Instrument for detecting auditory evoked nerve responses

Abstract

The present application discloses an instrument for detecting an auditory evoked nerve response, the instrument comprising: a stimulation generator configured to generate at least one stimulation or a plurality of consecutive stimuli according to a test protocol; at least one output unit including a transducer, the output unit configured to receive at least one stimulus or a plurality of consecutive stimuli and provide the stimuli to the subject; comprising one or more sensors for measuring one or more evoked responses of the subject in response to the provided stimuli; an analysis unit configured to receive and analyze the measured one or more evoked responses, where the analysis unit is configured to determine a probability p that each of the responses is driven by potential background noise during operation of the instrument, the analysis unit is configured to determine the probability p based on a value F of the measured one or more evoked responses determined as a ratio between a variance of an average of the one or more evoked responses and a variance of residual background noise.

Description

Technical Field

[0001] The present application relates to apparatus for detecting evoked responses.

[0002] The present application also relates to a method for detecting an induced response.

[0003] The application also relates to a detector unit.

[0004] The present application also relates to a data processing system comprising a processor and a program code causing the processor to execute at least part of the steps of the method of the present application.

[0005] The present application also relates to a computer program comprising instructions which, when the program is run by an apparatus, instruct the apparatus to perform the steps of the above-described method. Background Art

[0006] Objective detection of auditory evoked neural activity plays a key role in the diagnosis and early intervention of hearing loss in neonates and infants, and in the neural assessment of auditory function during surgery such as cochlear implantation or acoustic neuroma removal.

[0007] For newborns and infants, hearing screening can be performed with the aid of auditory evoked neural responses (i.e., electrophysiological responses) recorded by electroencephalography (EEG) using electrodes placed on the baby's scalp. These responses include auditory brainstem responses (ABR), cortical auditory potentials (CAP), auditory steady-state responses (ASSR), otoacoustic emissions (OAE), etc. By presenting sounds (stimuli) such as clicks, narrow-band chirps, or speech signals, etc., evoked responses can be obtained by averaging many presentations of the aforementioned stimuli, so that background noise is reduced and evoked responses are observed.

[0008] For decades, the detection of evoked neural responses has been achieved either by visual inspection by a professional audiologist or clinician, or by statistical methods (i.e., objective testing). The latter rely on statistical parameters and their associated probabilities, which indicate the likelihood that the measured parameters are driven by underlying background noise (null hypothesis) rather than representing actual evoked responses (alternative hypothesis).

[0009] These objective methods include parametric methods (F test, Hotelling-T2), non-parametric methods (Q samples), and machine learning-based methods. Each of these methods has advantages and disadvantages. For example, the F test is very efficient and intuitive because it measures the ratio between the energy of the signal with noise and the energy of the noise, which is easy for anyone to understand. However, the disadvantage of the F test is that it does not take into account the various types of potential background noise in the EEG recording, which leads to less accuracy and slower detection.

[0010] The Q-sample is a nonparametric test that relies on bootstrapping or pre-calculated statistical tables (usually derived from simulations using artificial noise signals). While bootstrapping can lead to very good estimates of the underlying background noise, it is computationally expensive and conservative, tending to detect fewer true positive reactions when the number of iterations used to bootstrap the data is limited. Therefore, Q-samples using bootstrapping cannot be used in real-time measurements with standard clinical computers. This can be partially solved by comparing the Q-sample statistics with pre-established tables obtained through simulations using different types of noise. However, this approach does not take into account the individual characteristics of the background noise in real recordings.

[0011] In addition, the statistics obtained through Q samples and Hotelling's T2 method are abstract and difficult for clinicians to understand.

[0012] Finally, machine learning methods depend on training datasets and are prone to replicating biases inherent in those training datasets.

[0013] In recent years, there has been an increasing interest in developing detectors (or detection methods) that are easy to interpret, statistically reliable, and applicable in real-time situations. The impact of these developments is important in producing optimal measurements (fast, reliable, and with a manageable false positive rate), which in turn directly affects the accuracy, efficiency, and cost of clinical centers.

[0014] Currently, no such detector or neural response detection method has been reported, and objective detection of brain activity is achieved by the above-mentioned methods despite the corresponding disadvantages.

[0015] Thus, there is a need for improved objective detection of auditory evoked neural responses, or any other type of response that relies on averaging to reduce potential background noise when detecting a signal. Summary of the invention

[0016] instrument

[0017] In one aspect of the present application, an apparatus (device / apparatus, or a detector of an apparatus) for detecting an induced response is provided.

[0018] For example, the apparatus (device / apparatus, or detector of the apparatus) may be adapted to detect auditory evoked neural responses.

[0019] The instrument may also be configured to estimate the hearing ability of a subject (patient).

[0020] The apparatus may include a stimulation generator.

[0021] The stimulus generator may be configured to generate at least one stimulus. The at least one stimulus may be generated according to a test protocol.

[0022] The stimulus generator may be configured to generate a plurality of sequential stimuli according to a test protocol.

[0023] The apparatus may include at least one output unit.

[0024] The at least one output unit may include a converter.

[0025] The output unit may be configured to receive at least one stimulus or a plurality of consecutive stimuli from the stimulus generator.

[0026] The output unit may be configured to provide / present at least one stimulus or a plurality of consecutive stimuli into the ear of the subject via the transducer.

[0027] The output unit may be configured to provide / present at least one stimulus or a plurality of consecutive stimuli to the subject as tactile or visual stimulation.

[0028] The output unit may be configured to present acoustic stimuli.

[0029] The output unit may be configured to present an electrical stimulus.

[0030] The output unit may be configured to present tactile stimulation.

[0031] The output unit may be configured to present a visual stimulus.

[0032] For example, auditory nerve responses can be induced using cochlear implants (electrically) or bone conductor devices.

[0033] For example, the transducer may include a receiver (speaker) for presenting the stimulus as an acoustic signal to the subject.

[0034] The output unit may include an ear probe, a (separate) speaker, or a headset. A headset may refer to an in-ear, on-ear, or on-head headset, earphone, etc., which is configured to introduce audio into the user's ear.

[0035] The apparatus may comprise at least one recording unit comprising one or more sensors for placement on the scalp of a subject.

[0036] For example, the sensor may comprise an electrode or an optode or a magnetometer or a microphone.

[0037] Thus, the apparatus may comprise at least one recording unit comprising one or more electrodes or optodes or magnetometers or microphones for placement on the scalp of the subject.

[0038] Thus, the one or more electrodes or optodes or magnetometers or microphones may be configured to measure one or more evoked responses of the subject.

[0039] One or more evoked responses may be measured in response to at least one stimulus or a plurality of successive stimuli being provided.

[0040] The evoked response can be an electrophysiological response (i.e., auditory evoked neural response), a visual evoked potential, a sound evoked response (such as OAE), etc.

[0041] The one or more electrodes may be configured to measure one or more electrophysiological responses of the subject.

[0042] The one or more electrodes may be configured to measure one or more electrophysiological responses of the subject in response to the at least one stimulus or a plurality of sequential stimuli provided.

[0043] In other words, the one or more sensors may be configured (or prompted, for example, by the control unit) to measure the response after the stimulus is provided / presented to the ear of the subject (immediately or with some delay).

[0044] For example, one or more sensors may refer to any number of sensors from 1 to several hundred sensors.

[0045] For example, one or more optodes may be used in functional near-infrared spectroscopy (fNIRS).

[0046] For example, one or more magnetometers may be used in magnetoencephalography (MEG).

[0047] At least one recording unit may be configured to receive (and record) one or more evoked responses measured via one or more electrodes.

[0048] For example, the response may be measured / recorded via electroencephalography (EEG) using one or more electrodes placed on the subject's scalp.

[0049] For example, an instrument comprising one or more sensors may be used in any EEG system / instrument and may be extended to any other domain where responses are obtained by averaging responses measured to repeatedly presented stimuli (e.g., vestibular evoked responses, evoked otoacoustic emissions, magnetoencephalography, etc.) (i.e., afferent data).

[0050] The apparatus may comprise an analytical unit.

[0051] The analysis unit may be configured to receive the measured one or more evoked responses.

[0052] The analyzing unit may be configured to receive the measured one or more evoked responses from the recording unit.

[0053] The analyzing unit may be configured to analyze the measured one or more evoked responses.

[0054] The analysis unit may be configured to determine, during operation of the instrument (eg, on-line), a probability p that each of said reactions is driven by underlying background noise.

[0055] The analysis unit may be configured to determine said probability p based on an F value of the measured one or more evoked responses determined as a ratio between a variance of an average of the one or more evoked responses and a variance of the residual background noise, each variance having statistical degrees of freedom (DOF) v and r, respectively.

[0056] In other words, the analyzing unit may be configured to estimate the probability p (ie the likelihood) that the measured response is a result of underlying background noise in the measured response and not induced by the stimulus presented in the ear of the subject.

[0057] In other words, the analyzing unit may be configured to determine the probability from an F-value of the measured one or more evoked responses determined as the ratio between the variance of the average evoked response and the variance of the residual background noise.

[0058] More specifically, the ratio may be the ratio between the variance of the mean of one or more evoked responses (with statistical degrees of freedom (DOF) v) and the variance of the residual background noise (with statistical DOF r).

[0059] Thereby, an improved objective detection of evoked responses is provided, which is able to take into account background noise (obtained from the subject) during the measurement and reduces the time required for detection / measurement.

[0060] The provided solution can be implemented in real time and leads to better estimation of the underlying background noise than prior art techniques.

[0061] The provided solution is applicable to any time-domain based evoked responses (brainstem responses and cortical responses) and can be used for diagnosis with higher accuracy and improved detection time than existing solutions.

[0062] Furthermore, the provided solution can be applied to any frequency-domain based evoked response (brainstem responses and cortical responses, e.g., auditory steady-state responses (ASSRs), frequency following responses (FFRs), otoacoustic emissions (OAEs), and others) and can be used for diagnosis with higher accuracy and improved detection time than existing solutions.

[0063] In other words, the determination of the above-mentioned probability p can be done during operation of the instrument.

[0064] One or more sensors (electrodes, optodes, magnetometers, or microphones) may be configured to measure one or more evoked responses including a ratio of the variance of a mean response evoked by a stimulus to the variance of the residual background noise.

[0065] In other words, the evoked response measured after presentation of a stimulus may reflect the potential background noise (signal) combined with the response evoked by the stimulus (in the case where it is perceived by the subject). Thus, the measured evoked response may partially include the response evoked by the stimulus and partially include the potential background noise, wherein the term partially covers a percentage of 0 or higher.

[0066] For example, we can measure / record evoked responses (e.g., EEG data) in response to presented sounds. During each period of presented sounds, a trial of duration T can be recorded using one or several electrodes.

[0067] A test may refer to the presentation of a stimulus to the ear of a subject and the subsequent measurement of the evoked response.

[0068] Thus, the measured one or more evoked responses may have a duration T.

[0069] For example, the stimulus may be presented 1000 times repeatedly every 40 ms (e.g. 0 ms, 40 ms, 80 ms, ...). The duration of the evoked response may be set to be 20 ms long, after each stimulus onset. In this case, there will be 1000 individual trials, each lasting 20 ms.

[0070] Each trial from a continuous data stream can be denoted by s 1 (t), s 2 (t), ..., s N (t), and can be obtained for analysis, where N is the total number of trials recorded. We are interested in obtaining the stimulus-evoked response (i.e., the evoked neural response) X(t), which, for simplicity, can be assumed to be the same for each stimulus presented. Thus, the problem can be expressed as follows:

[0071]

[0072] Here, X(t) is the stimulus-evoked response (signal), BN i (t) is the potential background noise (signal), both of which are functions of time t.

[0073] To reduce potential background noise, the average evoked response of all individual trials was estimated by:

[0074]

[0075] where RN(t) is the residual background noise in the average response (ie, residual noise), and w corresponds to the vector of estimated weights used to average the data.

[0076] Time Domain

[0077] In the case of standard averaging, in the time domain, w is given by:

[0078]

[0079] In the case of weighted averaging, w is given by:

[0080]

[0081] in, is the variance of the potential background noise in the measurement results (recordings).

[0082] Variance of underlying background noise It can be estimated as 1) the average variance of the individual variances at a fixed time point tracked across trials, or 2) the variance of multiple points within each trial.

[0083] For the first method, It is estimated from a subset of trials by:

[0084]

[0085] Among them, N k For calculation Here, it is used to calculate N k Each of the trials is assumed to be recorded under the same quasi-static underlying background noise, i.e., N k Each of the trials is Weighted.

[0086] For the second method, each test Calculate using the following formula:

[0088]

[0089] where P is the number of points used to estimate the potential background noise and can include from 1 to all points within a trial. If the neural activity is much smaller than the potential background noise, δ can be 0; or, if the neural response is expected to be large at the trial level, δ can be 1. For this method, each trial has its own

[0090]

[0091] Once you have obtained Therefore, the weight (w s or w ), the residual background noise in the average evoked response can be estimated by:

[0092]

[0093] Now, the signal-to-noise ratio (SNR) can be estimated by:

[0094]

[0095] in, is the average record (i.e., the average evoked response) The variance of and The ratio between.

[0096] The significance of SNR now depends entirely on the F value, which is expressed in the numerator The statistical degrees of freedom (DOF) (v) and the denominator The DOF(r) is determined.

[0097] Thus, the present solution provides a novel estimation of these parameters.The provided solution estimates these DOFs (v and r) for each individual measurement online while the data is being recorded, ie during operation of the instrument.

[0098] The analysis unit may be configured to determine the probability based at least on a statistical degree of freedom (DOF) of the variance. The probability may be determined during operation of the instrument.

[0099] In other words, the present application enables the statistical DOF v to be estimated online.

[0100] The analyzing unit may be configured to determine the probability based at least on a statistical DOF of a variance of an average of the one or more evoked responses and a statistical DOF of a variance of a residual background noise measured in the one or more evoked responses.

[0101] The analysis unit may be configured to determine the F value of the measured one or more evoked responses as a ratio between a variance of an average of the one or more evoked responses and a variance of the residual background noise, each variance having a statistical DOF v and r, respectively.

[0102] For example, without going into formal mathematical proofs, for the numerator of the F ratio, the following DOF is derived:

[0103]

[0104]

[0105] Where v is DOF, N ω for The number of frequency bins, w i is the angular frequency of each frequency bin.

[0106] is the power spectral density and is obtained by equation (10), which corresponds to the weighted average of the squared magnitude of the Fourier transform of each trial.

[0107]

[0108] where δ is 0 if the neural activity is much smaller than the underlying background noise, or 1 if the neural response is expected to be large at the trial level.

[0109] Weight w i It is given by equation (3) or equation (4).

[0110] The DOF of the denominator is given by:

[0111]

[0112] Where P is the number of tracking points used to estimate the background noise, fs is the sampling rate of the recorded data, and T is the duration of each trial.

[0113] Notice, A longer time window T′ can be used for estimation, and then scaled down for a smaller analysis window of length T:

[0114]

[0115] v′ is calculated using equation (9).

[0116] Once the DOF of the numerator and denominator have been obtained, the probability that the calculated F value is driven by the underlying background noise can be calculated as follows:

[0117] p=1-F CDF (F, v, r) (14)

[0118] where FCDF(F, v, r) is the cumulative distribution function of an F-distributed random variable with DOFs v and r.

[0119] Thus, the analysis unit may be configured to determine the probability p by:

[0120] p=1-F CDF (F, v, r)

[0121] Among them, F CDF (F,v,r) is the cumulative distribution function of an F-distributed random variable with v and r, v is the statistical DOF of the variance of the mean of the (measured) evoked response or responses, and r is the statistical DOF of the variance of the residual background noise.

[0122] The provided solution can measure the F ratio (F value, i.e., the ratio of the energy of the average measurement result to the energy of the estimated residual background noise). Ongoing activities can be weighted based on the changes in the potential background activity. These weights, obtained by consistent tracking across multiple points in the experiment and / or within the experiment being measured, can be used to estimate the energy of the residual background noise in the measurement result (the denominator of the F ratio). The energy in the numerator can be estimated from the variance of the average evoked response in progress. The provided solution uses a new method to obtain the correct statistical degrees of freedom (DOF), which determines the sensitivity of finding statistically significant evoked responses. In addition, the provided solution can be used in real time and its sensitivity can be automatically adjusted for each individual situation, thereby providing the best personalized accuracy. Since the F value is a parametric quantity, with a DOF parameter in the numerator of the ratio (referred to as v here) and a DOF parameter in the denominator (referred to as r), its performance depends entirely on the estimation accuracy of these two parameters.

[0123] Frequency Domain

[0124] In the case of standard averaging, in the frequency domain, w is given by:

[0125]

[0126] In the case of weighted averaging, w is given by:

[0127]

[0128] in is the variance of the potential (residual) background noise in the measurement results (records).

[0129] An important property that will be used in the following equations is that the variance of a signal in the time domain can be estimated in the frequency domain by means of Parseval's theorem as the sum of the power spectral densities of all positive frequencies. This is shown below:

[0130]

[0131] Where T is the duration of a trial (or a period), refers to the Fourier transform, and P is the number of samples or sample subsets within each trial (if the required time window is smaller than an epoch).

[0132] Using this property, the variance of the frequency-specific underlying background noise It can be estimated in the frequency domain as the sum of the power spectral density components at the frequencies of interest (e.g., angular frequencies):

[0133]

[0134] Where T is the duration of the test, refers to Fourier transform, P is the number of samples or sample subsets in each trial (period) (if the required time window is less than one period), ω j ∈[ω 1 ,ω 2 ,…,ω L ] is the set of angular frequencies of interest. δ can be 0 if the evoked activity is much smaller than the background noise, or 1 if the evoked response is expected to be large at the trial level.

[0135] If all frequencies are used to estimate the variance in Equation 18, we will obtain the same variance as in the time domain. However, if only a subset of the frequency components are used, we will obtain the variance due only to those frequencies, i.e., the frequency-specific variance.

[0136] Once you have obtained Therefore, the weight (w s or w ), the residual background noise in the average response can be estimated by:

[0137]

[0138] Now, the signal-to-noise ratio (SNR) can be estimated by:

[0139]

[0140] in, is the average record (i.e., the average evoked response) given by The frequency-specific variance of :

[0141]

[0142] Where T′ is the duration of the analysis window (with P′ samples), and F is and The ratio between.

[0143] The significance of SNR now depends entirely on the F value, which is expressed in the numerator The statistical DOF(v) and the denominator The DOF(r) is determined.

[0144] The novelty of the proposed solution relies on the estimation of these parameters.The analysis unit can estimate these DOFs (v and r) for each individual recording / measurement result online while the data is being recorded, from the specific frequency of interest.

[0145] The DOF of the numerator of the F value (i.e. F ratio) can be obtained as follows:

[0146]

[0147]

[0148] Where v is For the angular frequency ω of interest i DOF.

[0149] is the power spectral density and is obtained by equation 23, which corresponds to the weighted average of the magnitude of the Fourier transform of each trial:

[0150]

[0151] where δ can be 0 if the neural activity is much smaller than the underlying background noise, or 1 if the neural response is expected to be large at the trial level.

[0152] Weight ω i It is given by Equation 15 or Equation 16.

[0153] The DOF of the denominator is given by:

[0154]

[0155] It should be noted that in many applications, the The duration of the analysis window (P′ samples) will be the same as the one used to estimate (P samples) to ensure the same frequency resolution for F-value estimation.

[0156] Once the DOF of the numerator and denominator have been obtained, the probability that the calculated F value is driven by the underlying background noise can be calculated as follows:

[0157] p=1-F CDF (F, v, r) (26)

[0158] Among them, F CDF (F, v, r) is the cumulative distribution function of an F-distributed random variable with DOFs v and r.

[0159] Thus, the analysis unit may be configured to determine the probability p by:

[0160] p=1-F CDF (F, v, r)

[0161] Among them, F CDF (F,v,r) is the cumulative distribution function of an F-distributed random variable with v and r, v is the statistical DOF of the variance of the mean of the (measured) evoked response or responses, and r is the statistical DOF of the variance of the residual background noise.

[0162] There are many applications where one or a few frequencies need to be evaluated to assess whether their energy is significantly greater than the energy of the residual background noise, such as ASSR, steady-state visual evoked potentials, etc.

[0163] To deal with this problem, the equation given above can be used. However, if the number of frequencies to be tested is small (i.e. a small sparse number of frequencies of interest), the estimate of the residual background noise may be inaccurate due to the limited number of data points (frequency samples) contributing to the background noise estimate. To address this problem, the following adjustment can be used.

[0164] For example, consider a situation where there are L sparsely distributed frequencies of interest (ω 1 ,ω 2 , ..., ω L ).

[0165] Next, consider that we want to evaluate whether the combined energy of these frequencies is significantly greater than the potential background noise at the same frequencies.

[0166] Since a low number of frequencies will be analyzed, the estimate of the background noise variance (Equation 18) may not be accurate relative to the full spectrum of the data. To avoid this problem, we can use the assumption that the energy of the frequencies of interest should be similar to the energy of their neighboring frequencies. Under this assumption, we can obtain a good approximation of the energy of the potential background noise at each frequency of interest by averaging the energy of their neighboring frequencies.

[0167] Right now:

[0168]

[0169] in, For estimation The number of frequency bins around each frequency of interest.

[0170] Note that in Equation 27, the number of samples, and thus the number of DOFs, will be higher since more samples contribute to the estimate.

[0171] The number of DOF given by Equation 25 will now include all frequencies used in Equation 27 to estimate the variance of the underlying background noise. That is, r can be determined by:

[0172]

[0173] Thus, Equation 26 can again be used to estimate the probability that the calculated F value is driven by residual background noise (and therefore by latent background noise).

[0174] In other words, in the case where the instrument is used in ASSR measurement, Equation 27 can be used to estimate For example, if the response frequency is 40Hz, It can be estimated as the average energy in the frequency interval 40Hz±5Hz.

[0175] The provided solution can also measure the F ratio (ratio of the energy of the average measurement result to the energy of the estimated residual background noise) in a frequency-specific manner. The ongoing activity can be weighted based on the changes in the underlying background activity. These weights can be derived from the energy in the frequency domain and can be used to estimate the frequency-specific energy of the residual background noise in the measurement result / recording (the denominator of the F ratio). The energy in the numerator can be estimated in the frequency domain from the energy of the ongoing average reaction. The provided solution can estimate the correct statistical degrees of freedom (DOF) based on the frequency components of interest and thus determine the sensitivity for finding statistically significant evoked reactions. In addition, the provided solution can be used in real time and its sensitivity is automatically adjusted for each individual situation, thereby providing the best personalized accuracy. Since the F value is a parametric quantity, with a DOF parameter in the numerator of the ratio (referred to as v here) and a DOF parameter in the denominator (referred to as r), its performance is completely dependent on the estimation accuracy of these two parameters.

[0176] Thus, the analysis unit may be configured to convert the variance of the frequency-specific potential background noise into is estimated as the sum of the power spectral density components at the frequencies of interest (eg, angular frequencies).

[0177] The analysis unit can be configured to be based on the variance of the frequency specific underlying background noise and the estimated weight vector (w s or ω ) estimates the residual background noise. In other words, the residual background noise variance in the average response It can be estimated by Equation 7 or Equation 19.

[0178] The analysis unit may be configured to estimate the energy of the residual background noise at each frequency of interest by averaging the energy of the residual background noise at neighboring (adjacent) frequencies of the frequency of interest.

[0179] The analysis unit can be configured based on The DOF is for the angular frequency ω of interest i Determine the numerator of the F value (see Equation 9 (time domain) and Equation 22 (frequency domain)).

[0180] The analysis unit can be configured based on The DOF determines the denominator of the F value for the angular frequency of interest (see Equation 12 (time domain) and Equations 25 and 28 (frequency domain)).

[0181] The analysis unit can be configured based on and The ratio of estimates SNR and thus estimates the F value (see equation 20).

[0182] For example, It can be determined from equation 21(2).

[0183] For example, It can be determined from Equation 19.

[0184] For example, the frequency of interest for CAP measurements may be 2 Hz to 30 Hz.

[0185] For example, the frequencies of interest for ASSR measurements may be 40 Hz, 80 Hz, and other harmonics.

[0186] The analysis unit may be configured to adaptively determine said probability during operation of the instrument.

[0187] The probability may be determined in the time domain and / or in the frequency domain.

[0188] Thus, a rapid and continuous probability determination is provided to the instrument user while the instrument is operating.

[0189] The at least one stimulus or the plurality of consecutive stimuli may comprise an acoustic stimulus.

[0190] At least one stimulus or the plurality of consecutive stimuli may comprise a click.

[0191] At least one stimulus or the plurality of consecutive stimuli may comprise a narrowband chirp.

[0192] At least one stimulus or the plurality of consecutive stimuli may comprise a speech signal.

[0193] The at least one stimulus or the plurality of consecutive stimuli may comprise electrical stimulation or any repetitive auditory stimulation.

[0194] For example, at least one stimulus or a plurality of consecutive stimuli may be based on an auditory change complex (ACC) stimulus.

[0195] The one or more evoked responses measured may include auditory brainstem responses (ABRs).

[0196] The one or more evoked responses measured may include cortical auditory potentials (CAPs).

[0197] The one or more evoked responses measured may include an auditory steady-state response (ASSR).

[0198] For example, the one or more electrophysiological responses measured may include any response that relies on averaging to reduce background noise when detecting a target signal, such as otoacoustic emissions, visual evoked responses, vestibular responses, and the like.

[0199] The analysis unit may be configured to determine the hearing ability of the subject in response to the probability being below a predetermined threshold.

[0200] The predetermined threshold may be, for example, the minimum stimulus amplitude at which a sound will be audible, the minimum difference in acoustic features for distinguishing between sounds, the minimum difference for discerning speech in the presence of noise, the minimum difference for discerning the spatial location of sounds, or the minimum difference for detecting anomalous sounds from repetitive sounds.

[0201] The stimulus generator may be configured to modify at least one stimulus or a plurality of consecutive stimuli in response to the determined probability.

[0202] The stimulus generator may be configured to modify the test protocol in response to the determined probability.

[0203] The analyzing unit may be configured to determine the stimulus or a plurality of consecutive stimuli based on an analysis of previously measured evoked responses.

[0204] The analyzing unit may comprise a neural network. The analyzing unit may be configured to determine the probability p using the neural network.

[0205] Detector unit

[0206] In one aspect, the present application also provides a detector unit.

[0207] The detector unit may be configured to determine a probability p that a measured evoked response is driven by underlying background noise (signal).

[0208] For example, the analyzing unit described above may comprise the detector unit.

[0209] The detector unit may be configured to determine said probability p based on an F value of the measured evoked response, the F value being determined as the ratio between the variance of the mean evoked response and the variance of the residual background noise, each variance having statistical degrees of freedom (DOF) v and r, respectively.

[0210] In other words, the detector unit may be configured to determine the probability p by:

[0211] p=1-F CDF (F, v, r)

[0212] Among them, F CDF (F,v,r) is the cumulative distribution function of an F-distributed random variable with v and r, v is the statistical DOF of the variance of the mean of the (measured) evoked response, r is the statistical DOF of the variance of the residual background noise.

[0213] use

[0214] In one aspect, there is provided the use of an apparatus as described above and defined in the claims.

[0215] method

[0216] In one aspect, the present application also provides a method for detecting an induced response.

[0217] For example, the method may be a method of detecting auditory evoked neural responses.

[0218] The method may also include estimating the subject's hearing ability.

[0219] The method may include generating at least one stimulus according to a testing protocol.

[0220] The method may include generating a plurality of sequential stimuli according to a testing protocol.

[0221] The method may include providing the at least one stimulus to the subject.

[0222] The method may include providing the at least one stimulus via a transducer into an ear of the subject.

[0223] The method may include providing the plurality of sequential stimuli to the subject.

[0224] The method may include providing the plurality of sequential stimuli to an ear of the subject via a transducer.

[0225] The method may include placing one or more electrodes on the scalp of the subject.

[0226] The method may include placing one or more optodes on the scalp of the subject.

[0227] The method may include placing one or more microphones in the ear of the subject.

[0228] The method may include placing one or more magnetometers on the scalp of the subject.

[0229] The method may include measuring one or more evoked responses in the subject.

[0230] One or more evoked responses of the subject may be measured / recorded in response to the provision of at least one stimulus or a plurality of successive stimuli.

[0231] The method may include receiving, by an analysis unit, the measured one or more induced responses.

[0232] The method may include analyzing the measured one or more evoked responses by an analyzing unit.

[0233] The analysis unit may determine during instrument operation the probability p that each of the reactions is driven by residual (latent) background noise.

[0234] The analysis unit may determine said probability p based on an F value of the measured one or more evoked responses, the F value being determined as the ratio between the variance of the mean of the one or more evoked responses and the variance of the residual background noise.

[0235] The analysis unit may determine said probability p based on an F value of the measured one or more evoked responses, the F value being determined as the ratio between the variance of the mean of the one or more evoked responses and the variance of the residual background noise, each variance having statistical degrees of freedom v and r, respectively.

[0236] For example, the analyzing unit may determine the probability by at least equations 7, 8, 9, 12 and 14 (in the time domain).

[0237] For example, the analyzing unit may determine the probabilities by at least equations 19, 20, 21, 22, 25, 28, and 26 (in the frequency domain).

[0238] When appropriately replaced by corresponding procedures, some or all structural features of the apparatus described above or defined in the claims may be combined with the implementation of the method of the invention, and vice versa. The implementation of the method has the same advantages as the corresponding apparatus.

[0239] Computer readable medium or data carrier

[0240] The present invention further provides a tangible computer-readable medium (data carrier) storing a computer program including program code (instructions), which, when the computer program is run on a data processing system (computer), causes the data processing system to perform (implement) at least part (such as most or all) of the steps of the method described above and defined in the claims.

[0241] Computer Programs

[0242] The present application further provides a computer program (product) comprising instructions, which, when executed by a device (eg an instrument, a detector of an instrument, or a computer), cause the device to perform (the steps of) the method described above and defined in the claims.

[0243] Data processing system

[0244] In one aspect, the present application further provides a data processing system comprising a processor and a program code, wherein the program code causes the processor to execute at least part (such as most or all) of the steps of the method described above and defined in the claims.

[0245] system

[0246] In another aspect, there is provided a system comprising an apparatus as described above and defined in the claims and comprising an auxiliary device.

[0247] The system may be adapted to establish a communication link between the instrument and an auxiliary device so that information (eg, control and / or status signals) may be exchanged or forwarded from one device to another.

[0248] The auxiliary device may include a remote controller, a mobile device, a smart phone, or other portable electronic device such as a tablet computer, or may include a server device, etc.

[0249] The communication link may be a wired or wireless link, for example based on Bluetooth or some other standardized scheme.

[0250] APP

[0251] In another aspect, the present invention also provides a non-transient application referred to as an APP. The APP may include executable instructions configured to be run, for example, on the instrument or on an auxiliary device to implement, for example, a control interface. BRIEF DESCRIPTION OF THE DRAWINGS

[0252] Various aspects of the present invention will be best understood from the detailed description below in conjunction with the accompanying drawings. For clarity, the drawings are schematic and simplified, and only the details necessary for understanding the present invention are given, while other details are omitted. Throughout the specification, the same reference numerals are used for the same or corresponding parts. The various features of each aspect may be combined with any or all features of the other aspects. These and other aspects, features and / or technical effects will be apparent from and illustrated in conjunction with the following figures, in which:

[0253] Figure 1 An example of an instrument for detecting auditory evoked neural responses is shown;

[0254] Figure 2 shows an example of measurement results and recording of electrophysiological responses obtained by a recording unit comprising four electrodes;

[0255] Figure 3 shows an example of the distribution of p-values ​​as a function of trials in the presence of Gaussian noise;

[0256] Figure 4 shows an example of the distribution of F values ​​as a function of trials in the presence of Gaussian noise;

[0257] Figure 5 shows an example of the distribution of F values ​​as a function of trials in the presence of Gaussian noise and simulated "neural" signals;

[0258] Fig. 6A and 6BAnother example of the distribution of p-values ​​as a function of trials in the presence of Gaussian noise is shown;

[0259] Fig. 7A and 7B Another example of the distribution of p-values ​​as a function of trials in the presence of Gaussian noise is shown;

[0260] Fig. 8A and 8B An example of detection rate as a function of epoch number in the presence of simulated ASSR is shown.

[0261] By the detailed description given below, the further scope of application of the present invention will be apparent. However, it should be understood that while the detailed description and specific examples show the preferred embodiments of the present invention, they are only provided for illustrative purposes. For those skilled in the art, based on the following detailed description, other embodiments of the present invention will be apparent. DETAILED DESCRIPTION

[0262] The detailed description proposed below in conjunction with the accompanying drawings is used as a description of a variety of different configurations. The detailed description includes specific details for providing a thorough understanding of a plurality of different concepts. However, it will be apparent to those skilled in the art that these concepts can be implemented without these specific details. Several aspects of the apparatus, system, and method are described by a plurality of different blocks, functional units, modules, components, steps, processes, algorithms, etc. (collectively referred to as "elements"). Depending on the specific application, design limitations, or other reasons, these elements may be implemented using electronic hardware, computer programs, or any combination thereof.

[0263] Figure 1 An example of an instrument for detecting auditory evoked neural responses is shown.

[0264] exist Figure 1 In FIG. 2 , it is shown that the apparatus 1 may include a stimulation generator SG, an output unit OU, a recording unit RU and an analysis unit AU. The stimulation generator SG, the output unit OU, the recording unit RU and the analysis unit AU may be located in the base unit 2 .

[0265] The stimulus generator SG may be configured to generate at least one stimulus 3 or a plurality of consecutive stimuli 3 according to a test protocol.

[0266] The output unit OU can be connected to the stimulation generator SG. The output unit OU can include a converter. Figure 1 In the embodiment, the transducer is arranged in an ear probe 4 and may comprise a receiver (loudspeaker) for providing the stimulus 3 as an acoustic signal to an ear 5 of a subject 6 .

[0267] The output unit OU may be configured to receive the stimuli 3 from the stimulus generator SG and to provide the stimuli 3 via the transducer into the ear 5 of the subject 6. It has been shown that a plurality of consecutive stimuli 3 may be presented in the ear canal of the subject 6 via the ear probe 4.

[0268] exist Figure 1 In FIG. 6 , the ear probe 4 is shown inserted into the right ear of the subject 6, however, it is contemplated that either the right ear or the left ear may be tested, or both ears may be tested simultaneously.

[0269] The recording unit RU may include one or more electrodes 7a, 7b, 7c, 7d, which are shown arranged / placed on the scalp of the subject 6. The one or more electrodes 7a, 7b, 7c, 7d may measure one or more electrophysiological responses 8 of the subject 6 (during time period T) in response to one or more stimuli 3 presented to the ear 5 of the subject 6. The recording unit RU may record the electrophysiological responses 8.

[0270] The analyzing unit AU may be configured to receive one or more electrophysiological responses 8 measured (and recorded) from the recording unit RU (or, as an alternative, directly from the electrodes 7a, 7b, 7c, 7d) (see solid line between RU and AU).

[0271] The analysis unit AU may further be configured to receive data / information about the stimulation 3 (eg time, type etc.) from the stimulation generator SG (see solid line between SG and AU).

[0272] The analyzing unit AU may be configured to analyze one or more electrophysiological reactions 8 measured by the electrodes 7a, 7b, 7c, 7d. The analysis may be based on said data / information about the stimulation 3 received, for example, from the stimulation generator SG.

[0273] The analysis unit AU may be configured to determine a probability p that each of said reactions is (or is not) driven by the underlying background noise. The determination may also be based at least on the variance of the one or more electrophysiological reactions 8 measured.

[0274] exist Figure 1 In FIG. 2 , it is shown that the instrument 1 may additionally comprise a processor PR (eg located in the base unit 2 ). The processor PR may be configured to control the operation of the instrument 1 , in other words, to control the execution of the test protocol and the analysis of the measured data.

[0275] As indicated by dotted lines, the processor PR may be connected at least to the stimulation generator SG, the recording unit RU and the analyzing unit AU.

[0276] As indicated by the dotted lines, the output unit OU may transmit the stimulus 3 to the ear probe 4 via a wired or wireless connection 9 , and the recording unit RU may receive the measured electrophysiological response 8 from the electrodes 7a , 7b , 7c , 7d via a wired or wireless connection 10 .

[0277] Figure 2 An example of measurement results and recording of electrophysiological responses obtained by a recording unit comprising four electrodes is shown.

[0278] exist Figure 2 In the embodiment, the electrophysiological response may be EEG data. The EEG data may be measured by four electrodes 7a, 7b, 7c, 7d arranged on the scalp of the subject 6, which is substantially similar to Figure 1 The example shown in .

[0279] Thus, four channels can be formed from the electrodes 7a, 7b, 7c, 7d to the recording unit RU for transmitting EEG data. The time series of EEG data is illustrated in the figure in the recording unit RU.

[0280] The measured electrophysiological response 8 (ie the target neural response) is graphically illustrated in the upper part of the figure by the (black) waveform X(t).

[0281] The measured and recorded data of each channel, contaminated by potential background noise, is shown by respective colored traces, i.e., the first (blue, CH1) and third (cyan, CH3) curves measured from top to bottom at the left ear and the second (red, CH2) and fourth (orange, CH4) curves measured from top to bottom at the right ear. The electrodes 7a, 7b arranged at the left ear may represent channels CH1 and CH3, and the electrodes 7a, 7b arranged at the right ear may represent channels CH2 and CH4. The electrodes 7c arranged at the side and the electrode 7d arranged at the top of the scalp may represent the ground and reference channels / electrodes.

[0282] The onset of new stimulus presentation is indicated in the figure by a vertical line, from which the trial (s 1 (t),s 2 (t),…,s N (t)) has a duration T.

[0283] P(P 1 …P m ) denotes multiple (m) points being tracked and used to determine the probability p.

[0284] Figure 3 Shown is the distribution of p-values ​​as a function of trials in the presence of Gaussian noise.

[0285] Since the F-value is estimated using a number of (measurement) points m and the DOF (which characterizes the potential background noise of each recording), the F-value determination is named Fmpi.

[0286] To illustrate the accuracy of the Fmpi detector (or detection method), a total of 10,000 simulations were obtained, each with 360 trials. The simulations were performed using a typical setup for cortical response recordings. There were no neural responses and the underlying background noise in the simulated recordings was Gaussian noise. To compare the performance of the Fmpi detector, we processed the same dataset using a standard single-point F detector [1][2] with a standard mean (Fsp-sa). This detector uses a single point to estimate the residual noise of the F ratio and assumes DOF ​​of r=N (total number of trials) and v=5 (based on empirical analysis of human EEG noise).

[0287] exist Figure 3 In this paper, three different instruments (including three different detectors / methods for estimating F values) are used. The three different detectors (methods / algorithms) include Fmpi using weighted averaging (Fmpi-wa), Fmpi using standard averaging (Fmpi-sa), and Fsp using standard averaging with v=5 [1][2]. The trial length is 1 second, the sampling rate is 500Hz, the analysis window is 300ms, and the data is bandpass filtered between 2 and 30Hz. All points within a trial are used by Fmpi. Using the Fmpi detector, the p-values ​​are evenly distributed across all trials (which is the ultimate goal of correctly estimating the underlying noise distribution).

[0288] Figure 4 An example of the distribution of F values ​​as a function of trials in the presence of Gaussian noise is shown.

[0289] exist Figure 4 In , a histogram of the obtained F values ​​(statistics, 11) is shown in gray. The probability density functions estimated using three detectors (Fmpi using weighted average (Fmpi-wa), Fmpi using standard average (Fmpi-sa), and Fsp[1][2] using standard average with v=5) are shown (blue, 12). The empirical probability distribution is shown in red 13. The trial length is 1 second, the sampling rate is 500 Hz, the analysis window is 300 ms, and the data is bandpass filtered between 2 and 30 Hz. All points within the trial are used by Fmpi.

[0290] With the Fmpi detector using the standard average (Fmpi-sa, Equation 3) or the weighted average (Fmpi-wa, Equation 4), the estimated F distribution fits the empirical distribution of the data almost identically. On the other hand, with the classical F detector, a skewed distribution ( Figure 3 , Fsp-sa) and poor compliance ( Figure 4 , Fsp-sa).

[0291] Figure 5An example of the distribution of F values ​​as a function of trials in the presence of Gaussian noise and a simulated "provoking" signal is shown.

[0292] The simulation is similar to the previous example ( Figure 3 and Figure 4 ) but now including the target "evoked" signal. The results show that the estimate of the underlying distribution of the noise is the same as that obtained when the "neural" response is not included (compare black solid and dashed lines). The Fmpi detector correctly estimates the 5% chance point of false positive detections (vertical line), leading to earlier and more accurate detections than the standard Fsp-sa detector, for which the "5%" chance point is overestimated (compared to the empirical distribution), leading to fewer detections at a lower number of trials.

[0293] For comparison, for all three detectors, namely Fmpi using weighted average (Fmpi-wa), Fmpi using standard average (Fmpi-sa), and Fsp using standard average with v=5 [1][2], the histogram of F values ​​(statistics) obtained in the absence of target signal is shown in light blue 14, and the histogram of F values ​​obtained in the presence of target "evoked" signal is shown in red 15. The noise potential distribution obtained in the absence of target neural signal is marked by the black dotted line 16, and the distribution obtained in the presence of target neural signal is marked by the black solid line 17. The vertical line shows the critical F value for determining the presence of a response. The values ​​on the right are shown as detected responses. The trial length was 1 second, the sampling rate was 500 Hz, the analysis window was 300 ms, and the data were bandpass filtered between 2 and 30 Hz. The predetermined average target SNR of 360 trials was 6 dB. All points within the trial were used by Fmpi.

[0294] Fig. 6A and 6B Another example of the distribution of p-values ​​(probability values) as a function of trials in the presence of Gaussian noise is shown.

[0295] To illustrate the accuracy of the F value (and probability) determined in the frequency domain (e.g. by a detector unit), a typical recording time of 200 epochs was simulated. The probability of detecting a response in an interval of 20 epochs was estimated. Each interval was performed 1000 times, totaling 10000 random trials in the whole simulation. The simulation was performed using a typical setup for cortical response recordings. Neural responses were not included so that it could be demonstrated how well the properties of the Gaussian underlying background noise could be predicted.

[0296] Fig. 6A The p-value distributions for these simulations are shown. The p-values ​​are evenly distributed across all trials (which is the ultimate goal of correctly estimating the underlying distribution).

[0297] Figure 6BThe distribution of F values ​​(statistics) in the presence of Gaussian noise is shown as a function of the trial. The histogram of the obtained F values ​​(statistics) is shown in gray 19. The probability density function estimated with the modified detector is shown in blue (line 20). The empirical probability density is shown in red (line 21). The trial length is 1 second, the sampling rate is 500 Hz, the analysis window is 500 ms, and the data is bandpass filtered between 2 and 30 Hz. The estimated F distribution conforms to the empirical distribution of the data almost identically.

[0298] Fig. 7A and 7B Another example of the distribution of p-values ​​as a function of trials in the presence of Gaussian noise is shown.

[0299] exist Fig. 7A In , the probability distribution (frequency-specific version; frequency domain) was tested by simulating auditory steady-state responses (ASSRs) by convolving the template response at a rate of 41 Hz in the presence of Gaussian noise.

[0300] The results of the simulations are compared using the proposed approach (frequency domain) and a commonly used method for this type of analysis, namely the Hotelling-T-squared (HT2) test.

[0301] Both tests found that the composite power of 12 frequencies (multiples of 41 Hz) was significantly greater than the composite power of the underlying background noise. Both simulations were first performed without "evoked responses", i.e., with only noise ( Fig. 7A and 7B ), and then in the presence of a simulated evoked response ( Fig. 8A ).

[0302] The results of noise simulations show that both our solution and the HT2 test predict the underlying distribution well. However, the HT2 method does not produce reliable estimators when the number of frequencies of interest is larger than the number of trials included, which limits its use, especially for smaller numbers of trials.

[0303] exist Fig. 7A In FIG, the distribution of p-values ​​as a function of trials in the presence of Gaussian noise using a frequency specific implementation of the proposed method (Fmpi-FFT) (red, 22) and the Hotelling T-squared test (HT2; blue, 23) are shown. Figure 7BThe F value distribution of Fmpi-FFT test and HT2 test in the presence of Gaussian noise is shown as a function of the test. The histogram of the obtained F value (statistic) is shown in gray (24). The probability density function estimated with Fmpi-FFT and HT2 is shown in blue (line 25). The empirical probability distribution is shown in red (line 26). The HT2 method provides reliable potential distribution estimation when higher than 30 trials, but fails to provide reliable estimators when lower than 30 periods, so those data points are not included. The trial length is 1 second, the sampling rate is 2000Hz, the analysis window is 1000ms, and the data are not filtered.

[0304] Fig. 8A and 8B An example of detection rate as a function of epoch number in the presence of simulated ASSR is shown.

[0305] Fig. 8A The Fmpi-FFT detection rate 27 and the HT2 detection rate 28 are shown as a function of the number of epochs in the presence of a simulated 41 Hz ASSR. Figure 8B ) shows the false positive rate (FPR) 29 for the Fmpi-FFT method and the false positive rate 30 for the HT2 method as a function of the number of epochs (same noise but without ASSR). A total of 12 frequencies (41 Hz and its harmonics) were included in the analysis. The trial length was 1 second, the sampling rate was 2000 Hz, the analysis window was 1000 ms, and the data was not filtered. The horizontal line in the right figure marks the target 5% false positive rate used in this simulation.

[0306] exist Fig. 8A In the figure, it can be seen that in the presence of simulated neural signals, Fmpi-FFT outperforms HT2, indicating more realistic detection (left) and lower false positives ( Figure 8B ).

[0307] The structural features of the apparatus and systems described above, described in detail in the "Detailed Description of the Invention", and defined in the claims may be combined with the steps of the method of the present invention when appropriately replaced by corresponding procedures.

[0308] Unless expressly stated, the singular forms "one", "the" used herein include the plural form (i.e., have the meaning of "at least one"). It should be further understood that the terms "having", "including" and / or "comprising" used in the specification indicate the presence of the described features, integers, steps, operations, elements and / or parts, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, parts and / or combinations thereof. It should be understood that, unless expressly stated, when an element is referred to as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be an intermediate intervening element. As used herein, the term "and / or" includes any and all combinations of one or more listed related items. Unless expressly stated, the steps of any method disclosed herein do not have to be performed in the exact order disclosed.

[0309] The claims are not limited to the various aspects shown herein, but rather have the full scope consistent with the claim language wherein, unless expressly stated otherwise, elements referred to in the singular do not mean "one and only one" but rather "one or more." Unless expressly stated otherwise, the term "some" means one or more.

[0310] References

[0311] [1]Don,M.and Elberling,C.(1994)'Evaluating residual background noise in human auditory brain-stem responses.',The Journal of the AcousticalSociety of America,96(5Pt 1),pp.2746–57.

[0312] [2]Elberling, C. and Don, M. (1984) 'Quality estimation of averagedauditory brainstem responses.', Scandinavian audiology, 13(3), pp.187–97.

Claims

1. An apparatus for detecting an evoked response, wherein the apparatus comprises: a stimulus generator configured to generate at least one stimulus or a plurality of sequential stimuli according to a test protocol; at least one output unit comprising a transducer, the output unit being configured to receive the at least one stimulus or the plurality of consecutive stimuli from the stimulus generator and to provide the at least one stimulus or the plurality of consecutive stimuli to a subject; at least one recording unit comprising one or more sensors for measuring one or more evoked responses of the subject in response to at least one stimulus or a plurality of successive stimuli provided; An analysis unit configured to receive and analyze one or more measured evoked reactions, wherein the analysis unit is configured to determine a probability p that each of the reactions is driven by potential background noise during operation of the instrument, the analysis unit being configured to determine the probability p based on an F value of the measured one or more evoked reactions determined as the ratio between the variance of the average of the one or more evoked reactions and the variance of the residual background noise.

2. The apparatus according to claim 1, wherein: The one or more sensors are configured to measure one or more evoked responses including a ratio of a variance of a mean response evoked by the stimulus to a variance of a residual background noise.

3. An apparatus according to any preceding claim, wherein: The analysis unit is configured to determine the probability p by: p=1-F CDF (F,v,r) Among them, F CDF (F,v,r) is the cumulative distribution function of an F-distributed random variable with v and r, v is the statistical degrees of freedom for the variance of the mean of the one or more evoked responses, and r is the statistical degrees of freedom for the variance of the residual background noise.

4. An apparatus according to any preceding claim, wherein: The analysis unit is configured based on The degrees of freedom are for the frequency of interest ω i Determine the numerator of the F value, and based on The degrees of freedom are for the frequency of interest ω i Determine the denominator of the F value.

5. An apparatus according to any preceding claim, wherein: The analyzing unit is configured to estimate the energy of the residual background noise at each frequency of interest by averaging the energy of the residual background noise at neighboring frequencies of the frequency of interest.

6. An apparatus according to any preceding claim, wherein: The analysis unit is configured to adaptively determine the probability during operation of the instrument.

7. An apparatus according to any preceding claim, wherein: The at least one stimulus or the plurality of continuous stimuli may comprise acoustic stimuli such as clicks, narrow-band chirps, auditory variable complex stimuli, and / or speech signals, and / or electrical stimuli.

8. An apparatus according to any preceding claim, wherein: The one or more evoked responses measured may include auditory brainstem responses, cortical auditory potentials, auditory steady-state responses, otoacoustic emissions, or any response that relies on averaging to reduce potential background noise when detecting a target signal.

9. An apparatus according to any preceding claim, wherein: The analysis unit is configured to determine the hearing ability of the subject in response to the probability being below a predetermined threshold.

10. An apparatus according to any preceding claim, wherein: The stimulus generator is configured to modify the at least one stimulus or the plurality of consecutive stimuli in response to the determined probability.

11. A method for detecting an induced response, wherein the method comprises: generating at least one stimulus or a plurality of consecutive stimuli according to a test protocol; providing the at least one stimulus or the plurality of consecutive stimuli to the subject; placing one or more sensors on the subject; measuring one or more evoked responses of the subject in response to the provided at least one stimulus or a plurality of successive stimuli; receiving and analyzing, by an analysis unit, the measured one or more evoked responses; The probability p is determined during operation of the instrument by determining the probability p that each of the measured one or more evoked responses is driven by underlying background noise based on an F value determined as the ratio between the variance of the average of the one or more evoked responses and the variance of the residual background noise, each variance having statistical degrees of freedom v and r, respectively.

12. A data processing system comprising a processor and a program code, wherein the program code causes the processor to execute at least some of the steps of the method according to claim 11.

13. A computer program product, comprising a computer program containing instructions, which, when the computer program is run by an apparatus according to any one of claims 1 to 10, causes the apparatus to perform the steps of the method according to claim 11.

Images