the logistic regression curves, steady-state curves, improvement levels and underestimation values corresponding to the initial kurtosis-adjusted models, a target kurtosis equivalent sound level model is determined; an equivalent sound level correction module is used to adjust the equivalent sound level of the test subject using the target kurtosis equivalent sound level model to obtain the kurtosis-adjusted equivalent sound level of the test subject, so as to evaluate the hearing loss of the test subject based on the kurtosis-adjusted equivalent sound level.

[0006] Another aspect of the embodiments of the present disclosure provides an electronic device, including: a memory for storing a computer program; a processor for executing the computer program stored in the memory, and when the computer program is executed, the above-mentioned hearing loss prediction method is implemented.

[0007] In another aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned hearing loss prediction method is implemented.

[0008] In the embodiment of the present disclosure, by setting a plurality of initial kurtosis equivalent sound level models representing the correspondence between the equivalent sound level and the kurtosis-adjusted equivalent sound level, the steady-state noise equivalent sound level, the steady-state permanent hearing threshold shift and the steady-state high-frequency noise hearing loss ratio of each training data are used to determine the target kurtosis equivalent sound level model by means of logistic regression. Thus, the noise exposure kurtosis is added to the equivalent sound level through the target kurtosis equivalent sound level model, thereby realizing the introduction of noise exposure kurtosis as a quantitative indicator of noise time domain characteristics in the equivalent sound level, overcoming the problem of traditional equivalent noise indicators underestimating the hazards of impact and intermittent noise, and improving the accuracy of hearing loss prediction.

[0009] The technical solution of the present disclosure is further described in detail below through the accompanying drawings and examples. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0011] The present disclosure can be more clearly understood from the following detailed description with reference to the accompanying drawings, in which:

[0012] Figure 1 It is a flowchart of a hearing loss prediction method provided by an exemplary embodiment of the present disclosure.

[0013] Figure 2 4 is a flowchart of a hearing loss prediction method provided by another exemplary embodiment of the present disclosure.

[0014] Figure 3 3 is a flowchart of a hearing loss prediction method provided by another exemplary embodiment of the present disclosure.

[0015] Figure 4 3 is a schematic diagram of a logistic regression curve between the proportion of high-frequency noise hearing loss and the equivalent sound level provided by an exemplary embodiment of the present disclosure.

[0016] Figure 5 2 is a schematic structural diagram of a hearing loss prediction device provided by an exemplary embodiment of the present disclosure.

[0017] Figure 6 The figure is a schematic structural diagram of an application embodiment of the electronic device disclosed herein. DETAILED DESCRIPTION

[0018] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that unless otherwise specifically stated, the relative arrangement of components and steps, numerical expressions and numerical values set forth in these embodiments do not limit the scope of the present disclosure.

[0019] Those skilled in the art will understand that the terms "first" and "second" in the embodiments of the present disclosure are only used to distinguish different steps, devices or modules, and do not represent any specific technical meanings, nor do they indicate a necessary logical order between them.

[0020] It should also be understood that in the embodiments of the present disclosure, “a plurality of” may refer to two or more than two, and “at least one” may refer to one, two, or more than two.

[0021] It should also be understood that any component, data or structure mentioned in the embodiments of the present disclosure can generally be understood as one or more, unless explicitly limited or otherwise indicated in the context.

[0022] In addition, the term "and / or" in this disclosure is merely a description of the association relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this disclosure generally indicates that the related objects are in an "or" relationship.

[0023] It should also be understood that the description of the various embodiments in this disclosure focuses on the differences between the various embodiments, and the same or similar aspects thereof can be referenced with each other. For the sake of brevity, they will not be described one by one.

[0024] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0025] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.

[0026] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0027] It should be noted that like reference numerals and letters refer to like items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0028] The embodiments of the present disclosure can be applied to electronic devices such as terminal devices, computer systems, and servers, and can operate in conjunction with numerous other general-purpose or specialized computing system environments or configurations. Examples of well-known terminal devices, computing systems, environments, and / or configurations suitable for use with terminal devices, computer systems, servers, and other electronic devices include, but are not limited to, personal computer systems, server computer systems, thin clients, thick clients, handheld or laptop devices, microprocessor-based systems, set-top boxes, programmable consumer electronics, network personal computers, minicomputer systems, mainframe computer systems, and distributed cloud computing technology environments including any of the above systems, among others.

[0029] Electronic devices such as terminal devices, computer systems, and servers can be described in the general context of computer system-executable instructions (such as program modules) executed by a computer system. Generally, program modules can include routines, programs, object programs, components, logic, data structures, etc., which perform specific tasks or implement specific abstract data types. Computer systems / servers can be implemented in a distributed cloud computing environment, where tasks are performed by remote processing devices linked via a communication network. In a distributed cloud computing environment, program modules can be located on local or remote computing system storage media, including storage devices.

[0030] Figure 1 FIG. 1 is a flow chart of a hearing loss prediction method provided by an exemplary embodiment of the present disclosure. This embodiment can be applied to electronic devices, such as Figure 1 As shown, the hearing loss prediction method may include the following steps:

[0031] Step S100: Acquire a first data set and a plurality of initial kurtosis equivalent sound level models.

[0032] Among them, each initial kurtosis equivalent sound level model is used to express the corresponding relationship between the equivalent sound level and the kurtosis adjusted equivalent sound level. The initial kurtosis equivalent sound level model can be implemented as shown in the following formula (1):

[0033]

[0034] In formula (1), L EX，8h-K is the kurtosis-adjusted equivalent sound level, L EX，8h is the equivalent sound level, β N To test the noise exposure kurtosis, β G is the preset steady-state noise exposure kurtosis, and λ is the adjustment coefficient. EX，8h -K and L EX，8h The unit is dB(A), β N and β G Unitless, λ is a constant.

[0035] The β in formula (1) can be G and λ are set to different values to obtain multiple initial kurtosis equivalent sound level models. For example, β G It can be 3, 6.6 and 10, and λ can be 3, 4.4, 6.3 and 7.9.

[0036] Each piece of training data includes the test equivalent sound level, test noise exposure kurtosis, test permanent hearing threshold shift, noise exposure duration, steady-state noise equivalent sound level, steady-state permanent hearing threshold shift, and each piece of training data can also include the reference sound pressure level and the frequency domain of the noise in the test noise environment. The reference sound pressure level is the benchmark value used to calculate the sound pressure level (SPL) in acoustics. It is a relative standard for measuring sound intensity and ensures that sound pressure measurements in different scenarios are comparable. The frequency domain refers to the dimension in which the acoustic energy of noise is decomposed and analyzed according to its frequency components.

[0037] The noise exposure duration is the duration of time in the test noise environment, and the unit of noise exposure duration is year. The test noise environment can be the working environment of the sample object, and the sample object can be the person participating in the hearing loss test. The equivalent sound level can be L EX,8h , L EX,8h The average noise energy level that workers are exposed to during an 8-hour workday can be quantified, while taking into account the simulation of the human ear hearing characteristics by the A-weighting network. EX,8h It can be used to represent the equivalent of the actual noise energy exposure to a steady-state A-weighted noise level lasting 8 hours. The test equivalent sound level is the equivalent sound level collected in the test noise environment, and the steady-state noise equivalent sound level is the equivalent sound level collected in the steady-state noise environment. A steady-state noise environment refers to an environment with continuous noise in which the sound pressure level fluctuation range is ≤5dB(A) within a certain period of time (usually ≥1 second). The noise exposure kurtosis can describe the statistical parameters of the impulse response of noise and is used to measure the time domain structural characteristics of noise. The test noise exposure kurtosis is the noise exposure kurtosis corresponding to the test noise environment. The permanent hearing threshold shift (Noise-Induced Permanent Threshold Shift, NIPTS) is used to indicate the hearing threshold increase caused by noise that cannot be restored to normal levels. For each training data, based on the test noise exposure kurtosis, noise exposure duration, frequency domain and reference sound pressure level in the training data, the test permanent hearing threshold shift is determined using formula (2). Steady-state permanent threshold shift can be measured using the Noise-Induced Permanent Threshold Shift at 3, 4, and 6kHz frequencies (NIPTS). 346 ).

[0038]

[0039] In formula (2), NIPTS 50 To test for permanent hearing threshold shift, L EX,8h is the test equivalent sound level, t is the noise exposure duration, t0 is 1, L0 is the reference sound pressure level, and u and v are coefficients. Different reference sound pressure levels and frequency domains correspond to different coefficients. The correspondence between reference sound pressure levels, frequency domains, and coefficients is detailed in Table 1. For each training data point, coefficients u and v are selected from Table 1 based on the frequency domain and reference sound pressure level in that training data point. The test permanent hearing threshold shift for that training data point is calculated using these coefficients.

[0040] Table 1

[0041]

[0042] Step S110 , based on the test equivalent sound level, test noise exposure kurtosis and noise exposure duration in each training data, determining the kurtosis-adjusted equivalent sound level group and the estimated permanent hearing threshold shift group corresponding to each initial kurtosis equivalent sound level model.

[0043] The kurtosis-adjusted equivalent sound level group includes a plurality of kurtosis-adjusted equivalent sound levels, and the estimated permanent hearing threshold shift group includes a plurality of estimated permanent hearing threshold shifts. Specifically, the kurtosis-adjusted equivalent sound level group corresponding to each initial kurtosis equivalent sound level model includes a plurality of kurtosis-adjusted equivalent sound levels calculated using the initial kurtosis equivalent sound level model, and the estimated permanent hearing threshold shift group corresponding to each initial kurtosis equivalent sound level model includes a plurality of estimated permanent hearing threshold shifts, wherein the plurality of estimated permanent hearing threshold shifts are obtained using the plurality of kurtosis-adjusted equivalent sound levels corresponding to the initial kurtosis equivalent sound level model.

[0044] In one embodiment, for each initial kurtosis equivalent sound level model, the test equivalent sound level in each training data is calculated in turn using the initial kurtosis equivalent sound level model to obtain multiple kurtosis-adjusted equivalent sound levels corresponding to the initial kurtosis equivalent sound level model.

[0045] The permanent hearing threshold shift prediction model is used to calculate the estimated permanent hearing threshold shift. The preset permanent hearing threshold shift prediction model can be obtained based on formula (2), and the preset permanent hearing threshold shift prediction model can be obtained by replacing the value in formula (2) with the value in formula (3).

[0046]

[0047] In formula (3), NIPTS′ 50To estimate the permanent hearing threshold shift, the multiple estimated permanent hearing threshold shifts determined from the multiple kurtosis-adjusted equivalent sound levels calculated using the initial kurtosis equivalent sound level model are used as the multiple estimated permanent hearing threshold shifts corresponding to the initial kurtosis equivalent sound level model. For each kurtosis-adjusted equivalent sound level, the coefficients u and v are selected from Table 1 based on the frequency domain and reference sound pressure level in the training data corresponding to the test equivalent sound level corresponding to the kurtosis-adjusted equivalent sound level.

[0048] Step S120 , determining the steady-state high-frequency noise hearing loss percentage corresponding to each steady-state noise equivalent sound level, and the estimated high-frequency noise hearing loss percentage corresponding to each kurtosis-adjusted equivalent sound level.

[0049] Wherein, a permanent hearing threshold shift exceeding 30dB HL (decibel hearing level) is determined as high frequency noise-induced hearing loss (HFNIHL). In one embodiment, each data of each training data has the same correspondence with each other;

[0050] Count all steady-state permanent hearing threshold shifts corresponding to each steady-state noise equivalent sound level, determine the number of steady-state high-frequency noise-induced hearing losses among all steady-state permanent hearing threshold shifts, and determine the ratio of the number of steady-state high-frequency noise-induced hearing losses to the number of all steady-state permanent hearing threshold shifts as the proportion of steady-state high-frequency noise-induced hearing loss.

[0051] Count all the estimated permanent hearing threshold shifts corresponding to each kurtosis-adjusted equivalent sound level, determine the number of estimated high-frequency noise hearing losses among all the estimated permanent hearing threshold shifts, and determine the ratio of the number of estimated high-frequency noise hearing losses to the estimated number of high-frequency noise hearing losses as the estimated high-frequency noise hearing loss proportion.

[0052] Step S130 : determining a target kurtosis equivalent sound level model from a plurality of initial kurtosis equivalent sound level models based on a preset model screening method.

[0053] The target kurtosis equivalent sound level model is used to express the corresponding relationship between the equivalent sound level and the kurtosis-adjusted equivalent sound level. The form of the target kurtosis equivalent sound level model can be found in formula (1).

[0054] Preset model screening methods include:

[0055] S1, based on the estimated permanent hearing threshold shift groups corresponding to each initial kurtosis adjustment model, as well as the test permanent hearing threshold shift and steady-state permanent hearing threshold shift of each training data, determining the underestimation and improvement degree corresponding to each initial kurtosis adjustment model;

[0056] The difference between the estimated permanent hearing threshold shift group corresponding to each initial kurtosis adjustment model and the test permanent hearing threshold shift of each training data set is determined as the underestimated value corresponding to the initial kurtosis adjustment model. For example, the difference between the average value of the estimated permanent hearing threshold shift group corresponding to each initial kurtosis adjustment model and the average value of the test permanent hearing threshold shift of each training data set is determined as the underestimated value corresponding to the initial kurtosis adjustment model.

[0057] The difference between the estimated permanent hearing threshold shift group corresponding to each initial kurtosis adjustment model and the steady-state permanent hearing threshold shift of each training data is determined as the improvement level corresponding to the initial kurtosis adjustment model. For example, the difference between the average value of the estimated permanent hearing threshold shift group corresponding to each initial kurtosis adjustment model and the average value of the steady-state permanent hearing threshold shift of each training data is determined as the improvement level corresponding to the initial kurtosis adjustment model.

[0058] S2. Perform logistic regression curve fitting on each steady-state noise equivalent sound level and the corresponding steady-state high-frequency noise hearing loss percentage value to obtain a steady-state curve. Also perform logistic regression curve fitting on each kurtosis-adjusted equivalent sound level and the estimated high-frequency noise hearing loss percentage value to obtain a logistic regression curve corresponding to each initial kurtosis-adjusted equivalent sound level.

[0059] Among them, the kurtosis-adjusted equivalent sound level was used as the independent variable (i.e., the X-axis), and the proportion of high-frequency noise hearing loss was used as the dependent variable (i.e., the Y-axis), and logistic regression curve fitting was performed.

[0060] Based on each steady-state noise equivalent sound level and the corresponding percentage of steady-state high-frequency noise hearing loss for each steady-state noise level, a logistic regression curve was drawn and determined as the steady-state curve; the logistic regression curve was fitted using each kurtosis-adjusted equivalent sound level and the estimated percentage of high-frequency noise hearing loss corresponding to each kurtosis-adjusted equivalent sound level to obtain the logistic regression curve corresponding to each initial kurtosis-adjusted model.

[0061] S3. Determine the target kurtosis equivalent sound level model based on the logistic regression curve, steady-state curve, improvement degree and underestimation corresponding to each initial kurtosis adjustment model.

[0062] Among them, the initial kurtosis adjustment model with the logistic regression curve closest to the steady-state curve, the largest improvement value and the smallest underestimation value is determined as the target kurtosis equivalent sound level model.

[0063] Step S140 , adjusting the equivalent sound level of the test subject using the target kurtosis equivalent sound level model to obtain the kurtosis-adjusted equivalent sound level of the test subject, so as to evaluate the hearing loss of the test subject based on the kurtosis-adjusted equivalent sound level.

[0064] The test subject may be a person undergoing hearing loss assessment. The equivalent sound level of the test subject can be adjusted using the target kurtosis equivalent sound level model to obtain the test subject's kurtosis-adjusted equivalent sound level. Based on the kurtosis-adjusted equivalent sound level, formula (3) can be used to determine the test subject's permanent hearing threshold shift. Based on the test subject's permanent hearing threshold shift, the test subject's hearing loss (hearing disorders) can be determined. Hearing loss, also known as deafness or hearing level, is the number of decibels by which the human ear's hearing threshold at a certain frequency is higher than the normal hearing threshold. In the embodiment of the present disclosure, by setting a plurality of initial kurtosis equivalent sound level models representing the correspondence between the equivalent sound level and the kurtosis-adjusted equivalent sound level, the steady-state noise equivalent sound level, the steady-state permanent hearing threshold shift and the steady-state high-frequency noise hearing loss ratio of each training data are used to determine the target kurtosis equivalent sound level model by means of logistic regression. Thus, the noise exposure kurtosis is added to the equivalent sound level through the target kurtosis equivalent sound level model, thereby realizing the introduction of noise exposure kurtosis as a quantitative indicator of noise time domain characteristics in the equivalent sound level, overcoming the problem of traditional equivalent noise indicators underestimating the hazards of impact and intermittent noise, and improving the accuracy of hearing loss prediction.

[0065] Figure 2 FIG. 1 is a flow chart of a hearing loss prediction method provided by another exemplary embodiment of the present disclosure. Figure 2 As shown, the method of obtaining the first data set may include the following steps:

[0066] Step S200 , collecting test noise equivalent sound level, test noise exposure kurtosis, test permanent hearing threshold shift and noise exposure duration of each sample subject in a test noise environment.

[0067] Among them, 4,000 occupational noise-exposed workers were selected as sample subjects. The sample subjects had a length of service of 5 to 35 years, an age range of 20 to 60 years, and a gender ratio of male to female of approximately 2:1. The preset noise equivalent sound level can range from 75dB to 105dB. The test permanent hearing threshold shift of each sample subject in the test noise environment can be determined based on subjective behavioral audiometry in ISO 8253-1 "Acoustics - Basic Pure-Tone Air and Bone Conductance Audiometry," the Noise-Induced Hearing Loss Diagnosis Package (GBZ 49-2014), cortical auditory evoked potentials, and other methods. The test noise environment at each preset noise equivalent sound level can be understood as having a noise equivalent sound level of the preset noise equivalent sound level.

[0068] In one embodiment, the test noise exposure kurtosis may be obtained by the following method: for multiple sample subjects, collecting the noise in the test noise environment where the sample subjects are located, and determining the corresponding test noise exposure kurtosis of the sample subjects based on the respective noise amplitude values and the average noise amplitude value of the noise in the test noise environment.

[0069] The preset duration can be 8 hours. The noise of the preset duration in the test noise environment is divided into multiple noise segments of 60 seconds. For each noise segment, based on the noise amplitude values and the average noise amplitude value in the noise segment, the noise exposure kurtosis corresponding to the noise segment is determined using formula (4). The total noise exposure kurtosis is determined using formula (5), and the total test noise exposure kurtosis is determined as the test noise exposure kurtosis corresponding to the sample object.

[0070]

[0071] In formulas (4)-(5), x i is the i-th noise amplitude value, is the mean value of the flat noise, β j is the noise exposure kurtosis corresponding to the noise segment, and β is the total noise exposure kurtosis.

[0072] Step S210 : collecting the steady-state noise equivalent sound level and steady-state permanent hearing threshold shift of each sample person in a steady-state noise environment.

[0073] In one embodiment, the sample subjects are subjected to pure tone air conduction hearing threshold tests at 3kHz, 4kHz, and 6kHz in a soundproof room (steady-state noise environment) 24 hours after leaving the noise environment, wherein the soundproof room has noise of a preset noise equivalent sound level.

[0074] In a steady-state noise environment at each noise equivalent sound level, the age- and gender-corrected pure-tone hearing thresholds (HTLs) at noise frequencies of 3 kHz, 4 kHz, and 6 kHz were determined. The corresponding steady-state permanent hearing threshold shift for the sample subject was then determined based on formula (6). The pure-tone hearing threshold is the minimum intensity (sound pressure level) at which a test subject can detect the presence of a sound at a specific frequency using a standardized pure-tone audiometry method.

[0075]

[0076] In formula (6), NIPTS 346 Indicates the steady-state permanent hearing threshold shift, HTL 3kHz 、HTL 4kHz and HTL 6kGHz The pure tone hearing thresholds are 3kHz, 4kHz and 6kHz respectively.

[0077] In step S220 , the test noise equivalent sound level, test noise exposure kurtosis, test permanent hearing threshold shift, noise exposure duration, steady-state noise equivalent sound level, and steady-state permanent hearing threshold shift corresponding to each sample object are used as a piece of training data.

[0078] In one embodiment, for each piece of training data, based on the Synthetic Minority Oversampling Technique (SMOTE) algorithm, Z-score normalization is used to eliminate dimensional differences and eliminate abnormal training data outside ±3σ.

[0079] In some optional embodiments, the disclosed embodiments may include obtaining the duration of the test subject's noise exposure in a target noise environment. In one embodiment, the frequency domain and reference sound pressure level of the noise in the target noise environment are obtained. The target noise environment can be understood as the test subject's working environment.

[0080] Accordingly, the hearing loss of the test subject is assessed in the following manner: the equivalent sound level is adjusted based on the exposure duration and kurtosis, the permanent hearing threshold shift of the test subject is determined using a preset permanent hearing threshold shift prediction model, and the hearing loss of the test subject is estimated based on the permanent hearing threshold shift of the test subject.

[0081] Among them, obtain L in the target noise environment EX,8h (equivalent sound level) is used as the equivalent sound level of the test object, and then the equivalent sound level of the test object is processed based on the target kurtosis equivalent sound level model to obtain the kurtosis-adjusted equivalent sound level corresponding to the test object. Formula (3) is used to determine the permanent hearing threshold shift of the test object, and then the hearing loss of the test object is estimated based on the permanent hearing threshold shift of the test object. The value of the permanent hearing threshold shift of the test object can be determined as the value of the hearing loss.

[0082] In some optional implementations, determining the first data set in the embodiments of the present disclosure may include:

[0083] Constructing a validation set and multiple data sets, wherein the multiple data sets include a first data set, and the validation set includes multiple training data.

[0084] In one embodiment, a stratified five-fold cross-validation mechanism is used to generate a validation set and four data sets. Specifically, the plurality of training data are divided into five parts, four of which are used as the four data sets, and the remaining one is used as the validation set. The validation set and the four first data sets are mutually exclusive.

[0085] Figure 3FIG. 1 is a flow chart of a hearing loss prediction method provided by another exemplary embodiment of the present disclosure. Figure 3 As shown, the hearing loss prediction method may further include the following steps:

[0086] Step S300 : Determine the kurtosis-adjusted equivalent sound level corresponding to each training data in the validation set by using the target kurtosis equivalent sound level model and each test equivalent sound level in the validation set.

[0087] Among them, the test equivalent sound levels in the verification set are input into the target kurtosis equivalent sound level model in turn to obtain the kurtosis-adjusted equivalent sound levels corresponding to each test equivalent sound level.

[0088] Step S310 : determining the estimated permanent hearing threshold shift corresponding to each training data based on the kurtosis-adjusted equivalent sound level.

[0089] Among them, based on the kurtosis-adjusted equivalent sound level, the noise exposure time of each test data and the frequency domain of the noise, the estimated permanent hearing threshold shift corresponding to each training data is obtained using formula (3).

[0090] Step S320 : determining the accuracy of the target kurtosis equivalent sound level model based on the ratio of the measured error between the estimated permanent hearing threshold shift and the steady-state permanent hearing threshold shift corresponding to each training data.

[0091] Among them, for each test data, the absolute value of the difference between the estimated permanent hearing threshold shift and the steady-state permanent hearing threshold shift corresponding to the training data is determined, and the ratio between the absolute value of the difference and the steady-state permanent hearing threshold shift is determined as the measurement error ratio of the test data; the average value of the measurement error ratios corresponding to each test data is used as the accuracy of the target kurtosis equivalent sound level model.

[0092] Step S330, in response to the accuracy of the target kurtosis equivalent sound level model not meeting the preset accuracy condition, using the second data set and based on the preset model screening method, re-determine the target kurtosis equivalent sound level model from multiple initial kurtosis equivalent sound level models until the accuracy of the target kurtosis equivalent sound level model meets the preset accuracy condition.

[0093] The second data set is a data set other than the first data set among the multiple data sets.

[0094] In one embodiment, the preset accuracy condition includes a preset accuracy threshold. When the accuracy of the target kurtosis equivalent sound level model exceeds the preset accuracy threshold, it is determined that the accuracy of the target kurtosis equivalent sound level model does not meet the preset accuracy condition. When the accuracy of the target kurtosis equivalent sound level model does not exceed the preset accuracy threshold, it is determined that the accuracy of the target kurtosis equivalent sound level model meets the preset accuracy condition.

[0095] When the accuracy of the target kurtosis equivalent sound level model does not meet the preset accuracy condition, other first data sets are reselected from the four first data sets, and then steps S110 to S130 are executed to re-determine the target kurtosis equivalent sound level model from the multiple initial kurtosis equivalent sound level models.

[0096] In an application example, R Studio software version 4.4.1 is used as a statistical analysis tool to determine the target kurtosis equivalent sound level model. N / β G , respectively, using multiple linear regression model and nonlinear regression model to establish 6 groups of models. Specifically, using each test data to establish a multiple linear regression model Find L EX，8h The regression coefficient b1 and the adjustment term regression coefficient b2, the ratio b2 / b1 is the adjustment coefficient λ, b0 is a constant. For the nonlinear regression model, choose to establish the following equation a, b, c, f, h and g are constants, and L is calculated based on the equation results. EX，8h and adjustments Find the first-order derivative function f′(L EX，8h )and use To obtain λ. The specific results are as follows: The adjustment coefficients obtained by using the multiple linear regression method are all 3.0, and the coefficients obtained by using the multiple nonlinear regression method are: β G =3,λ=4.4;β G =6.6,λ=6.3;β G =10,λ=7.9. In all adjustment equations, when β N <β G When , λ is 0.

[0097] Based on β G =3,λ=4.4;β G =6.6,λ=6.3;β G When =10 and λ=7.9, six initial kurtosis equivalent sound level models are constructed, which are:

[0098] First, based on the test equivalent sound level, test noise exposure kurtosis and noise exposure duration in each training data, determine the kurtosis-adjusted equivalent sound level group and the estimated permanent hearing threshold shift group corresponding to each initial kurtosis equivalent sound level model. The difference between the average value of the estimated permanent hearing threshold shift group corresponding to each initial kurtosis adjustment model and the average value of the test permanent hearing threshold shift of each training data is determined as the underestimated value corresponding to the initial kurtosis adjustment model. The difference between the average value of the estimated permanent hearing threshold shift group corresponding to each initial kurtosis adjustment model and the steady-state permanent hearing threshold shift (NIPTS346 The difference between the average values of the initial kurtosis adjustment model and the average value of the initial kurtosis adjustment model is determined as the improvement degree of the initial kurtosis adjustment model. The underestimation and improvement degree are detailed in Table 2. NIPTS of each test data 346 Based on formula (6), the initial kurtosis equivalent sound level model line L can be obtained from Table 2. EX,8h -K6 corresponds to a larger improvement value and a smaller underestimation value. EX,8h -K6 achieved the best results.

[0099] Table 2

[0100]

[0101] Note: All L EX -K n The low estimates of the group are all related to L EX,8h The results were compared and statistically tested using the t test; EX,8h -K 1～3 λ=3.0,β G =3, 6.6 and 10, L EX,8h -K4 is λ=4.4, β G =3,L EX,8h -K5 is λ=6.3, β G =6.6, L EX,8h -K6 is λ=7.9, β G =10.

[0102] Second, the percentage of high-frequency noise hearing loss (HFNIHL%) was used as the dependent variable. In the steady-state noise group, L EX,8h As the independent variable, L is used in the non-stationary noise group. EX,8h and L EX,8h -K n Establish a dose-response relationship for the independent variable. In this application example, the test equivalent sound level of each test data is used as the data on the X axis, and the test high-frequency noise hearing loss percentage (HFNIHL%) of each test data is used as the data on the Y axis to determine the non-stationary Logistic regression curve (based on the non-stationary noise L EX,8h Based on the steady-state noise equivalent sound level and the steady-state high-frequency noise hearing loss ratio (HFNIHL%) of each training data, a logistic regression curve is drawn to determine the steady-state curve (consisting of the steady-state noise L EX,8h Using the kurtosis-adjusted equivalent sound level group and the estimated high-frequency noise hearing loss ratio group corresponding to each initial kurtosis-adjusted model, the logistic regression curve corresponding to each initial kurtosis-adjusted model is drawn (respectively represented by the non-stationary noise L EX,8h -K1~non-steady-state noise L EX,8h -K6) indicates the non-stationary noise L EX,8h-K1~non-steady-state noise L EX,8h -K6, non-stationary noise L EX,8h and steady-state noise L EX,8h The regression curves in the expression are as follows Figure 4 As shown, Figure 4 FIG is a schematic diagram of a logistic regression curve between HFNIHL% and equivalent sound level provided by an exemplary embodiment of the present disclosure. Figure 4 In the figure, the X-axis represents the equivalent sound level, using L EX,8h The unit is [dB(A)], and the Y-axis represents the proportion of high-frequency noise hearing loss, using HFNIHL%.

[0103] based on Figure 4 Determine the regression logistic curves and determine Figure 4 Display steady-state noise L EX,8h L for two noise types in the range of 70-103dB(A) EX,8h -L EX,8h -K n Logistic regression curve of HFNIHL%, regression curve of steady-state noise (steady-state noise L EX,8h ) equation HFNIHL%=1 / [1+exp(-0.0836×L EX,8h +10.0)]. Regression curve of non-stationary noise (non-stationary noise L EX,8h ) equation HFNIHL%=1 / [1+exp(-0.0940×L EX,8h +9.7).

[0104] Regression curve L of non-stationary noise EX,8h -K1~L EX,8h -K6 equations are:

[0105] HFNIHL%=1 / [1+exp(-0.0968×L EX,8h- K1+10.3)]),

[0106] HFNIHL%=1 / [1+exp(-0.0949×L EX,8h -K2+10.1)]),

[0107] HFNIHL%=1 / [1+exp(-0.0940×L EX,8h -K3+9.9)]),

[0108] HFNIHL%=1 / [1+exp(-0.0938×L EX,8h -K4+10.2)]),

[0109] HFNIHL%=1 / [1+exp(-0.0943×LEX,8h -K5+10.5)],

[0110] HFNIHL%=1 / [1+exp(-0.0848×L EX,8h -K6+9.7)]).

[0111] L of non-stationary noise EX,8h (Red) in the L EX,8h The HFNIHL% in the range is 9.4% higher than that of the steady-state noise group (blue), which is a significant difference. EX,8h -K6 (blue) in the L EX,8h The difference between the HFNIHL% in the range and the HFNIHL% in the steady-state noise group (red) is the smallest at 4.1% (p=0.053), which is the closest to the steady-state noise regression curve. EX,8h -K6 corresponding logistic regression curve and steady-state noise L EX,8h The corresponding logistic regression curve (steady-state curve) is closest, so the initial kurtosis equivalent sound level model L is selected. EX,8h -K6 has the best effect. In summary, the initial kurtosis equivalent sound level model L EX,8h -K6 is closest to the steady-state curve, and the initial kurtosis equivalent sound level model line L EX,8h -K6 corresponds to a larger improvement value and a smaller underestimation value. EX,8h -K6 is determined as the target kurtosis equivalent sound level model.

[0112] Figure 5 FIG. 1 is a schematic diagram of a hearing loss prediction device provided by an exemplary embodiment of the present disclosure. Figure 5 As shown, the device of this embodiment may include:

[0113] A first data acquisition module 400 is configured to acquire a first data set and multiple initial kurtosis equivalent sound level models, wherein the first data set includes multiple training data, each training data including a test equivalent sound level, a test noise exposure kurtosis, a test permanent hearing threshold shift, a noise exposure duration, a steady-state noise equivalent sound level, and a steady-state permanent hearing threshold shift, and each initial kurtosis equivalent sound level model is configured to represent a correspondence between an equivalent sound level and a kurtosis-adjusted equivalent sound level;

[0114] A first data determination module 410 is configured to determine a kurtosis-adjusted equivalent sound level group and an estimated permanent hearing threshold shift group corresponding to each initial kurtosis equivalent sound level model based on the test equivalent sound level, the test noise exposure kurtosis, and the noise exposure duration in each training data, wherein the kurtosis-adjusted equivalent sound level group includes a plurality of kurtosis-adjusted equivalent sound levels, and the estimated permanent hearing threshold shift group includes a plurality of estimated permanent hearing threshold shifts.

[0115] The second data determination module 420 is configured to determine the percentage of steady-state high-frequency noise hearing loss corresponding to each steady-state noise equivalent sound level, and the estimated percentage of high-frequency noise hearing loss corresponding to each kurtosis-adjusted equivalent sound level;

[0116] A model screening module 430 is configured to determine a target kurtosis equivalent sound level model from the multiple initial kurtosis equivalent sound level models based on a preset model screening method. The preset model screening method includes: determining the underestimation and improvement degree corresponding to each initial kurtosis adjustment model based on the estimated permanent hearing threshold shift group corresponding to each initial kurtosis adjustment model, and the test permanent hearing threshold shift and steady-state permanent hearing threshold shift based on each training data; performing logistic regression curve fitting on each steady-state noise equivalent sound level and the steady-state high-frequency noise hearing loss percentage corresponding to each steady-state noise equivalent sound level to obtain a steady-state curve; and performing logistic regression curve fitting on each kurtosis-adjusted equivalent sound level and the estimated high-frequency noise hearing loss percentage corresponding to each kurtosis-adjusted equivalent sound level to obtain a logistic regression curve corresponding to each initial kurtosis adjustment model; and determining the target kurtosis equivalent sound level model based on the logistic regression curve, steady-state curve, improvement degree, and underestimation corresponding to each initial kurtosis adjustment model.

[0117] The equivalent sound level correction module 440 is configured to adjust the equivalent sound level of the test subject using the target kurtosis equivalent sound level model to obtain the kurtosis-adjusted equivalent sound level of the test subject, so as to assess the hearing loss of the test subject based on the kurtosis-adjusted equivalent sound level.

[0118] In some possible implementations of the present disclosure, the hearing loss prediction device in an embodiment of the present disclosure further includes:

[0119] The first data acquisition module is used to collect the test noise equivalent sound level, test noise exposure kurtosis, test permanent hearing threshold shift and noise exposure duration of each sample subject in the test noise environment;

[0120] The second data acquisition module is used to collect the steady-state noise equivalent sound level and steady-state permanent hearing threshold displacement of the sample object in a steady-state noise environment;

[0121] The training data generation module is used to use the test noise equivalent sound level, test noise exposure kurtosis, test permanent hearing threshold shift, noise exposure duration, steady-state noise equivalent sound level and steady-state permanent hearing threshold shift corresponding to each sample object as a piece of training data.

[0122] In some possible implementations of the present disclosure, the test noise exposure kurtosis of each sample object in a test noise environment with multiple noise intensities is collected in the embodiments of the present disclosure, and is further used to:

[0123] For the plurality of sample objects, collecting noise in a test noise environment where the sample objects are located;

[0124] The corresponding test noise exposure kurtosis of the sample object is determined based on the respective noise amplitude values and the average noise amplitude value of the noise.

[0125] In some possible implementations of the present disclosure, the hearing loss prediction device in an embodiment of the present disclosure further includes:

[0126] A second data acquisition module is used to obtain the noise exposure time of the test subject in the target noise environment;

[0127] an hearing threshold assessment module, configured to determine the permanent hearing threshold shift of the test subject based on the noise exposure duration and the kurtosis-adjusted equivalent sound level and using a preset permanent hearing threshold shift prediction model;

[0128] The hearing loss of the test subject is estimated based on the permanent hearing threshold shift of the test subject.

[0129] In some possible implementations of the present disclosure, the hearing loss prediction device in an embodiment of the present disclosure further includes:

[0130] The data set construction module is used to construct a validation set and multiple data sets, wherein the multiple data sets include a first data set, and the validation set includes multiple training data.

[0131] In some possible implementations of the present disclosure, the hearing loss prediction device in an embodiment of the present disclosure further includes:

[0132] A first verification module is configured to determine the kurtosis-adjusted equivalent sound level corresponding to each training data in the verification set by using the target kurtosis equivalent sound level model and each test equivalent sound level in the verification set;

[0133] A second verification module is used to determine the estimated permanent hearing threshold shift corresponding to each training data based on the kurtosis-adjusted equivalent sound level;

[0134] a third verification module, configured to determine the accuracy of the target kurtosis equivalent sound level model based on a ratio of the measured error between the estimated permanent hearing threshold shift and the steady-state permanent hearing threshold shift corresponding to each training data;

[0135] A fourth verification module is used to, in response to the accuracy of the target kurtosis equivalent sound level model not meeting the preset accuracy condition, use a second data set and, based on the preset model screening method, redetermine the target kurtosis equivalent sound level model among the multiple initial kurtosis equivalent sound level models until the accuracy of the target kurtosis equivalent sound level model meets the preset accuracy condition, wherein the second data set is a data set among the multiple data sets other than the first data set.

[0136] The hearing loss prediction device of the embodiment of the present disclosure corresponds to the hearing loss prediction method of the embodiment of the present disclosure with the above-mentioned equity relationship, and the relevant contents can be referenced to each other and will not be repeated here.

[0137] The beneficial technical effects corresponding to the exemplary embodiments of the hearing loss prediction device of the embodiments of the present disclosure can be referred to the corresponding beneficial technical effects in the above corresponding exemplary method part, and will not be repeated here.

[0138] In addition, an embodiment of the present disclosure further provides an electronic device, including:

[0139] memory for storing computer programs;

[0140] The processor is configured to execute the computer program stored in the memory, and when the computer program is executed, the hearing loss prediction method described in any one of the above embodiments of the present disclosure is implemented.

[0141] Figure 6 This is a schematic diagram of the structure of an application embodiment of the electronic device disclosed in the present invention. Figure 6 The electronic device according to the embodiment of the present disclosure is described. The electronic device may be either or both of the first device and the second device, or a standalone device independent of them, and the standalone device may communicate with the first device and the second device to receive collected input signals from them.

[0142] like Figure 6 As shown, the electronic device includes one or more processors and memory.

[0143] The processor may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.

[0144] The memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), a hard disk, or flash memory. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor may execute the program instructions to implement the hearing loss prediction methods of the various embodiments of the present disclosure described above and / or other desired functions.

[0145] In one example, the electronic device may further include an input device and an output device, and these components are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0146] In addition, the input device may also include, for example, a keyboard, a mouse, and the like.

[0147] The output device can output various information to the outside, including determined distance information, direction information, etc. The output device can include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, and the like.

[0148] Of course, to simplify, Figure 6 Only some of the components related to the present disclosure in the electronic device are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, the electronic device may further include any other appropriate components according to specific application scenarios.

[0149] In addition to the above-mentioned methods and devices, embodiments of the present disclosure may also be a computer program product, which includes computer program instructions. When the computer program instructions are executed by a processor, the processor performs the steps of the hearing loss prediction method according to various embodiments of the present disclosure described in the above part of this specification.

[0150] The computer program product may be written in any combination of one or more programming languages to implement the operations of the disclosed embodiments, including object-oriented programming languages such as Java, C++, and conventional procedural programming languages such as C or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0151] In addition, an embodiment of the present disclosure may also be a computer-readable storage medium having computer program instructions stored thereon. When the computer program instructions are executed by a processor, the processor performs the steps of the hearing loss prediction method according to various embodiments of the present disclosure described in the above part of this specification.

[0152] The computer-readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can, for example, include but is not limited to a system, device or component of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0153] Those skilled in the art will understand that all or part of the steps of implementing the above-mentioned method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above-mentioned method embodiment; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk, etc. Various media that can store program codes.

[0154] The basic principles of the present disclosure have been described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, and effects mentioned in this disclosure are merely illustrative and not restrictive, and should not be construed as necessarily possessed by each embodiment of the present disclosure. Furthermore, the specific details disclosed above are provided for illustrative purposes and to facilitate understanding, rather than as limitations. These details do not limit the present disclosure to necessarily being implemented using these specific details.

[0155] Each embodiment in this specification is described in a progressive manner, with each embodiment focusing on its differences from the other embodiments. References to the same or similar parts between the various embodiments are sufficient. For system embodiments, since they largely correspond to method embodiments, their description is relatively simple. For relevant parts, references to the description of the method embodiments are sufficient.

[0156] The block diagrams of the devices, devices, equipment, and systems involved in this disclosure are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As will be appreciated by those skilled in the art, these devices, devices, equipment, and systems can be connected, arranged, or configured in any manner. Words such as "include," "comprise," "have," and the like are open-ended words, meaning "including but not limited to," and can be used interchangeably therewith. The words "or" and "and" used herein refer to the words "and / or" and can be used interchangeably therewith, unless the context clearly indicates otherwise. The word "such as" used herein refers to the phrase "such as but not limited to," and can be used interchangeably therewith.

[0157] The methods and apparatus of the present disclosure may be implemented in many ways. For example, the methods and apparatus of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above unless otherwise specified. In addition, in some embodiments, the present disclosure may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the methods according to the present disclosure. Thus, the present disclosure also covers recording media that store programs for executing the methods according to the present disclosure.

[0158] It should also be noted that in the apparatus, device, and method of the present disclosure, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present disclosure.

[0159] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the aspects shown herein, but rather to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0160] The above description has been provided for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present disclosure to the forms disclosed herein. Although a number of example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

Claims

1. A hearing loss prediction method, characterized in that: include: Obtaining a first data set and multiple initial kurtosis equivalent sound level models, wherein the first data set includes multiple training data, each training data includes a test equivalent sound level, a test noise exposure kurtosis, a test permanent hearing threshold shift, a noise exposure duration, a steady-state noise equivalent sound level, and a steady-state permanent hearing threshold shift, and each initial kurtosis equivalent sound level model is used to represent a corresponding relationship between an equivalent sound level and a kurtosis-adjusted equivalent sound level; Based on the test equivalent sound level, the test noise exposure kurtosis and the noise exposure duration in each training data, determining a kurtosis-adjusted equivalent sound level group and an estimated permanent hearing threshold shift group corresponding to each initial kurtosis equivalent sound level model, the kurtosis-adjusted equivalent sound level group including a plurality of kurtosis-adjusted equivalent sound levels, and the estimated permanent hearing threshold shift group including a plurality of estimated permanent hearing threshold shifts; Determine the percentage of steady-state high-frequency noise hearing loss corresponding to each steady-state noise equivalent sound level, and the estimated percentage of high-frequency noise hearing loss corresponding to each kurtosis-adjusted equivalent sound level; Based on a preset model screening method, a target kurtosis equivalent sound level model is determined from the multiple initial kurtosis equivalent sound level models, the preset model screening method comprising: determining the underestimation and improvement degree corresponding to each initial kurtosis adjustment model based on the estimated permanent hearing threshold shift group corresponding to each initial kurtosis adjustment model, and the test permanent hearing threshold shift and steady-state permanent hearing threshold shift based on each training data; performing logistic regression curve fitting on each steady-state noise equivalent sound level and the steady-state high-frequency noise hearing loss percentage corresponding to each steady-state noise equivalent sound level to obtain a steady-state curve, and performing logistic regression curve fitting on each kurtosis-adjusted equivalent sound level and the estimated high-frequency noise hearing loss percentage corresponding to each kurtosis-adjusted equivalent sound level to obtain a logistic regression curve corresponding to each initial kurtosis adjustment model; determining the target kurtosis equivalent sound level model based on the logistic regression curve, steady-state curve, improvement degree and underestimation corresponding to each initial kurtosis adjustment model; The target kurtosis equivalent sound level model is used to adjust the equivalent sound level of the test subject to obtain the kurtosis-adjusted equivalent sound level of the test subject, so as to evaluate the hearing loss of the test subject based on the kurtosis-adjusted equivalent sound level.

2. The method according to claim 1, characterized in that Acquiring the first data set includes: Collecting test noise equivalent sound level, test noise exposure kurtosis, test permanent hearing threshold shift and noise exposure duration of each sample subject in a test noise environment; collecting the steady-state noise equivalent sound level and steady-state permanent hearing threshold shift of each sample object in a steady-state noise environment; The test noise equivalent sound level, test noise exposure kurtosis, test permanent hearing threshold shift, noise exposure duration, steady-state noise equivalent sound level and steady-state permanent hearing threshold shift corresponding to each sample object are used as a piece of training data.

3. The method according to claim 2, characterized in that The test noise exposure kurtosis of each sample object in a test noise environment with multiple noise intensities is collected, including: For the plurality of sample objects, collecting noise in a test noise environment where the sample objects are located; The corresponding test noise exposure kurtosis of the sample object is determined based on the respective noise amplitude values and the average noise amplitude value of the noise.

4. The method according to claim 1, wherein Also includes: Obtaining the noise exposure duration of the test subject in the target noise environment; The assessing of the hearing loss of the test subject based on the kurtosis-adjusted equivalent sound level comprises: Determining the permanent hearing threshold shift of the test subject using a preset permanent hearing threshold shift prediction model based on the noise exposure duration and the kurtosis-adjusted equivalent sound level; The hearing loss of the test subject is estimated based on the permanent hearing threshold shift of the test subject.

5. The method according to claim 2, characterized in that Also includes: A validation set and multiple data sets are constructed, where the multiple data sets include a first data set, and the validation set includes multiple training data.

6. The method according to claim 5, characterized in that Also includes: Determine the kurtosis-adjusted equivalent sound level corresponding to each training data in the validation set by using the target kurtosis equivalent sound level model and each test equivalent sound level in the validation set; Determine the estimated permanent hearing threshold shift corresponding to each training data based on the kurtosis-adjusted equivalent sound level; determining the accuracy of the target kurtosis equivalent sound level model based on a ratio of measured errors between the estimated permanent hearing threshold shift and the steady-state permanent hearing threshold shift corresponding to each training data; In response to the accuracy of the target kurtosis equivalent sound level model not meeting the preset accuracy condition, the target kurtosis equivalent sound level model is re-determined among the multiple initial kurtosis equivalent sound level models based on the preset model screening method using a second data set until the accuracy of the target kurtosis equivalent sound level model meets the preset accuracy condition, and the second data set is a data set among the multiple data sets except the first data set.

7. A hearing loss prediction device, characterized in that: include: a first data acquisition module, configured to acquire a first data set and a plurality of initial kurtosis equivalent sound level models, wherein the first data set includes a plurality of training data, each training data including a test equivalent sound level, a test noise exposure kurtosis, a test permanent hearing threshold shift, a noise exposure duration, a steady-state noise equivalent sound level, and a steady-state permanent hearing threshold shift, and each initial kurtosis equivalent sound level model is used to represent a correspondence between an equivalent sound level and a kurtosis-adjusted equivalent sound level; A first data determination module is configured to determine, based on the test equivalent sound level, the test noise exposure kurtosis, and the noise exposure duration in each training data, a kurtosis-adjusted equivalent sound level group and an estimated permanent hearing threshold shift group corresponding to each initial kurtosis equivalent sound level model, wherein the kurtosis-adjusted equivalent sound level group includes a plurality of kurtosis-adjusted equivalent sound levels, and the estimated permanent hearing threshold shift group includes a plurality of estimated permanent hearing threshold shifts; The second data determination module is used to determine the steady-state high-frequency noise hearing loss ratio corresponding to each steady-state noise equivalent sound level, and the estimated high-frequency noise hearing loss ratio corresponding to each kurtosis-adjusted equivalent sound level; a model screening module for determining a target kurtosis equivalent sound level model from the multiple initial kurtosis equivalent sound level models based on a preset model screening method, the preset model screening method comprising: determining the underestimation and improvement degree corresponding to each initial kurtosis adjustment model based on the estimated permanent hearing threshold shift group corresponding to each initial kurtosis adjustment model, and the test permanent hearing threshold shift and steady-state permanent hearing threshold shift based on each training data; performing logistic regression curve fitting on each steady-state noise equivalent sound level and the steady-state high-frequency noise hearing loss percentage corresponding to each steady-state noise equivalent sound level to obtain a steady-state curve, and performing logistic regression curve fitting on each kurtosis-adjusted equivalent sound level and the estimated high-frequency noise hearing loss percentage corresponding to each kurtosis-adjusted equivalent sound level to obtain a logistic regression curve corresponding to each initial kurtosis adjustment model; and determining the target kurtosis equivalent sound level model based on the logistic regression curve, steady-state curve, improvement degree, and underestimation corresponding to each initial kurtosis adjustment model; The equivalent sound level correction module is used to adjust the equivalent sound level of the test subject using the target kurtosis equivalent sound level model to obtain the kurtosis-adjusted equivalent sound level of the test subject, so as to evaluate the hearing loss of the test subject based on the kurtosis-adjusted equivalent sound level.

8. The device according to claim 7, characterized in that Also includes: The first data acquisition module is used to collect the test noise equivalent sound level, test noise exposure kurtosis, test permanent hearing threshold shift and noise exposure duration of each sample subject in the test noise environment; The second data acquisition module is used to collect the steady-state noise equivalent sound level and steady-state permanent hearing threshold displacement of the sample object in a steady-state noise environment; The training data generation module is used to use the test noise equivalent sound level, test noise exposure kurtosis, test permanent hearing threshold shift, noise exposure duration, steady-state noise equivalent sound level and steady-state permanent hearing threshold shift corresponding to each sample object as a piece of training data.

9. An electronic device, characterized in that: include: memory for storing computer programs; The processor is configured to execute the computer program stored in the memory, and when the computer program is executed, the hearing loss prediction method according to any one of claims 1 to 6 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the hearing loss prediction method according to any one of claims 1 to 6 is implemented.

Images